Collaborating Researcher
Institute of Computer Science, Foundation of Research & Technology (FORTH)
Haridimos Kondylakis is currently a Collaborating Researcher at Computational BioMedicine Laboratory (CBML), Institute of Computer Science, Foundation of Research & Technology-Hellas (FORTH). He holds a Ph.D and an M.Sc. in Computer Science from University of Crete.
He is also a visiting lecturer at the Computer Science Department, University of Crete and at the Department of Electric and Computer Engineeringat Hellenic Mediterranean University and at Hellenic Open University, teaching lessons related to Big Data Management, Semantics and Artificial Intelligence.
In the past he has participated in more that 27 EU research projects with a total budget over 100Mie and he is currently participating in nine H2020 EU projects (eCare - Technical Manager, ProCancer-I - WP Leader, InCareHeart - Task Leader, CareMatrix - Task Leader, EUCAIM - Task Leader, GenoMed4All - Task Leader, ODIN - Task Leader, SafePolyMed - Task Leader RadioVal - Task Leader) and one national research project (iQARuS as PI).
He has more than 210 publications in international conferences, books and journals including ACM SIGMOD, VLDB, JWS, KER, EDBT, ISWC, ESWC etc. He is the organizer of the International Workshop on Semantic Web Technologies for Health Data Management and the leader of Data Storage, Curation and Management Working Group of the EU commission for the AI for health imaging projects. He also acts as a regular reviewer and a PC member for a number of premier journals and conferences, such as WWW, EDBT, SIGMOD, VLDBJ, JWS, JODS, CIKM, EDBT, ISWC etc.
Institute of Computer Science, Foundation of Research & Technology (FORTH)
European Commission
Hellenic Open University
Computer Science Department, University of Crete
Department Electrical and Computer Engineering, Hellenic Mediterranean University
General Secretary of Research and Technology
Information Systems Division, Hellenic Army
Institute of Computer Science, Foundation of Research & Technology (FORTH)
PLANET S.A
Ph.D. in Information Systems
Computer Science Department, University of Crete, Greece
Master in Information Systems & Telecommunications
Computer Science Department, University of Crete, Greece
Bachelor in Computer Science
Computer Science Department, University of Crete, Greece
Streams of data are now massively, constantly arrive from Internet of Things Devices (IoT), sensors and social media, and require rapid processing, querying and integration with background knowledge in order to support further data analysis. To this direction, RDF Stream processing platforms are valuable tools, enabling query answering over RDF streams. However, so far, the current state-of-the-art in RDF Stream processing has provided either centralized engines that cannot deal with massive RDF data streams or distributed engines that offer limited reasoning capabilities. Summarization techniques on the other hand have already proved their value for indexing data, query answering, reasoning, source selection, graph visualization, and schema discovery. However, to the best of our knowledge they have not yet been exploited for stream data, which remains a completely unexplored area. iQARuS objective is to enable effective and efficient query answering over RDF stream data using summaries. Generated summaries will be smaller than the original graphs and as such they will reduce drastically the data space, enabling efficient query answering and reasoning. However, due to this reduced data space exact answers might not be always possible to be retrieved directly from summaries. We intend to explore approximate query answering and then to offer exploration operations that will allow expanding the summaries for exact query answering. In addition, incremental algorithms will enable summary updating to avoid the overhead of summary recomputation from scratch. The developed solution will cover and combine both recent window stream data and background, staged knowledge and will be evaluated extensively using both well-established RDF Stream Processing benchmarks and a new one to be generated during the lifetime of the project.
In Europe, prostate cancer (PCa) is the second most frequent type of cancer in men and the third most lethal. Current clinical practices, often leading to overdiagnosis and overtreatment of indolent tumors, suffer from lack of precision calling for advanced AI models to go beyond SoA by deciphering non-intuitive, high-level medical image patterns and increase performance in discriminating indolent from aggressive disease, early predicting recurrence and detecting metastases or predicting effectiveness of therapies. To date efforts are fragmented, based on single–institution, sizelimited and vendor-specific datasets while available PCa public datasets (e.g. US TCIA) are only few hundred cases making model generalizability impossible. The ProCAncer-I project brings together 20 partners, including PCa centers of reference, world leaders in AI and innovative SMEs, with recognized expertise in their respective domains, with the objective to design, develop and sustain a cloud based, secure European Image Infrastructure with tools and services for data handling. The platform hosts the largest collection of PCa multi-parametric (mp)MRI, anonymized image data worldwide (>17,000 cases), based on data donorship, in line with EU legislation (GDPR). Robust AI models are developed, based on novel ensemble learning methodologies, leading to vendor-specific and -neutral AI models for addressing 8 PCa clinical scenarios. To accelerate clinical translation of PCa AI models, we focus on improving the trust of the solutions with respect to fairness, safety, explainability and reproducibility. Metrics to monitor model performance and a causal explainability functionality are developed to further increase clinical trust and inform on possible failures and errors. A roadmap for AI models certification is defined, interacting with regulatory authorities, thus contributing to a European regulatory roadmap for validating the effectiveness of AI-based models for clinical decision making.
Hospitals must increase their efficiency and productivity and boost quality and safety, while containing and reducing costs. This cannot be an untaught linear reduction. For instance, the number of ICU beds per million of EU habitants was reduced of 75% in the past 30 years, also in response to the unneglectable need to invest on territory healthcare services in response to democratic challenges. This left EU Hospitals completely unprepared to the COVID-19 pandemics, proving that hospital budget cuts must be complemented with major organizational restructuring, making use of innovative technologies. We have identified 11 hospital critical challenges, which ODIN will face combining robotics, Internet of Things (IoT) and Artificial Intelligence (AI) to empower workers, medical locations, logistics and interaction with the territory. ODIN will deploy technologies along three lines of intervention: empowering workers (AI, cybernetics and bionics), introducing autonomous and collaborative robots and enhancing medical locations with IoT. These areas of intervention will be piloted in six hospitals (in Spain, France, Italy, Poland, The Netherland, Germany), via seven use cases, spanning from clinical to logistic, including patient management, disaster preparedness and hospital resiliency.ODIN pilot will be a federation of multicentre longitudinal cohort studies, demonstrating the safety, effectiveness and cost-effectiveness of ODIN technologies for the enhancement of hospital safety, productivity and quality. Use-case protocols will be approved by the local hospital ethical committees, in order to assure the highest quality of the study, while providing a pragmatic solution for the scaling-up of the ODIN technological solutions and business models in a variety of local ecosystems. ODIN vision is that as Evidence Based Medicine revolutionized medicine with data-driven procedures, so data-driven management (enabled by Industry 4.0 tech) can revolutionise hospital management
Personalized or precision medicine is a medical model that combines already established clinical-pathological parameters with advanced genomic profiling in order to create innovative diagnostic, prognostic and therapeutic strategies. In this context, haematology has been rapidly transformed by genome characterization. However, despite the existence of national collaborative groups for many hematological diseases (HDs), some of them cooperating at the EU level, national approaches for HDs clinical management and research are often ineffective, especially for rarest conditions. Development of infrastructures that can support collection and use of genomic information in the health-care community represents a research priority for HDs, as repositories of genomic and clinical information in Europe are unconnected and due to the large number of disorders and in some cases relative number of samples small, it is difficult to have central big data repositories as existing in other areas. Clinical networks are needed to address accrual of sufficient patients for both genomic profiling and conduction of clinical trials. In this context, Genomics and Personalized Medicine for all though Artificial Intelligence in Haematological Diseases (GENOMED4ALL) project proposal has been invited for the grant preparation phase under the call H2020 "DT-TDS-04: AI for Genomics and Personalised Medicine" for supporting the pooling of genomic, clinical data and other "-omics" health data (clinical data from Electronic Health Record, PET, MRI and CT, Next Generation Sequencing, etc.) through a secure and privacy respectful data sharing platform based on the novel Federated Learning scheme, to advance research in personalised medicine in haematological diseases thanks to advanced Artificial Intelligence (AI) models and standardized interoperable sharing of cross-border data, without needing to directly share any sensitive clinical / patients' data.
Design, develop and test innovative AI-based and data intensive solutions for tackling multimorbidity problems.
Design, develop and test innovative AI-based and data intensive solutions for tackling chronic heart failure problems.
Design, develop and test a plarform with relevant services for management of frailty
Desing of the common data model and the hyperontology and integration of clinical and imaging data for the development of AI models.
Desing of the common data model and integration of clinical and imaging data for the development of AI models.
Desing of the common data model and integration of clinical and imaging data for the development of AI models.
The BOUNCE project aims to research aspects that influence breast cancer patients’ resilience, and ability to return to normal/everyday life and work after the breast cancer treatments. The project starts by modelling the factors that influence resilience. Afterwards the functionality of the model is being tested in a multicenter pilot study. Patient information in the pilot is collected with mobile PRO (patient reported outcomes) follow-up application Noona. The aim of the project is to develop clinical tools and operation model that support patients’ ability to recover from cancer treatments. BOUNCE project will start by the end of the year, and will involve a broad group of early phase breast cancer patients and their well-being. Patients’ empowerment and participation to their treatment might also be cost efficient.
Design, develop and test a resilient support tool, to be applied in the field of patients, planned for surgery, with the aim of reducing stress and anxiety as well as improving the health condition of the patient during the complete care path.
Over the last two decades, substantial progress has been made in the early detection, diagnosis, and treatment of cancer resulting in more and more patients who may be freed of their cancer. The disease can now be managed as a chronic illness requiring long-term surveillance and, in some cases, maintenance treatment that stretches from prevention to the end of life, with early detection, diagnosis, treatment, and survivorship in between. Chronic cancer treatment places new demands on patients and families to manage their own care. A collaborative and interactive relationship between patients and health professionals can empower patients to take on responsibility for their condition with the appropriate clinical support. Overcoming the challenge that patients see themselves as passive recipients of care could translate in aiding them in becoming motivated to take responsibility for their own contribution, in order to improve their health and well-being status, and feeling more empowered to do so. New ICT driven models of self-management-support developed within this project will enable them to better know their condition and the various treatment options, to monitor and manage the adverse events before they occur but also other symptoms and signs of their illness, to manage the impact of the disease on emotions, physical functioning and interpersonal relationships, to engage in activities that maintain their health and to form partnerships with oncologists and primary care providers, in order to negotiate and follow a plan of care.
The iManageCancer project will provide a cancer specific self-management platform designed according to the needs of patient groups while in parallel focusing on the wellbeing of the cancer patient with special emphasis on avoiding, early detecting and managing adverse events of cancer therapy but also, importantly, on the psycho-emotional evaluation and self-motivated goals. The platform will be centred in a Personal Health Record that will exploit recent advances on Health Avatars for the individual cancer patient surrounded by mHealth applications designed to encourage the patient, enhance clinician-patient communication, maximise compliance to therapy, inform about drug interactions, and contribute to the management of pain and other side-effects of cancer treatment. The Health Avatar PHR will regularly monitor the psycho-emotional status of the patient and will periodically record the everyday life experiences of the cancer patient with respect to the therapy side effects, while different groups of patients and their families will share information through diaries. The PHR will be used to provide the clinician with valuable clinical information about the patients’ cancer state and will help assess adherence to therapy, physiological and psychological status while the platform will recommend targeted informative applications and serious games according to the disease type and psycho-emotional status of the patients in order to promote a positive and healthier psycho-emotional state, and thus reducing anxiety and depression. The disease management platform will be further complemented by an integrated expert system with formal self-management models executed by a Care Flow Engine that will be oriented to decision support, the management of side-effects, adherence to therapy and guidance for patients including drug dose self-adjustments where feasible.
The iManageCancer platform will be designed on clinical evidence and in close collaboration of clinical experts, IT specialists and patients and will be assessed in clinical pilots with adult and paediatric cancer patients.
The personalized services of the platform will be complemented by smart analytic data services that support research on anonymised patient data. A novel business model, the Health Data Cooperative, will be pursued as one alternative for the exploitation of the results of iManageCancer.
Owing to the highly fragmented health systems in European countries, gaining access to a consistent record of individual citizens that involves cross-border activities is very difficult. MyHealthAvatar is an attempt at a proof of concept for the digital representation of patient health status. It is designed as a lifetime companion for individual citizens that will facilitate the collection of, and access to, long-term health-status information. This will be extremely valuable for clinical decisions and offer a promising approach to acquire population data to support clinical research, leading to strengthened multidisciplinary research excellence in supporting innovative medical care.
MyHealthAvatar will be built on the latest ICT technology with an aim of engaging public interest to achieve its targeted outcomes. In addition to data access, it is also an interface to access integrative models and analysis tools, utilizing resources already created by the VPH community. Overall, it will contribute to individualized disease prediction and prevention and support healthy lifestyles and independent living. It is expected to exert a major influence on the reshaping of future healthcare in the handling of increased life expectancy and the ageing population in Europe. This complies with the priority and strategy of FP7 ICT for healthcare, and constitutes a preparatory action aiming at the grand challenge on a “Digital Patient”, which is currently the subject of a roadmap in the VPH community..
MyHealthAvatar places a special emphasis on engaging the public. It has huge implications to the society both socially and economically. The initiative of designing a personal avatar can potentially change the way we think, communicate and search for information. Meanwhile, the acceptance of the avatars by the public will open opportunities for many industrial sectors, leading to the reinforced leadership of European industry.
Identifying individual patient characteristics that contribute to long-term mental health deterioration following diagnosis of breast cancer (BC) is critical in clinical practice. The present study employed a supervised machine learning pipeline to address this issue in a subset of data from a prospective, multinational cohort of women diagnosed with stage I-III BC with a curative treatment intention. Patients were classified as displaying stable HADS scores (Stable Group; n=328) or reporting a significant increase in symptomatology between BC diagnosis and 12 months later (Deteriorated Group; n=50). Sociodemographic, life-style, psychosocial, and medical variables collected on the first visit to their oncologist and three months later served as potential predictors of patient risk stratification. The flexible and comprehensive machine learning (ML) pipeline used entailed feature selection, model training, validation and testing. Model-agnostic analyses aided interpretation of model results at the variable- and patient-level. The two groups were discriminated with a high degree of accuracy (Area Under the Curve=0.864) and a fair balance of sensitivity (0.85) and specificity (0.87). Both psychological (negative affect, certain coping with cancer reactions, lack of sense of control/positive expectations, and difficulties in regulating negative emotions) and biological variables (baseline percentage of neutrophils, thrombocyte count) emerged as important predictors of mental health deterioration in the long run. Personalized break-down profiles revealed the relative impact of specific variables toward successful model predictions for each patient. Identifying key risk factors for mental health deterioration is an essential first step toward prevention. Supervised ML models may guide clinical recommendations toward successful illness adaptation.
The explosion in the size and the complexity of the available Knowledge Graphs on the web has led to the need for efficient and effective methods for their understanding and exploration. Semantic summaries have recently emerged as methods to quickly explore and understand the contents of various sources. However, in most cases, they are static, not incorporating user needs and preferences, and cannot scale. In this paper, we present iSummary, a novel, scalable approach for constructing personalized summaries. As the size and the complexity of the Knowledge Graphs for constructing personalized summaries prohibit efficient summary construction, in our approach we exploit query logs. The main idea behind our approach is to exploit knowledge captured in existing user queries for identifying the most interesting resources and linking them, constructing as such high-quality, personalized summaries. We present an algorithm with theoretical guarantees on the summary’s quality, linear in the number of queries available in the query log. We evaluate our approach using three real-world datasets and several baselines, showing that our approach dominates other methods in terms of both quality and efficiency.
Semantic summaries try to extract compact information from the original knowledge graph (KG) while reducing its size for various purposes such as query answering, indexing, or visualization. Although so far several techniques have been exploited for summarizing individual KGs, to the best of our knowledge, there is no approach summarizing the interests of the users in exploring those KGs, capturing also how these evolve. SummaryGPT fills this gap by enabling the exploration of users' interests as captured from their queries over time. For generating these summaries we first extract the nodes appearing in query logs, captured from a specific time period, and then we classify them into different categories in order to generate quotient summaries on top. For the classification, we explore both the KG type hierarchy (if existing) and also a large language model, i.e. ChatGPT. Exploring different time periods enables us to identify shifts in user interests and capture their evolution through time.
Artificial Intelligence (AI) is transforming the field of medical imaging and has the potential to bring medicine from the era of ‘sick-care’ to the era of healthcare and prevention. The development of AI requires access to large, complete, and harmonized real-world datasets, representative of the population and diseases diversity. However, to date, efforts are fragmented, based on single–institution, size- and annotation-limited datasets. Available public datasets (e.g., US TCIA) are limited in scope, making model generalizability really difficult. In this direction, five EU projects are currently working on the development of big data infrastructures that will enable European, ethically and GDPR-compliant, quality-controlled, cancer-related, medical imaging platforms, in which both large-scale data and AI algorithms will co-exist. The vision is to create sustainable AI cloud-based platforms for the development, implementation, verification and validation of trustable, usable, and reliable AI models for addressing specific unmet needs regarding cancer care provision. In this paper, we present an overview of the development efforts highlighting challenges and approaches selected providing valuable feedback to future attempts in the area.
Proper and well-timed interventions may improve breast cancer patient adaptation and quality of life (QoL) through treatment and recovery. The challenge is to identify those patients who would benefit most from a particular intervention. The aim of this study was to measure whether the machine learning prediction incorporated in the clinical decision support system (CDSS) improves clinicians’ performance to predict patients’ QoL during treatment process. We conducted two user experiments in which clinicians used a CDSS to predict QoL of breast cancer patients. In both experiments each patient was evaluated both with and without the aid of a machine learning (ML) prediction. In Experiment I, 60 breast cancer patients were evaluated by 6 clinicians. In Experiment II, 90 patients were evaluated by 9 clinicians. The task of clinicians was to predict the patient’s quality of life at either 6 (Experiment I) or 12 months post-diagnosis (Experiment II). Taking into account input from the CDSS considerably improved clinicians’ prediction accuracy. Accuracy of clinicians for predicting QoL of patients at 6 months post-diagnosis was .745 (95% CI .668-.821) with the aid of the prediction provided by the ML model and .696 (95% CI .608-.781) without the aid. Clinicians’ prediction accuracy at 12 months was .739 (95% CI .667-.812) with the aid and .709 (95% CI .633- .783) without the aid. When the machine learning model’s prediction was correct, the average accuracy of the clinicians for predicting QoL at 6 months was .793 (95% CI .739-.838) with the aid and .720 (95% CI .636-.798) without the aid. Corresponding prediction accuracy of QoL at 12 months was .909 (95% CI .881-.936) and .827 (95% CI .782-.871).
Knowledge graphs are becoming more and more prevalent on the web, ranging from small taxonomies, to large knowledge bases containing a vast amount of information. To construct such knowledge graphs either automatically or manually, tools are necessary for their quick exploration and understanding. Semantic summaries have been proposed as a key technology enabling the quick understanding and exploration of large knowledge graphs. Among the methods proposed for generating summaries, structural methods exploit primarily the structure of the graph in order to generate the result summaries. Approaches in the area focus on identifying the most important nodes and usually employ a single centrality measure, capturing a specific perspective on the notion of a node’s importance. Moving from one centrality measure to many however, has the potential to generate a more objective view on nodes’ importance, leading to better summaries. In this paper, we present SumMER, the first structural summarization technique exploiting machine learning techniques for RDF/S KGs. SumMER explores eight centrality measures and then exploits machine learning techniques for optimally selecting the most important nodes. Then those nodes are linked formulating a subgraph out of the original graph. We experimentally show that combining centrality measures with machine learning effectively increases the quality of the generated summaries.
The explosion of the data on the semantic web has led to many weakly structured, and irregular data sources, becoming available every day. The schema of these sources is useful for a number of tasks, such as source selection, query answering, exploration and summariza- tion. However, although semantic web data might contain schema in- formation, in many cases this is completely missing or partially defined. Schema discovery consists in extracting schema-related information from the original semantic graph, which some applications can exploit instead of or along with the original graph, to perform some tasks more efficiently. This tutorial presents a structured analysis and comparison of existing works in the area of semantic schema discovery helping researchers and practitioners to understand the challenges in the area; it is based upon a recent survey we authored.
The rapid explosion of linked data has resulted in many weakly struc- tured and incomplete data sources, where typing information might be missing. On the other hand, type information is essential for a number of tasks such as query answering, integration, summarization, and partitioning. Existing ap- proaches for type discovery, either completely ignore type declarations avail- able in the dataset (implicit type discovery approaches), or rely only on exist- ing types, in order to complement them (explicit type enrichment approaches). In this demonstration, we present HInT, the first incremental and hybrid type dis- covery system for RDF datasets. To achieve this goal HInT identifies the patterns of the various instances, and then indexes and groups them to identify the types. Besides discovering new types, HInT exploits type information if available, to improve the quality of the discovered types by guiding the classification of the new instance in the correct group and by refining the groups already built.
More and more weakly structured, and irregular data sources are becoming available every day. The schema of these sources is useful for a number of tasks, such as query answering, exploration and summarization. However, although semantic web data might contain schema information, in many cases this is completely missing or partially defined. In this paper, we present a survey of the state of the art on schema information extraction approaches. We analyze and classify these approaches into three families: (i) approaches that exploit the implicit structure of the data, without assuming that some explicit statements on the schema are provided in the dataset; (ii) approaches that use the explicit schema statements contained in the dataset to complement and enrich the schema, and (iii) those that discover structural patterns contained in a dataset. We compare these studies in terms of their approach, advantages and limitations. Finally we discuss the problems that remain open.
Background: Major chronic diseases such as cardiovascular disease, diabetes, and cancer impose a significant burden on people and the healthcare systems around the globe. Recently, Deep Learning (DL) has shown great potential towards the development of intelligent mobile health (mHealth) interventions for chronic diseases which could revolutionize the delivery of healthcare anytime-anywhere. Objective: To present a systematic review of studies that have used DL based on mHealth data for the diagnosis, prognosis, management, and treatment of major chronic diseases, and advance our understanding of the progress made in this rapidly developing field. Methods: Search was conducted on the bibliographic databases of Scopus and PubMed in order to identify papers with a focus on the deployment of DL algorithms, using data captured from mobile devices (e.g., smartphones, smartwatches, and other wearable devices), targeting cardiovascular disease, diabetes, or cancer. The identified studies were synthesized according to the target disease, the number of enrolled participants and their age, the study period, as well as the employed DL algorithm, the main DL outcome, the dataset used, the features selected, and the achieved performance. Results: 20 studies were included in the review. 7 DL studies (35%) targeted cardiovascular disease, 9 studies (45%) targeted diabetes, and 4 studies (20%) targeted cancer. The most common DL outcome was the diagnosis of the patient condition for the cardiovascular disease studies, prediction of blood glucose values for studies in diabetes, and early detection of cancer. Most of the DL algorithms employed were convolutional neural networks in studies for cardiovascular disease and cancer, and recurrent neural networks in studies for diabetes. The performance of DL was found overall to be satisfactory reaching more than 84% accuracy in the majority of the studies. In comparison with classic machine learning approaches, DL was found to achieve better performance almost in all studies which reported such comparison outcomes. The vast majority of the studies did not provide details on the explainability of DL outcomes. Conclusions: The use of DL can facilitate the diagnosis, management, and treatment of major chronic diseases through harnessing mHealth data. Prospective studies are now required to demonstrate the value of applied DL in real-life mHealth tools and interventions.
Given the huge amount of imaging data that are becoming available worldwide, and the incredible range of possible improvements that artificial intelligence algorithms can provide in clinical care for diagnosis and decision support, it has become essential to properly manage and handle these medical images and to define which metadata have to be considered, in order for the images to provide their full potential. In medical imaging, metadata are additional data associated with the images, that provide a complete description of the image acquisition, curation, analysis, and of the relevant clinical variables associated with the images. Currently, several data models are available to describe one or more subcategories of metadata, but a unique, common, and standard data model capable of fully representing the heterogeneity of medical metadata has not been yet developed. This paper reports the state of the art on metadata models for medical imaging, the current limitations and further developments, and describes the strategy adopted by the HORIZON 2020 “AI for Health Imaging” projects, which are all dedicated to the creation of imaging biobanks.
Informal care is considered to be important for the wellbeing and resilience of the elderly. However, solutions for the effective collaboration of healthcare professionals, patients, and informal caregivers are not yet widely available. The purpose of this paper is to present the development of a digital platform that uses innovative tools and artificial intelligence technologies to support care coordination and shared care planning for elder care, with a particular focus on frailty. The challenges of shared care planning in the coordination of frailty care are demonstrated, followed by presentation of the design and technical architecture of an integrated platform. The platform incorporates all elements essential for the support of daily activities, coordinated care, and timely interventions in case of emergency and need. This paper describes the challenges involved in implementing the platform and concludes by reporting the necessary steps required in order to establish effective smart care for the elderly.
Background: Stress and anxiety are psychophysiological responses commonly experienced by patients during a perioperative process that can increase pre- and post-surgical complications to a comprehensive and positive recovery. Preventing and intervening stress and anxiety can help patients achieve positive outcomes on health and wellbeing. Similarly, provision of education about the surgery can be a crucial component and it is inversely correlated to preoperative anxiety levels. However, few patients receive stress and anxiety relief support prior to a surgery and, resource constraints make face-to-face education sessions untenable. Digital health (DH) interventions can be helpful in empowering patients and enhancing a more positive experience. DH interventions are showing to help patients feel informed about possible benefits and risks of available treatment options. However, currently they focus only on providing informative content, neglecting the importance of personalization and patient empowerment. Objective: This study aims to explore the feasibility of a DH intervention called CARINAE designed to provide personalized stress- and anxiety-management evidence-based methods enabled by a comprehensive digital ecosystem that incorporates wearable, mobile and virtual reality technologies. CARINAE includes the use of advanced data-driven techniques for tailored patient education and lifestyle support. Methods: The trial will include 5 hospitals across 3 European countries, and will use a randomized-controlled design including 30 intervention participants and 30 control group participants. The involved surgeries are cardiopulmonary and coronary artery bypass surgeries, cardiac valve replacement, prostate or bladder cancer surgeries, hip and knee replacement, maxillofacial surgery, or scoliosis. The control group will receive the standard care and the intervention group will additionally be exposed to the CARINAE intervention. Results: The recruitment process starts by January 2022, and the primary impact analysis is expected to be conducted by May 2022. Conclusions: This manuscript details a comprehensive protocol for a study that will provide valuable information about the CARINAE intervention such as the measurement of comparative intervention effects on stress, anxiety and pain management and usability by patients, caregivers and healthcare professionals. This will contribute to the evidence planning process for the future adoption of diverse DH solutions in the surgery field. Clinical Trial: Trial was registered in ClinicalTrials.gov: NCT05184725.
The term frailty is often used to describe a particular state of health, related to the ageing process, often experienced by older people. The most common indicators of frailty are weakness, fatigue, weight loss, low physical activity, poor balance, low gait speed, visual impairment and cognitive impairment. The objective of this work is the creation of a serious games mobile application to conduct elderly frailty assessments in an accurate and objective way using mobile phone capabilities. The proposed app includes three games (memory card, endless runner, and clicker) and three questionnaires, aiming towards the prediction of signs of memory and reflection deterioration, as well as endurance and strength. The games, when combined with a set of qualified questionnaires, can provide an efficient tool to support adults in identifying frailty symptoms and in some cases prevent further deterioration. At the same time the app can support older adults in improving physical and mental fitness, while gathering useful information about frailty.
Prostate cancer (PCa) is one of the most prevalent cancers in the male population. Current clinical practices lead to overdiagnosis and overtreatment necessitating more effective tools for improving diagnosis, thus the quality of life of patients. Recent advances in infrastructure, computing power and artificial intelligence enable the collection of tremendous amounts of clinical and imaging data that could assist towards this end. ProCAncer-I project aims to develop an AI platform integrating imaging data and models and hosting the largest collection of PCa (mp)MRI, anonymized image data worldwide. In this paper, we present an overview of the overall architecture focusing on the data ingestion part of the platform. We describe the workflow followed for uploading the data and the main repositories for storing imaging data, clinical data and their corresponding metadata.
Being diagnosed with breast cancer (BC) is perceived as a traumatic experience for the patient who is often confronted with mental disorders and psychiatric morbidities such as depression. In order to ensure support for successful adaptation to this illness, it is crucial to understand how and why depressive symptoms may evolve differently for each individual, from diagnosis through treatment and survivorship. In the present work, first, a trajectory-based unsupervised clustering based on K-means is performed to capture the dynamic patterns of change in patient’s depressive symptoms during the first year after BC diagnosis and to identify distinct trajectory clusters. Then a supervised learning approach was employed to build a classification model of depression progression and to identify potential predictors. Patients were clustered into 4 groups of stable low, stable high, improving and worsening depressive symptoms. In a nested cross-validation pipeline, the performance of the Support Vector Machine model for discriminating between “good” and “poor” progression was 0.78 ± 0.05 in terms of AUC. Several psychological variables emerged as highly predictive of the evolution of depressive symptoms with the most important ones being negative affectivity and anxious preoccupation.
In this work, we provide an overview of the state-of-the-art in the area of traceability for trustworthy AI along with a proposal for the minimal information to be kept for constructing a multi-layer AI passport for the developed models. Finally, we will report experiences from the relevant developments within the ProCAncer-I project.
Proper and well-timed interventions may improve breast cancer patient adaptation, resilience and quality of life (QoL) during treatment process and time after disease. The challenge is to identify those patients who would benefit most from a particular intervention. The aim of this study was to measure whether the machine learning prediction incorporated in the clinical decision support system (CDSS) improves clinicians’ performance to predict patients’ QoL during treatment process. We conducted an experimental setup in which six clinicians used CDSS and predicted QoL for 60 breast cancer patients. Each patient was evaluated both with and without the aid of machine learning prediction. The clinicians were also open-ended interviewed to investigate the usage and perceived benefits of CDSS with the machine learning prediction aid. Clinicians’ performance to evaluate the patients’ QoL was higher with the aid of machine learning predictions than without the aid. AUROC of clinicians was .777 (95% CI .691−.857) with the aid and .755 (95% CI .664−.840) without the aid. When the machine learning model’s prediction was correct, the average accuracy (ACC) of the clinicians was .788 (95% CI .739−.838) with the aid and .717 (95% CI .636−.798) without the aid.
Mental health impairment after breast cancer diagnosis may persist for months or years. The present work leverages on novel machine learning techniques to identify distinct trajectories of mental health progression in a 18-month period following BC diagnosis and develop an explainable predictive model of mental health progression using a large list of clinical, sociodemographic and psychological variables. The modelling process was conducted in two phases. The first modeling step included an unsupervised clustering to define the number of trajectory clusters, by means of a longitudinal Kmeans algorithm. In the second modeling step an explainable ML framework was developed, on the basis of Extreme Gradient Boosting (XGBoost) model and SHAP values, in order to identify the most prominent variables that can discriminate between good and unfavorable mental health progression and to explain how they contribute to model’s decisions. The trajectory analysis revealed 5 distinct trajectory groups with the majority of patients following stable good (57%) or improving (21%) trends, while for others mental health levels either deteriorated (12%) or remained at unsatisfactory levels (11%). The model’s performance for classifying patient mental health into good and unfavorable progression achieved an AUC of 0.82 ± 0.04. The top ranking predictors driving the classification task were the higher number of sick leave days, aggressive cancer type (triplenegative) and higher levels of negative affect, anxious preoccupation, helplessness, arm and breast symptoms, as well as lower values of optimism, social and emotional support and lower age.
Prostate cancer incidence has doubled in Europe between 1995 and 2018, surpassing both lung and colorectal cancer incidence. As such, it is a major health challenge, necessitating precision care through the whole disease continuum. The ProCAncer-I EU project aspires to create the largest interoperable, high-quality mpMRI (multi-parametric Magnetic Resonance Imaging) dataset worldwide comprising more than 11.000 retrospective and more than 6.000 prospective mpMRI examinations for the study of prostate cancer.
More and more weakly structured, and irregular data sources are becoming available every day. The schema of these sources is useful for a number of tasks, such as query answering, exploration and summarization. However, although semantic web data might contain schema information, in many cases this is completely missing or partially defined. In this paper, we present a survey of the state of the art on schema information extraction approaches. We analyze and classify these approaches into three families: (i) approaches that exploit the implicit structure of the data, without assuming that some explicit statements on the schema are provided in the dataset; (ii) approaches that use the explicit schema statements contained in the dataset to complement and enrich the schema, and (iii) those that discover structural patterns contained in a dataset. We compare these studies in terms of their approach, advantages and limitations. Finally we discuss the problems that remain open.
Analytical queries over RDF data are becoming prominent as aresult of the proliferation of knowledge graphs. Yet, RDF databasesare not optimized to perform such queries efficiently, leading tolong processing times. A well known technique to improve theperformance of analytical queries is to exploitmaterialized views.Although popular in relational databases, view materialization forRDF and SPARQL has not yet transitioned into practice, due to thenon-trivial application to the RDF graph model. Motivated by a lackof understanding of the impact of view materialization alternativesfor RDF data, we demonstrateSofos, a system that implements andcompares several cost models for view materialization.Sofosis, tothe best of our knowledge, the first attempt to adapt cost models,initially studied in relational data, to the generic RDF setting, and topropose new ones, analyzing their pitfalls and merits.Sofostakesan RDF dataset and an analytical query for some facet in the data,and compares and evaluates alternative cost models, displayingstatistics and insights about time, memory consumption, and querycharacteristics
Databases are at the core of virtually any soft-ware product. Changes to database schemas cannot be madein isolation, as they are intricately coupled with applicationcode. Such couplings enforcecollateral evolution, which is arecognized, important research problem. In this demonstration,we show a new dimension to this problem, in software thatsupports alternative database backends: vendor-specific SQLdialects necessitate a simultaneous evolution of both, databaseschemaandprogram code, for all supportedDB variants. Thesenear-same changes impose substantial manual effort for softwaredevelopers. We introduceDeBinelle, a novel framework anddomain-specific language forsemantic patchesthat abstractsDB-variant schema changes and coupled program code into asingle, unified representation. DeBinelle further offers a novelalternative to manually evolving coupled schemas and code.DeBinelle considerably extends established, seminal results insoftware engineering research, supporting several programminglanguages, and the many dialects of SQL. It effectively eliminatesthe need to perform vendor-specific changes, replacing them withintuitive semantic patches. We demonstrate the benefits of usingDeBinelle, based on real-world use cases from reference systemsfor schema evolution, and we allow conference participants toverify by hand the simplicity and the usefulness of our approach.
Knowledge graphs have now become common on the web, ranging from small taxonomies for categorizing web sites, to large knowledge bases that contain a vast amount of structured content. To enable their quick understanding and explo-ration semantic summaries have been proposed. A key issue of structural seman-tic summaries is the identification of the most important nodes. Works in the area, usually employee a single centrality measure, capturing a specific perspective on the notion of a node’s importance. However, combining multiple centrality measures could give a more objective view, on which nodes should be selected as the most important ones. In this paper, we present SumMER, a novel framework that explores machine learning techniques for optimally combining multiple cen-trality measures for selecting the most important nodes. The experiments per-formed show the benefit of our approach, effectively increasing the quality of the generated summaries.
As more and more data become available as linked data, the need for efficient and effective methods for their exploration becomes apparent. Semantic summaries try to extract meaning from data, while reducing its size. State of the art structural semantic summaries, focus primarily on the graph structure of the data, trying to maximize the summary’s utility for query answering, i.e. the query coverage. In this poster paper, we present an algorithm, trying to maximize the aforementioned query coverage, using ideas borrowed from result diversification. The key idea of our algorithm is that, instead of focusing only to the “central” nodes, to push node selection also to the perimeter of the graph. Our experiments show the po-tential of our algorithm and demonstrate the considerable advantages gained for answering larger fragments of user queries.
The rapid explosion of linked data has resulted into many weakly structured and incomplete data sources, where typing information might be missing. On the other hand, type information is essential for a number of tasks such as query answering, integration, summarization and partitioning. Existing approaches for type discovery, either completely ignore type declarations available in the dataset (implicit type discovery approaches), or rely only on existing types, in order to complement them (explicit type enrichment approaches). Implicit type discovery approaches are based on instance grouping, which requires an exhaustive comparison between the instances. This process is expensive and not incremental. Explicit type enrichment approaches on the other hand, are not able to identify new types and they can not process data sources that have little or no schema information. In this paper, we present HiNt, the first incremental and hybrid type discovery system for RDF datasets, enabling type discovery in datasets where type declarations are missing. To achieve this goal, we incrementally identify the patterns of the various instances, we index and then group them to identify the types. During the processing of an instance, our approach exploits its type information, if available, to improve the quality of the discovered types by guiding the classification of the new instance in the correct group and by refining the groups already built. We analytically and experimentally show that our approach dominates in terms of efficiency, competitors from both worlds, implicit type discovery and explicit type enrichment while outperforming them in most of the cases in terms of quality.
Semantic summaries try to extract compact information from the original RDF graph, while reducing its size. State of the art structural semantic summaries, focus primarily on the graph structure of the data, trying to maximize the summary’s utility for a specific purpose, such as indexing, query answering and source selection. In this paper, we present an approach that is able to construct high quality summaries, exploiting a small part of the query workload, maximizing their utility for query answering, i.e. the query coverage. We demonstrate our approach using two real world datasets and the corresponding query workloads and we show that we strictly dominates current state of the art in terms of query coverage.
Creating a holistic view on patient data comes withmany challenges but also brings many benefits for dis-ease prediction, prevention, diagnosis and treatment.Especially in the COVID-19 era, this is more importantthan ever before. The third International Workshop onSemantic Web Meets Health Data Management (SWH)was aimed at bringing together an interdisciplinary audi-ence who was interested in the fields of Semantic Web,data management and health informatics. The workshopgoal was to discuss the challenges in healthcare datamanagement and to propose new solutions for the nextgeneration data-driven healthcare systems. In this arti-cle, we summarize the outcomes of the workshop, andwe present a number of key observations and researchdirections that emerged from presentations.
This article studies opinion mining from social media with probabilistic logic reasoning. As it is known, Twitter is one of the most active social networks, with millions of tweets sent daily, where multiple users express their opinion about travelling, economic issues, political decisions etc. As such, it offers a valuable source of information for opinion mining. In this paper we present OpinionMine, a Bayesian-based framework for opinion mining, exploiting Twitter Data. Initially, our framework imports Tweets massively by using Twitter’s API. Next, the imported Tweets are further processed automatically for constructing a set of untrained rules and random variables. Then, a Bayesian Network is derived by using the set of untrained rules, the random variables and an evidence set. After that, the trained model can be used for the evaluation of new Tweets. Finally, the constructed model can be retrained incrementally, thus becoming more robust. As application domain for the development of our methodology we have selected tourism because it is one of the most popular topics in social media. Our framework can predict users’ intention to visit a place. Among the advantages of our framework is that it follows an incremental learning strategy. That is, the derived model can be retrained incrementally with new training sets thus becoming more robust. Further, our framework can be easily adapted to opinion mining from social media on other topics, whereas the rules of the derived model are constructed in an efficient way and automatically.
The rapid advancement of clinical research has resulted into numerous therapeutic options currently available for most of the diseases. During the patient therapeutic journey, many health-related decisions are necessary requiring patients to choose between the potential health benefits of an intervention, versus the countervailing risk of serious adverse health outcomes. Patient preference studies aim to elicit preferences with the common objective to generate information that facilitates comparing the importance of attributes of interest. The world has experienced a dramatic change in patient’s preferences during the pandemic. Despite the importance of patient preference studies in healthcare decision making, there is a lack of effective storage and accessibility of relevant data for wider use. In this paper the authors present the design of a platform to systematically collect, curate, annotate, index, synthesize and make available pertinent information of patient preference studies so that they can be further exploited by decision support tools.
Ontologies are widely used nowadays. However, the plethora of ontologies currently available online, makes it really difficult to identify which ontologies are appropriate for a given task and to decide on their quality characteristics. This is further complicated, by the fact that multiple quality criteria have been proposed for ontologies making it even more difficult to decide which ontology to adopt. In this direction, in this paper, we present Delta, a modular online tool for analyzing and evaluating ontologies. The interested user can upload an ontology to the tool, which, then automatically analyzes it and graphically visualizes numerous statistics, metrics, and pitfalls. Those visuals presented include a diverse set of quality dimensions, further guiding users to understand the benefits and the drawbacks of each individual ontology and how to properly develop and extend it.
HIFUN is a high-level query language for expressing analytic queries over big data sets, offering a clear separation between the conceptual layer, where analytic queries are defined independently of the nature and location of data, and the physical layer where queries are evaluated. In this paper, we present a methodology, based on the HIFUN language, and the corresponding algorithms for the incremental evaluation of continuous queries. In essence, our approach, is able to process the most recent data batch, exploiting already computed information, without requiring the evaluation of the query over the complete data set. We present the generic algorithm which subsequently, we translate to both SQL and MapReduce using SPARK, implementing various query rewriting methods. We demonstrate the effectiveness of our approach in achieving query answering efficiency. Finally, we show that by exploiting the formal query rewriting methods of HIFUN, we can further reduce the computational cost, adding another layer of query optimization to our implementation.
COVID-19 pandemic has affected nearly every aspect of life. Observing online the spread of the virus can offer a complementary view to the cases that are daily officially recorded and reported. In this article, we present an approach that exploits information available on social media to predict whether a patient has been infected with COVID-19. Our approach is based on a Bayesian model that is trained using data collected online. Then the trained model can be used for evaluating the possibility that new patients are infected with COVID-19. The experimental evaluation presented shows the high quality of our approach. In addition, our model can be incrementally retrained, so that it becomes more robust in an efficient way.
Informal care is fundamental in the wellbeing and resilience of elderly and people with chronic conditions. However, solutions for the effective collaboration of healthcare professionals, patients and informal carers are not yet widely available. CareKeeper builds on a state-of-the-art personal health system, augmenting it with Artificial Intelligence and Big Data technologies, to boost informal care coordination. In this paper we report on the design of the platform with the aim of providing a light-weighted communication solution to support practical challenges about sharing the responsibility of caring, such as the frequency of visits, support to routinely activities and timely intervention in case of emergency and need.
Sentiment Analysis is an actively growing field with demand in both scientific and industrial sectors. Political sentiment analysis is used when a data analyst wants to determine the opinion of different users on social media platforms regarding a politician or a political event. This paper presents Athena Political Popularity Analysis (AthPPA), a tool for identifying political popularity over Twitter. AthPPA is able to collect in real-time tweets and for each tweet to extract metadata such as number of likes, retweets per tweet etc. Then it processes their text in order to calculate their overall sentiment. For the calculation of sentiment analysis, we have implemented a sentiment analyzer that is able to identify the grammatical issues of a sentence as well as a lexicon of negative and positive words designed specifically for political sentiment analysis. An analytic engine processes the collected data and provides different visualizations that provide additional insights on the collected data. We show how we applied our framework to the three most prominent Greek political leaders in Greece and present our findings there.
Breast cancer diagnosis has been associated with serious mental health problems, significantly impairing patients’ quality of life and their way to cope with therapy. In order to ensure successful illness adaptation, it is of paramount importance to identify the most prominent factors affecting patients’ mental well-being and accurately predict their mental health status. In the present work we exploit a rich set of clinical, psychological, socio-demographic, and lifestyle data from a large multi-centric study, in order to classify patients diagnosed with breast cancer based on their mental health status and further identify the potential predictors. For this purpose, a supervised learning pipeline using cross-sectional data was implemented for the formulation of a classification scheme of mental health status 6 months after diagnosis.
Coping with breast cancer has become more than ever before a major socio-economic challenge. In this short paper we report on activities within the BOUNCE EU project focusing on understanding, predicting and increasing resilience for breast cancer patients throughout their cancer management continuum.
Advances in computers and communications have significantly changed almost every aspect of our daily activity. In this maze of change, governments around the world cannot remain indif-ferent. Public administration is evolving and taking on a new form through e-Government. A large number of organizations have set up websites, establishing an online interface with the citizens and businesses with which it interacts. However, most organizations, especially the de-centralized agencies of the ministries and local authorities, despite the fact that they provide many information services they are not integrated with other e-government services and are not offered electronically. Besides, these services are mainly focused on serving citizens and busi-nesses and less on providing services to employees. In this paper, we describe the process of de-veloping an Ontology to support the administrative procedures of decentralized government organizations. Finally, we describe the development of an e-government portal that provides employees services that are processed online, using the above ontology for modeling and data management.
The lives of millions of people have been affected during the coronavirus pandemic that spread throughout the world in 2020. Society is changing establishing new norms for healthcare education, social life, and business. Digital health has seen an accelerated implementation throughout the world in response to the pandemic challenges. In this perspective paper, the authors provide an overview of the digital platform Safe in COVID-19 to highlight the necessary requirements for incorporating digital apps in integrated healthcare systems. Semantics and data management could highly contribute, along with personal health apps, to the secondary use of available data in order to support effective prevention, prediction, and disease management, while at the same time increase collective knowledge.
The advancements in health-care have brought to the foreground the need for flexible access to health-related information and created an ever-growing demand for efficient data management infrastructures. To this direction, many challenges must be first overcome, enabling seamless, effective and efficient access to several health data sets and novel methods for exploiting the existing information. The second international workshop on semantic Web technologies for health data management aimed at putting together an interdisciplinary audience that is interested in the fields of semantic web, data management and health informatics to discuss the challenges in health-care data management and to propose new solutions for the next generation data-driven health-care systems. In this article, we summarize the outcomes of the workshop, and we present a number of key observations and research directions that emerge.
It is well known that the mental and emotional state of cancer patients plays an important role in the treatment of their disease. As such, for building prediction tools, patient’s psychology should also be considered, along with medical, clini-cal, biological and lifestyle data. However, for modelling patients psychological status, only a limited set of terms is available in existing ontologies. The BOUNCE Psychological Ontology (BPO) is an attempt to model all relevant psychological constructs, for cancer patients, effectively capturing patients' emo-tional and mental disposition in order to further study methods for coping with and recovering from the disease.
COVID-19 is a disease that has infected almost the whole world and has been pronounced as a global pandemic. The digital health domain has already tried to respond to the pandemic challenges by developing algorithms and applications that make predictions on the infection and the corresponding outcome in case of infection. In this direction, in this paper, we present "COVID-19 Detect & Predict", an application that can detect and predict COVID-19 infection through probabilistic logic reasoning, and in case of infection, it can also predict the outcome. We demonstrate the effectiveness of our solution on realistic datasets, showing its potential benefits. Our approach is the first, which detects and predicts COVID-19 infection through probabilistic logic reasoning to the best of our knowledge.
Sentiment analysis over social media platforms has been an active case of study for more than a decade. This occurs due to the constant rising of internet users over these platforms as well as to the increasing interest of companies for monitoring the opinion of customers over commercial products. Most of these plat-forms provide free online services such as the creation of interactive web communities, multimedia content uploading etc. This new way of communication has affected human societies as it shaped the way by which an opinion can be expressed, sparking the era of digital revolution. One of the most profound examples of social networking platforms for opinion mining is Twitter as it is a great source for extracting news and a platform which politicians tend to use frequently. In addition to that, the character limitation per posted tweet (maximum of 280 characters) makes it easier for automated tools to extract its underlying sentiment. In this review paper, we present a variety of lexicon-based tools as well as machine learning algorithms used for sentiment extraction. Furthermore, we present additional implementations used for political sentiment analysis over Twitter as well as additional open topics. We hope the review will help readers to under-stand this scientifically rich area, identify best options for their work and work on open topics.
Background: Patients undergoing elective surgery often face symptoms of anxiety and stress. Healthcare systems have limited time and resources to provide individualized stress relief interventions. Research has shown that stress relief interventions and educational resources can improve health outcomes and speed recovery. Objectives: Digital health tools can provide valuable assistance in stress relief and educational support to patients and family. This paper reports on the design of a novel digital health infrastructure for improving the health condition of patients, during the care path, using virtual reality (VR) and other information and communication technologies (ICT). Methods: Digital tools developed and integrated into a platform of modules that can be used by patients before but also after an operation, enabling better self-management and self-empowerment. Results: The designed platform aims at improving the knowledge of patients about their condition, providing stress relief tools, helping them adhere to treatment, as well as providing for effective communication channels between patients and clinicians. Conclusion: The proposed solution has the potential to improve physical and emotional reactions to stress and increase the levels of calmness and a sense of wellbeing. Information provided through the platform enhances health literacy and digital competence, and increases the participation of the patient in the decision-making process. Integration with third-party applications can facilitate the exchange of important information between patients and physicians as well as between personal applications and clinical health systems.
Background: A vast amount of mobile apps have been developed during the past few months in an attempt to “flatten the curve” of the increasing number of COVID-19 cases. Objective: This systematic review aims to shed light into studies found in the scientific literature, which have used and evaluated mobile apps for the prevention, management, treatment, and/ or follow-up of COVID-19. Methods: We searched the bibliographic databases of COVID-19 global literature on coronavirus disease, PubMed and Scopus, to identify papers focusing on mobile apps for COVID-19 that (i) show evidence on their real-life use and (ii) have been developed involving clinical professionals in their design or validation. Results: Mobile apps have been implemented for training, information sharing, risk assessment, self-management of symptoms, contact tracing, home-monitoring and decision making, rapidly offering effective and usable tools against the COVID-19 pandemic. Conclusions: Mobile apps are considered to be a valuable tool for citizens, health professionals and decision makers in facing critical challenges imposed by the pandemic, such as reducing the burden of hospitals, providing access to credible information, tracking symptoms and the mental health of individuals, and discovering new predictors of COVID-19.
Background: This article discusses the status, the challenges, the opportunities and the recommendations aimed at accelerating the rate of progress in the data driven management of cancer care. Two international workshops in June 2019, in Cordoba, Spain and then in October 2019, in Athens, Greece were organized by four HORIZON2020 EU funded projects: BOUNCE, CATCH ITN, DESIREE and MyPal. The issues covered include patient engagement, knowledge and data-driven decision support systems, patient journey, rehabilitation, personalized diagnosis, trust, assessment of guidelines, interoperability of ICT platforms and others. A series of recommendations is provided as the complex landscape of data-driven technical innovation in cancer care is being portrayed. Objective: Provide information on the current state of the art of technology and data-driven innovation for management of cancer care through the work of four EU H2020 funded projects. Methods: Reporting of the results of two international workshops on ICT in management of cancer care. Results: Technical and data-driven innovation provide promising tools for the management of cancer care. However, several challenges need to be successfully addressed, such as patient engagement, interoperability of ICT-based systems, knowledge management, trust and others. The paper analyses these challenges, which are seen as opportunities for further research and practical implementation and provides practical recommendations for future work. Conclusions: Technology and data-driven innovations are becoming an integral part of cancer care management. In the process, specific challenges need to be addressed, such as increasing trust and engaging the whole stakeholder ecosystem to fully benefit from what these innovations have to offer.
.
Chatbots, also known as conversation agents, are programs being able to simulate and reproduce an intelligent conversation with humans. Although this type of programs is not new, the explosion of the available information and the rapid increase of the users seeking for this information, has renewed the interest for their development. In this paper, we present R2D2 an intelligent chatbot, relying on semantic web technologies to enable intelligent question answering over the information available in DBPedia. The chatbot accepts structured input, allowing users to enter triple-pattern like queries, which are being answered by the underlying engine. While typing, an auto-complete service guides users on creating the triple-patterns, suggesting resources available in the DBPedia. Based on user input, the corresponding SPARQL queries are automatically formulated. The queries are submitted to the corresponding DBPedia SPARQL endpoint, the result is received by R2D2 and augmented with maps and visuals and eventually presented to the user. The usability evaluation performed shows the advantages of our solution and its usefulness.
Providing useful resources to patients is essential in achieving the vision of participatory medicine. However, the problem of identifying pertinent content for a group of patients is even more difficult than identifying information for just one. Nevertheless, studies suggest that the group-dynamics-based principles of behavior change have a positive effect on the patients' welfare. Along these lines, in this paper, we present a multidimensional recommendation model in the health domain using collaborative filtering. We propose a novel semantic similarity function between users, going beyond patient medical problems, considering additional dimensions such as the education level, the health literacy and the psycho-emotional status of the patients. Exploiting those dimensions we are interested in providing recommendations that are both high relevant and fair to groups of patients. Consequently, we introduce the notion of fairness and we present a new aggregation method, accumulating preference scores. We experimentally show that our approach is able to perform better recommendations to small group of patients for useful information documents.
Over the last decade, there have been many changes in the field of political analysis at a global level. Through social networking platforms, millions of people have the opportunity to express their opinion and capture their thoughts at any time, leaving their digital footprint. As such, massive datasets are now available, which can be used by analysts to gain useful insights on the current political climate and identify political tendencies. In this paper, we present TwiFly, a framework built for analyzing Twitter data. TwiFly accepts a number of accounts to be monitored for a specific time-frame and visualizes in real time useful extracted information. As a proof of concept, we present the application of our platform to the most recent elections of Greece, gaining useful insights on the election results.
As a result of recent advances in cancer research and “precision medicine” approaches, i.e. the idea of treating each patient with the right drug at the right time, more and more cancer patients are being cured, or might have to cope with a life with cancer. For many people, cancer survival today means living with a complex and chronic condition. Surviving and living with or beyond cancer requires the long-term management of the disease, leading to a significant need for active rehabilitation of the patients. In this paper, we present a novel methodology employed in the iManageCancer project for cancer patient empowerment in which personal health systems, serious games, psychoemotional monitoring and other novel decision-support tools are combined into an integrated patient empowerment platform. We present in detail the ICT infrastructure developed and our evaluation with the involvement of cancer patients on two sites, a large-scale pilot for adults and a small-scale test for children. The evaluation showed mixed evidences on the improvement of patient empowerment, while ability to cope with cancer, including improvement in mood and resilience to cancer, increased for the participants of the adults’ pilot.
A huge amount of data is generated each day from various sources. Analysis of these massive data is difficult, and requires new forms of processing to enable enhanced decision making, in-sight discovery and process optimization. In addition, besides their ever increasing volume, da-tasets change frequently, and as such, results to continuous queries have to be updated at short intervals. In this paper, we address the problem of evaluating continuous queries over big data streams that are frequently updated, adopting HIFUN, a high-level query language introduced recently. HIFUN offers a clear separation between the conceptual layer, where analytic queries are defined independently of the nature and location of data, and the physical layer where queries are evaluated, by encoding them as map-reduce jobs or as SQL group-by queries. Using HIFUN, we devise an algorithm for incremental processing of continuous queries, processing only the most recent data partition, and exploiting already computed information, without requiring evalu-ating the query over the complete dataset. Subsequently, we translate the generic algorithm to both SQL and MapReduce using SPARK, exploiting the query rewriting method provided by HIFUN. The experiments performed show the advantages of our solution in terms of query answering efficiency.
Diagnosis is one of the most important tasks of health providers with a tremendous impact on patient's health and well-being. Although many diagnostic decision support systems have been proposed so far, with several advantages and disadvantages, we have not yet seen approaches trying to integrate the decision of individual systems to improve the quality of the final diagnosis. To this direction, we present INTEGRA a novel decision support system, allowing multiple underlying diagnostic decision-support sub-systems to work in parallel, effectively integrating the individual decisions, increasing as such the quality of the final diagnosis. In addition, INTEGRA enables health providers to get explanation of the decisions based on the combined diagnosis and further explore the recommendations of the individual decision support sub-systems.
During the burst of the coronavirus pandemic, in early-midst 2020, public health authorities worldwide considered appropriate identification, isolation and contact tracing as the most appropriate strategy for infection containment. This work presents an outbreak response tool, designed for public health authorities, to effectively track suspect, probable and confirmed incidence cases in a pandemic by means of a mobile app used by citizens to provide immediate feedback. It is developed based on an already existing personal health record app, which has been extended to properly accommodate specific needs that emerged during the crisis. The aim is to better support human tracers and should not be confused with proximity tracking apps. It respects safety and security regulations, while at the same time it conforms to international standards and widely accepted medical protocols. Issues relevant to privacy concerns, and interoperability with available patient registries and data analytics tools are also examined to better support public healthcare delivery and contain the spread of the infection.
The early and accurate identification of a disease is important for its effective treatment. However, medical errors represent a serious problem and pose a threat to patient safety. To this direction, appropriate and continuous education of the medical personnel has been widely recognized as an important mean to reduce medical errors and increase the quality of the health system. In this paper, we present MediExpert, an expert system targeting on continuous and remote/online education of health personnel, providing also guidelines to persons that either cannot easily move due to age related comorbidities, or because they are away from healthcare units, further recommending users to talk with their doctors. It is based on differential diagnosis, employs ontologies for effective classification of health related problems and intelligent algorithms to enhance continuous education, offering as such a useful tool to both health providers and patients, allowing educating them at their own pace and location. We present the various components of the system and we elaborate on the benefits gained when using it for education.
Many modern applications produce massive streams of data series that need to be analyzed, requiring efficient similarity search operations. However, the state-of-the-art data series indexes that are used for this purpose do not scale well for massive datasets in terms of performance, or storage costs. We pinpoint the problem to the fact that existing summarizations of data series used for indexing cannot be sorted while keeping similar data series close to each other in the sorted order. This leads to two design problems. First, traditional, efficient algorithms for bulk-loading and updating an index cannot be used because they rely on sorting. Instead, index construction and updates takes place through slow top-down insertions, which create a non-contiguous index that results in many random I/Os. This also prevents an index from maintaining temporal partitions of the data to support window queries, as the cost of merging temporal partitions via top-down insertions (rather than through sorting) is prohibitive. Second, data series cannot be sorted and split across nodes evenly based on their median value; thus, most leaf nodes are in practice nearly empty. This further slows down query speed and amplifies storage costs. To address these problems, we present Coconut, the first data series index based on sortable summarizations, and the first efficient solution for indexing and querying streaming series. The first innovation in Coconut is an inverted, sortable data series summarization that organizes data series based on a z-order curve, keeping similar series close to each other in the sorted order. As a result, Coconut is able to use bulk-loading and updating techniques that rely on sorting to quickly build and maintain a contiguous index using large sequential disk I/Os. We then explore prefix-based and median-based splitting policies for bottom-up bulkloading, showing that median-based splitting outperforms the state of the art, ensuring that all nodes are densely populated. Finally, we explore the impact of sortable summarizations on variable size window queries, showing that they can be supported in the presence of updates through efficient merging of temporal partitions. Overall, we show analytically and empirically that Coconut dominates the state-of-theart data series indexes in terms of construction speed, query speed, and storage costs
The explosion in the amount of the available RDF data has lead to the need to explore, query and understand such data sources. Due to the complex structure of RDF graphs and their heterogeneity, the exploration and understanding tasks are significantly harder than in relational databases, where the schema can serve as a first step toward understanding the structure. Summarization has been applied to RDF data to facilitate these tasks. Its purpose is to extract concise and meaningful information from RDF knowledge bases, representing their content as faithfully as possible. There is no single concept of RDF summary, and not a single but many approaches to build such summaries; each is better suited for some uses, and each presents specific challenges with respect to its construction. This survey is the first to provide a comprehensive survey of summarization method for semantic RDF graphs. We propose a taxonomy of existing works in this area, including also some closely related works developed prior to the adoption of RDF in the data management community; we present the concepts at the core of each approach and outline their main technical aspects and implementation. We hope the survey will help readers understand this scientifically rich area and identify the most pertinent summarization method for a variety of usage scenarios.
Many modern applications produce massive streams of data series and maintain them in indexes to be able to explore them through nearest neighbor search. Existing data series indexes, however, are expensive to operate as they issue many random I/Os to storage. To address this problem, we recently proposed Coconut, a new infrastructure that organizes data series based on a new sortable format. In this way, Coconut is able to leverage state-of-the-art indexing techniques that rely on sorting for the first time to build, maintain and query data series indexes using fast sequential I/Os. In this demonstration, we present Coconut Palm, a new exploration tool that allows to interactively combine different indexing techniques from within the Coconut infrastructure and to thereby seamlessly explore data series from across various scientific domains.We highlight the rich indexing design choices that Coconut opens up, and we present a new recommender tool that allows users to intelligently navigate them for both static and streaming data exploration scenarios.
Background: Mobile health (mHealth) technology has the potential to play a key role in improving the health of patients with chronic non-communicable diseases. Objectives: We present a review of systematic reviews of mHealth in chronic disease management, by showing the features and outcomes of the studied interventions, along with associated challenges in this rapidly growing field. Methods: We searched the bibliographic databases of PubMed, Scopus, and Cochrane to identify systematic reviews of mHealth interventions with advanced technical capabilities (e.g., Internet-linked apps, interoperation with sensors, communication with clinical platforms, etc.) utilized in randomized clinical trials. Reviewed studies were synthesized according to their intervention features, the targeted diseases, the primary outcome, the number of participants and their average age, as well as the total follow-up duration. Results: We identified 5 reviews respecting our inclusion and exclusion criteria, which examined 30 mHealth interventions. The highest percentage of the interventions targeted patients with diabetes (n=19, 63%), followed by patients with psychotic disorders (n=7, 23%), lung diseases (n=3, 10%), and cardiovascular disease (n=1, 3%). 14 studies showed effective results: 9 in diabetes management, 2 in lung function, and 3 in mental health. Significantly positive outcomes were reported in 8 interventions (n=8, 47%) from 17 studies assessing glucose concentration, one intervention assessing physical activity, 2 interventions (n=2, 67%) from 3 studies assessing lung function parameters, and 3 mental health interventions assessing N-back performance, medication adherence, and number of hospitalizations. Divergent features were adopted in 14 interventions with significantly positive outcomes, such as personalized goal setting (n=10, 71%), motivational feedback (n=5, 36%), and alerts for caregivers health professionals (n=3, 21%). The most significant found challenges in the development and evaluation of mHealth interventions include the design of studies with high quality, the construction of robust standalone interventions and in combination with health professional inputs, and the identification of tools and methods to improve patient adherence. Conclusions: This review found mixed evidence regarding the health benefits of mHealth interventions for patients living with chronic diseases. Further rigorous studies are needed to assess the outcomes of personalized mHealth interventions toward the optimal management of chronic diseases.
In the present study, we present a model-based reference system for predicting resilience in silico, as part of personalizing precision medicine, to better under-stand the needs for improved therapeutic protocols of each patient is proposed. The BOUNCE environment will help clinicians and care-givers to predict the pa-tient’s resilience trajectory throughout cancer continuum. The overall proposed system architecture contributes to clinical outcomes and patient well-being by tak-ing into consideration biological, social, environmental, occupational and lifestyle factors for resilience prediction in women with breast cancer. Supervised learning algorithms are adopted with the inherent ability to represent the time-varying be-haviour of the underlying system which allows for a better representation of spa-tiotemporal input-output dependencies. The in silico resilience prediction ap-proach accommodates numerous diverse factors contributing to multi-scale mod-els of cancer, in order to better specify clinically useful aspects of recovery, treatment and intervention.
The huge diversity and quantity of data and information, and the requirements for knowledge extraction out of them put new challenges for knowledge management, synthesis, conflict detection and reasoning This paper elaborates on the design and development of COSMOS, an intelligent system which supports a) collaboration of experts for developing common knowledge bases and b) diagnosis derivation which takes into account time and uncertainty. The health domain is used for illustration and discussion of the features of our approach. Initially, we present the syntax and the semantics of the rules which incorporate time (temporal rules) and the data items upon which reasoning is performed. Then we introduce its inference engine, able to perform reasoning on top of rules and data, handling also the embedded time and uncertainty. We proceed further to define a conflict detection policy for supporting the difficult and error prone task of rule generation. The complexities of the aforementioned tasks are hidden from users via a well-designed and user friendly web interface that possesses strong collaboration features enabling multiple experts to work on defining a rule and on developing a common knowledge base. Evaluation of COSMOS has been performed using a) students studying expert systems and b) health experts in order to demonstrate the usability of the approach and the considerable advantages gained. To the best of our knowledge, COSMOS is one of the very few systems combining temporal rules, a powerful inference engine handling uncertainty and conflict detection and progresses beyond state of the art by adopting strong collaboration features and paradigms.
Anxiety and stress are very common symptoms of patients facing a forthcoming surgery. However, limited time and resources within healthcare systems make the provision of stress relief interventions difficult to provide. Research has shown that provision of preoperative stress relief and educational resources can improve health outcomes and speed recovery. Information and Communication Technology (ICT) can be a valuable tool in providing stress relief and educational support to patients and family before but also after an operation, enabling better self-management and self-empowerment. To this direction, this paper reports on the design of a novel technical infrastructure for a resilience support tool for improving the health condition of patients, during the care path, using Virtual Reality (VR). The designed platform targets, among others, at improving the knowledge on the patient data, effectiveness and adherence to treatment, as well as providing for effective communication channels between patients and clinicians.
Better information management is the key to a more intelligent health and social system. To this direction, many challenges must be first overcome, enabling seamless, effective and efficient access to the various health data sets and novel methods for exploiting the available information. The First International Workshop on Semantic Web Technologies for Health Data Management aimed at bringing together an interdisciplinary audience interested in the fields of semantic web, data management and health informatics to discuss the unique challenges in health-care data management and to propose novel and practical solutions for the next generation data-driven health-care systems. In this paper we summarize the outcomes of the first instance of the workshop, and we present interesting conclusions and key messages..
The explosion in the amount of the RDF on the Web has lead to the need to explore, query and understand such data sources. The task is challenging due to the complex and heterogeneous structure of RDF graphs which, unlike relational databases, do not come with a structure-dictating schema. Summarization has been applied to RDF data to facilitate these tasks. Its purpose is to extract concise and meaningful information from RDF knowledge bases, representing their content as faithfully as possible. There is no single concept of RDF summary, and not a single but many approaches to build such summaries; the summarization goal, and the main computational tools employed for summarizing graphs, are the main factors behind this diversity. This tutorial presents a structured analysis and comparison existing works in the area of RDF summarization; it is based upon a recent survey which we co-authored with colleagues. We present the concepts at the core of each approach, outline their main technical aspects and implementation. We conclude by identifying the most pertinent summarization method for different usage scenarios, and discussing areas where future effort is needed.
Database schemata are not static artifacts but subject to continuous change as new requirements daily occur and the modeling choices of the past should be updated or adapted. To this direction, multiple approaches available already, try to keep multiple co-existing schema versions in parallel or model schema evolution through Schema Modification Operations, known as SMOs. However, to the best of our knowledge, in the era of big data, where thousands of SMOs might appear, it is really hard for developers to identify the modeling choices of the past and to identify how a specific column or table has been evolved. In this demo, we present VESEL, the first system enabling the VisuaL Exploration of Schema Evolution using provenance queries. Our approach relies on a state of the art database evolution language, and can efficiently answer queries about when a specific table or column has been introduced, how - with which SMO operation and why - which is the sequence of changes that led to the creation of the specific table/column. In the demonstration we will present the architecture of our system and the various algorithms implemented, enabling end-users to visually explore schema evolution. Then we will allow conference participants to interact directly with the system to test its capabilities.
The widespread usage of electronic health record (EHR) systems and the prevalence of personal health apps has led to the digitization of huge quantities of health and medical data. However, the lack of plug and play interoperability between them is a major bottleneck, significantly restricting their exploitation potential. This paper presents a novel infrastructure developed for enabling the seamless integration between such systems, focusing on the use of state-of-the-art, widely accepted, interfaces. The infrastructure effectively supports end-user needs for integrated care, bringing linked EHR information from multiple providers at the point of care by means of personal health apps. Integration is provided through cross-enterprise document sharing (XDS) and fast healthcare interoperability resources (FHIR) services. The described infrastructure facilitates interoperability by enabling citizens to use a personal health app of their choosing for accessing own medical record information stored at multiple sites, by combining the benefits and potential of both XDS and FHIR.
Breast cancer is the most common cancer disease in women and is rapidly becoming a chronic illness due recent advances in treatment methods. As such, coping with cancer has become a major socio-economic challenge leading to an increasing need for predicting resilience of women to the variety of stressful experiences and practical challenges they face. In this paper, we present the data infrastructure developed for this purpose, demonstrating the various components that will contribute to the developing the resilience trajectory predictor. Special emphasis is given to the semantic tier, presenting the project solution already implemented for effectively collecting, ingesting, cleaning, modelling and processing data that will be used throughout the lifetime of the project.
In the present study, a reference system for predicting resilience in silico, as part of personalizing precision medicine, to better understand the needs for improved therapeutic protocols of each patient is proposed. The BOUNCE environment will help clinicians and care-givers to predict the patient’s resilience trajectory throughout cancer continuum. The overall proposed system architecture contributes to clinical outcomes and patient well-being by taking into consideration biological, social, environmental, occupational and lifestyle factors for resilience prediction in women with breast cancer. Supervised learning algorithms are adopted with the inherent ability to represent the time-varying behaviour of the underlying system which allows for a better representation of spatiotemporal input-output dependencies. The in silico resilience prediction approach accommodates numerous diverse factors contributing to multi-scale models of cancer, in order to better specify clinically useful aspects of recovery, treatment and intervention.
Processing and managing sensitive health data requires a high standard of security and privacy measures to ensure that all ethical and legal requirements are respected. Data anonymization is one of the key technologies to this purpose. However, the plethora and the complexity of the available anonymization techniques make it difficult for a non-expert to select and apply the appropriate technique. In this paper, we report on Shiny Database Anonymizer, a tool enabling the easy and flexible anonymization of available health data, providing access to state of the art anonymization techniques, incorporating also multiple data analysis visualization paradigms. In addition, a number of encryption and hashing techniques are presented.
The explosion of the web and the abundance of linked data, demand for effective and efficient methods for storage, management and querying. Apache Spark is one of the most widely used engines for big data processing, with more and more systems adopting it, for efficient querying over distributed data. Specifically for query answering over RDF data, existing works partition triples using simplistic horizontal and/or vertical partitioning of triples . This results in poor query answering performance. The main reason for this is that simplistic data partitioning, fails to identify data locality and group data that are usually queried together. We show this both analytically and experimentally and we present LAWA, a novel platform that accepts as input an RDF dataset and effectively partitions data, maximizing data locality. To achieve this, LAWA first identifies the top-k most important nodes as centroids and distributes the other schema nodes to the centroid they mostly depend on. Then it allocates the corresponding instance nodes to the schema nodes they are instantiated under. This method, named Locality Aware Partitioning (LAP), in most cases results in a balanced data distribution over the generated partitions, however, without offering any guarantees on it. In order to further study the trade off between data locality and data distribution we introduce another variant, the Bounded Locality Aware Partitioning (BLAP), enforcing a balanced data distribution over the generated partitions. Based on the aforementioned partitioning methods, we implement an indexing scheme in order to speed up query processing. Using the constructed index LAWA can easily located the primary partition that the queried classes and instances reside. Then, we provide a query execution engine, on top of Spark SQL module, that exploits the generated indexes and builds the query to be finally evaluated by Spark. We show that our approach offers an optimal fine-tuning between data distribution, replication and data access reduction, dominating existing approaches. More specifically, we evaluate our approach using both synthetic and real workloads, showing that we improve query answering in most of the cases orders of magnitude over existing state of the art.
Real-time analytics have received significant attention in recent years, being one of the most active research areas with a number of challenging problems. Storing and rapidly analyzing large amounts of data enables better decision making and as such the research community has responded by proposing formal models and frameworks enabling large-scale distributed real-time data analytics. A major challenge in this context is the rapid processing of continuous queries. A continuous query is a query which is evaluated automatically and periodically over a dataset that changes in time [1]. The results of continuous queries are usually fed to dashboards in large enterprises to provide support in the decision making process. As new data and updates are arriving at a high rate, the data sets grow rapidly and re-evaluation of the query incurs delays. Therefore the problem is how to evaluate the query incrementally that is given the answer of the query at time t, on dataset D, how to find the answer of the query at time t’ on dataset D’, assuming that the answer at time t has been saved and results become stale and stagnant over a time. Incremental processing is an auspicious approach to refreshing mining results as it uses previously saved results to avoid the cost of re-computation from scratch. We study this problem in the context of HIFUN, a recently proposed high level functional language of analytic queries [2]. Two distinctive features of HIFUN are that (a) analytic queries and their answers are defined and studied in the abstract, independently of the structure and location of the data and (b) each HIFUN query can be mapped either as a SQL group-by query or as a Map-Reduce job. In this work we follow the HIFUN approach: we study algorithms for the incremental evaluation of continuous queries in the abstract and then map them to concrete algorithms using SQL or Map-Reduce based platforms (namely, Hadoop and Spark). Our objective is to study the performance of these algorithms and their scaling in the context of real systems.
Many modern applications produce massive amounts of data series that need to be analyzed, requiring effiient similarity search operations. However, the state-of-the-art data series indexes that are used for this purpose do not scale well for massive datasets in terms of performance, or storage costs. We pinpoint the problem to the fact that existing summarizations of data series used for indexing cannot be sorted while keeping similar data series close to each other in the sorted order. This leads to two design problems. First, traditional bulk-loading algorithms based on sorting cannot be used. Instead, index construction takes place through slow top-down insertions, which create a non-contiguous index that results in many random I/Os. Second, data series cannot be sorted and split across nodes evenly based on their median value; thus, most leaf nodes are in practice nearly empty. This further slows down query speed and amplifis storage costs. To address these problems, we present Coconut. The first innovation in Coconut is an inverted, sortable data series summarization that organizes data series based on a z-order curve, keeping similar series close to each other in the sorted order. As a result, Coconut is able to use bulk-loading techniques that rely on sorting to quickly build a contiguous index using large sequential disk I/Os. We then explore prefix-based and median-based splitting policies for bottom-up bulk-loading, showing that median-based splitting outperforms the state of the art, ensuring that all nodes are densely populated. Overall, we show analytically and empirically that Coconut dominates the state-of-the-art data series indexes in terms of construction speed, query speed, and storage costs.
Chronic pain is one of the most common health problems affecting daily activity, employment, relationships and emotional functioning. Unfortunately, limited access to pain experts, the high heterogeneity in terms of clinical manifestation and treatment results, contribute in failure to manage efficiently and effectively pain. Information and Communication Technology (ICT) can be a valuable tool, enabling better self-management and self-empowerment of pain. To this direction, this paper reports on the design of a novel technical infrastructure for chronic pain self-management based on an Intelligent Personal Health Record platform. The designed platform targets, among others, at improving the knowledge on the patient data, effectiveness and adherence to treatment and providing effective communication channels between patients and clinicians.
The explosion of the web and the abundance of linked data demand for effective and efficient methods for storage, management and querying. More specifically, the everincreasing size and number of RDF data collections raises the need for efficient query answering, and dictate the usage of distributed data management systems for effectively partitioning and querying them. To this direction, Apache Spark is one of the most active big-data approaches, with more and more systems adopting it, for efficient, distributed data management. The purpose of this paper is to provide an overview of the existing works dealing with efficient query answering, in the area of RDF data, using Apache Spark. We discuss on the characteristics and the key dimension of such systems, we describe novel ideas in the area, and the corresponding drawbacks, and provide directions for future work.
Nowadays, the number of people who search for information related to health has significantly increased, while the time of health professionals for recommending online useful sources of information has been reduced to a great extend. FairGRecs aims to offer valuable information to users, in the form of suggestions, via their caregivers, and improve as such the opportunities that users have to inform themselves online about health problems and possible treatments. Specifically, we propose a model for group recommendations incorporating the notion of fairness, following the collaborative filtering approach. For computing similarities between users, we define a novel measure that is based on the semantic distance between users’ health problems. Our special focus is on providing valuable suggestions to a caregiver who is responsible for a group of users. We interpret valuable suggestions as ones that are both highly related and fair to the users of the group. As such, we introduce a new aggregation design, incorporating fairness, and we compare it with current state-of-the-art. Our experiments demonstrate the advantages of both the semantic similarity measure and the fair aggregation design.
The huge diversity, big quantity of data and information, and the requirements for knowledge extraction out of them put new challenges for knowledge management, synthesis, conflict detection and reasoning. In this paper, we present COSMOS, a knowledge system that fully addresses these challenges, in an efficient way, paving the way for a new generation of knowledge systems. Using our approach, it is possible for domain experts to generate temporal knowledge rules. As those rules are saved to our knowledge base, a conflict detection mechanism detects and solves rule conflicts. Then, an inference engine is able to perform efficiently, accurate decisions, based on available factual information using reasoning and handling uncertainty. Ontologies are used to model both the factual information and the data items in the rules enabling also interoperability with existing systems. To validate our approach, as an application scenario, we deploy our infrastructure in a health environment where doctors provide rules that are activated over a patient health record. Preliminary results indicate the benefits of our approach for decision support based on health data, successfully identifying adverse events and enabling intelligent patient monitoring.
Significant efforts have been dedicated recently to the development of architectures for storing and querying RDF data in distributed environments. Several approaches focus on data partitioning, which are able to answer queries efficiently, by using a small number of computational nodes. However, such approaches provide static data partitions. Given the increase on the continuous and rapid flow of data, nowadays there is a clear need to deal with streaming data. In this work, we propose a framework for incremental data partitioning by exploiting machine learning techniques. Specifically, we present a method to learn the structure of a partitioned database, and we employ two machine learning algorithms, namely Logistic Regression and Random Forest, to classify new streaming data.
Semantic Web and ontology engineering can play significant role in the area of education. In this paper we focus on the conceptualization of educational knowledge structures in an academic setting. More specifically, we present the methodology and the development process of an educational ontology. The de-veloped ontology can be reused and applied to any type of course in different in-stitutions and contribute to several curriculum tasks and course activities.
Ontologies are constantly evolving as new requirements daily occur and the modeling choices of the past should be updated or adapted. Rapidly understanding and exploring this evolution will enable the ontology engineers to understand the modeling choices of the past and maintainers of the depending artifacts to make decisions about possible changes. However, recent research focuses only on detecting changes between ontology versions, overloading end-users with hundreds or even thousands of changes between ontology versions, making it impossible to explore what has happened. To this direction, in this paper, we present EvoRDF, a novel framework for exploring ontology evolution using provenance queries. Our approach uses a high-level language of changes and effectively answers queries about when a specific resource and how – by which change operations. Even more, why queries can identify the sequence of changes that led to the creation of a specific resource in the latest ontology version or track the evolution of a specific resource from a past ontology version. The evaluation performed shows the feasibility of our solution and the great advantages gained.
Developments in information and communication technology have changed the way healthcare processes are experienced by both patients and healthcare professionals: more and more services are now available through computers and mobile devices. Smartphones are becoming useful tools for managing one’s health, and today, there are many available apps meant to increase self-management, empowerment and quality of life. However, there are concerns about the implications of using mHealth and apps: data protection issues, concerns about sharing information online, and the patients’ capacity for discerning effective and valid apps from useless ones. The new general data protection regulation has been introduced in order to give uniformity to data protection regulations among European countries but shared guidelines for mHealth are yet to develop. A unified perspective across Europe would increase the control over mHealth exploitation, making it possible to think of mHealth as effective and standard tools for future medical practice.
Patients today have a wealth of information available on the internet. Despite, the potential benefits of internet health information seeking, several concerns have been raised about the quality of in-formation and about the patient’s capability to evaluate medical information and to relate it to their own disease and treatment. As such, novel tools are required able to effectively guide patients and provide high quality medical information in an intelligent and personalized manner. To this direction, this paper presents a platform, trying to empower patients by enabling them to search in a high quality document repository selected by experts, avoiding the information overload of the internet. In addition, the information provided to the patients is personalized, based on indi-vidual preferences, medical conditions and other profiling information. Despite the generality of our approach, we apply our solution to a Personal Health Record constructed for the cancer patients and we report our initial findings. To the best of our knowledge, our platform is the only one combin-ing advanced natural language processing, ontologies and personal information to offer a unique user experience.
n the last decade, clinicians have started to shift from an individualistic perspective of the patient towards family-centred models of care, due to the increasing evidence from research and clinical practice of the crucial role of significant others in determining the patient's adjustment to cancer disease and management. eHealth tools can be considered a means to compensate the services gap and support outpatient care flows. Within the works of the European H2020 iManageCancer project, a review of the literature in the field of family resilience was conducted, in order to determine how to monitor the patient and his/her family's resilience through an eHealth platform. An analysis of existing family resilience questionnaires suggested that no measure was appropriate for cancer patients and their families. For this reason, a new family resilience questionnaire (named FaRe) was developed to screen the patient's and caregiver's psycho-emotional resources. Composed of 24 items, it is divided into four subscales: Communication and Cohesion, Perceived Family Coping, Religiousness and Spirituality, and Perceived Social Support. Embedded in the iManageCancer eHealth platform, it allows users and clinicians to monitor the patient's and the caregivers' resilience throughout the cancer trajectory.
Clinical decision support systems can play a crucial role in healthcare delivery as they promise to improve health outcomes and patient safety, reduce medical errors and costs and contribute to patient satisfaction. Used in an optimal way, they increase the quality of healthcare by proposing the right information and intervention to the right person at the right time in the healthcare delivery process. This paper reports on a specific approach to integrated clinical decision support and patient guidance in the cancer domain as proposed by the H2020 iManageCancer project. This project aims at facilitating efficient self-management and management of cancer according to the latest available clinical knowledge and the local healthcare delivery model, supporting patients and their healthcare providers in making informed decisions on treatment choices and in managing the side effects of their therapy. The iManageCancer platform is a comprehensive platform of interconnected mobile tools to empower cancer patients and to support them in the management of their disease in collaboration with their doctors. The backbone of the iManageCancer platform comprises a personal health record and the central decision support unit (CDSU). The latter offers dedicated services to the end users in combination with the apps iManageMyHealth and iSupportMyPatients. The CDSU itself is composed of the so-called Care Flow Engine (CFE) and the model repository framework (MRF). The CFE executes personalised and workflow oriented formal disease management diagrams (Care Flows). In decision points of such a Care Flow, rules that operate on actual health information of the patient decide on the treatment path that the system follows. Alternatively, the system can also invoke a predictive model of the MRF to proceed with the best treatment path in the diagram. Care Flow diagrams are designed by clinical experts with a specific graphical tool that also deploys these diagrams as executable workflows in the CFE following the Business Process Model and Notation (BPMN) standard. They are exposed as services that patients or their doctors can use in their apps in order to manage certain aspects of the cancer disease like pain, fatigue or the monitoring of chemotherapies at home. The mHealth platform for cancer patients is currently being assessed in clinical pilots in Italy and Germany and in several end-user workshops.
The advancements in healthcare have brought to the fore the need for flexible access to health-related information and created an ever-growing demand for efficient data management infrastructures. To this direction, in this paper, we present an effective and efficient data management infrastructure implemented for the iManageCancer EU project. The architecture focuses on enabling data access to multiple, heterogeneous and diverse data source that are initially available in a data lake. Parts of these data are integrated and semantically uplifted using a modular ontology. This integration can be either at run-time or through an ETL process ensuring efficient access to the integrated information. A unique feature of out platform is that it allows the uninterrupted, continuous evolution of ontologies/terminologies. Finally, summarization tools enable the quick understanding of the available information, whereas APIs and anonymization services ensure the secure access to the requested information.
Basic research and prospective clinical trials have led to higher cure rates of patients with cancer. Cancer is now frequently managed as a chronic disease. There is an increasing need for cancer patients to take an active, informed and leading role in their ongoing care to improve their physical, psychological, and social aspects of health thus resulting in a better quality of life. Identification of self-management processes for cancer can help to guide future research and clinical practice to improve patient’s outcome. In iMC a Personal Health Record platform (iPHR) is developed, were different tools for self-management are available including serious games for children and adults. Data security and privacy are part of the platform.
The vision of personalized medicine has led to an unprecedent-ed demand for acquiring, managing and exploiting health related information, which in turn has led to the development of many e-Health systems and applications. However, despite this increasing trend only a limited set of information is cur-rently being exploited for analysis and this has become a major obstacle towards the advancement of personalized medicine. To this direction, this paper presents the design and implemen-tation of a content aware health data-analytics framework. The framework enables first the seamless integration of the availa-ble data and their efficient management through big data management systems and staging environments. Then the integrated information is further anonymized at run-time and accessed by the data analysis algorithms in order to provide appropriate statistical information, feature selection correlation and clustering analysis.
Given the explosive growth in the size and the complexity of the Data Web, there is now more than ever, an increasing need to develop methods and tools in order to facilitate the understanding and exploration of RDF/S Knowledge Bases (KBs). To this direction, summarization approaches try to produce an abridged version of the original data source, highlighting the most representative concepts. Central questions to summarization are: how to identify the most important nodes and then how to link them in order to produce a valid sub-schema graph. In this paper, we try to answer the first question by revisiting six wellknown measures from graph theory and adapting them for RDF/S KBs. Then, we proceed further to model the problem of linking those nodes as a graph Steiner- Tree problem (GSTP) employing approximations and heuristics to speed up the execution of the respective algorithms. The performed experiments show the added value of our approach since a) our adaptations outperform current state of the art measures for selecting the most important nodes and b) the constructed summary has a better quality in terms of the additional nodes introduced to the generated summary.
The evolution of ontologies is a reality in current research community. The problem of understanding and exploring this evolution is a fundamental problem as maintainers of depending artifacts need to take a decision about pos-sible changes and ontology engineers need to understand the reasons for this evo-lution. Recent research focuses on identifying and statically visualizing deltas be-tween ontology versions using various low- or high-level language of changes. In this paper we argue that this is not enough and we provide a complete solution enabling the active, dynamic exploration of the evolution of RDF/S ontologies using provenance queries. To this direction we construct an ontology of changes for modeling the language of changes and we store all changes as instances of this ontology in a triple store. On top of this triple store two visualization mod-ules, one individual app and one protégé plugin allow the exploration of the evo-lution using provenance queries. To the best of our knowledge our approach is unique in allowing the dynamic exploration of the evolution using provenance queries.
During the last decade, the number of users who look for health-related information has impressively increased. On the other hand, health professionals have less and less time to recommend useful sources of such information online to their patients. To this direction, we target at streamlining the process of providing useful online information to patients by their caregivers and improving as such the opportunities that patients have to inform themselves online about diseases and possible treatments. Using our system, relevant and high quality information is delivered to patients based on their profile, as represented in their personal healthcare record data, facilitating an easy interaction by minimizing the necessary manual effort. Specifically, in this paper, we propose a model for group recommendations following the collaborative filtering approach. Since in collaborative filtering is crucial to identify the correct set of similar users for a user in question, in addition to the traditional ratings, we pay particular attention on how to exploit healthrelated information for computing similarities between users. Our special focus is on providing valuable suggestions to a caregiver who is responsible for a group of users. We interpret valuable suggestions as suggestions that are both highly related and fair to the users of the group. In this line, we propose an algorithm for identifying the top-z most valuable recommendations, and present its implementation in MapReduce.
As knowledge bases are constantly evolving, there is a clear need for monitoring and analyzing the changes that occur on them. Traditional approaches for studying the evolution of data focus on providing humans with deltas that include loads of information. In this work, we envision a processing model that recommends evolution measures taking into account particular socio-technical challenges, such as relatedness, transparency, diversity, fairness and anonymity. We target at supporting humans with complementary measures that offer high-level overviews of the changes in order to help them understand how data of interest evolve.
The ever-growing demand for acquiring, managing and exploiting patient health related information has led to the development of many e-Health systems and applications. However, despite the number of systems already developed and the apparent need for such systems, end users can only collect online and exploit, only a limited set of information for health purposes in the context of personalized, preventive and participatory medicine. To this direction, this paper initially presents the personal health record (PHR) concept, related work and best practices for the development of PHR systems in a standardized manner. It also outlines the proposal for meaningful use criteria in the United States (US) and the health level seven (HL7) personal health record system functional model (PHR-S FM). Focus is put on trying to link core functionality modules of the Integrated Care Solutions TM PHR system, designed to support the citizen, paying emphasis on wellbeing, home care and the management of chronic diseases with PHR-S FM personal health functions, in a preliminary effort towards the exploration of functional models to support interoperability. Based on accumulated experiences from many European Union (EU) research projects, the paper concludes by providing directions towards achieving wider PHR adoption and meaningful use.
Δεδομένης της εκρηκτικής αύξησης του μεγέθους και της πολυπλοκότητας των δεδομένων στο διαδίκτυο, υπάρχει τώρα περισσότερο από ποτέ, μια αυξανόμενη ανάγκη για την ανάπτυξη μεθόδων και εργαλείων προκειμένου να διευκολυνθεί η κατανόηση και η εξερεύνηση βάσεων δεδομένων RDF/S. Προς αυτή την κατεύθυνση, μέθοδοι δημιουργίας συνόψεων επιδιώκουν την παραγωγή μιας συνοπτικής έκδοσης της αρχικής πηγής δεδομένων, αναδεικνύοντας τις πιο αντιπροσωπευτικές έννοιες. Βασικά ερωτήματα για την παραγωγή μιας σύνοψης είναι: το πως θα προσδιοριστούν οι σημαντικότεροι κόμβοι ενός συνόλου και εν συνεχεία, το πώς θα συνδεθούν προκειμένου να παραχθεί ένας έγκυρος υπογράφος. Σε αυτή την εργασία, προσπαθούμε να απαντήσουμε το πρώτο ερώτημα με την χρήση έξι μέτρων σημαντικότητας από την θεωρία γράφων, και την προσαρμογή τους για βάσεις δεδομένων RDF/S. Έπειτα μοντελοποιούμε το πρόβλημα της διασύνδεσης των κόμβων ως ένα Δέντρο Στάινερ σε γράφημα, διερευνώντας προσεγγιστικούς αλγορίθμους και ευρηστικά τεχνάσματα για να επιταχύνουμε την εκτέλεση τους των αλγορίθμων. Τα πειράματα που διεξήχθησαν, δείχνουν την προστιθέμενη αξία της προσέγγισής μας δεδομένου ότι α) οι προσαρμογές μας ξεπερνούν τις κορυφαίες τρέχουσες τεχνικές για την επιλογή των πιο σημαντικών κόμβων και β) η παραγόμενη σύνοψη έχει καλύτερη ποιότητα, εισάγοντας μικρότερο αριθμό πρόσθετων κόμβων.
Over the last years we are experiencing an explosion of biomedical research tools, data formats and analysis methods. While the global scientific output is doubling every nine years, researches still use generic Google-like searches in order to locate useful tools or rely on specialized web forums to seek technical and analytical advices. It is not an exaggeration to state that science on that matter has not changed over the last 20 years. This is surprising given that the results of a plethora of published papers in experimental science are generated through an analysis pipeline of some kind. We introduce Calchas (calchas.ics.forth.gr), a web based framework that takes advantage of domain specific ontologies (e.g. the EDAM ontology is utilized edamontology.org), and Natural Language Processing (NLP), aiming to empower exploration of biomedical resources via semantic-based querying and search. The NLP engine analyzes the input free-text query (description of the facts/data, the research question and the desired output) and translates it into targeted queries with terms from the underlying ontology. Each query is passed to the semantically-annotated tools repository, and based on similarity matches, it ranks the available resources. The current version of Calchas is an improvement, both in terms of the included resources and functionality, of an older version of it (Sfakianaki et al., 2015). In addition, Calchas features a rich GUI to present search results as a network of tools and input/output entries. The recommended tools are mapped on a directed graph where each node represents a data type or data format and each edge a tool that supports as input and output the source and target node, respectively. Different edge (tool) colors indicate whether the tool supports both the requested input and output (green), only the input (blue) or only the output (orange). Such a visualization provides a direct overview of putative pipelines that could serve the research quest, based on the identified tools and the engaged input/output formats. Other features include interaction with the network allowing to explore descriptions of the tools and matching scores of the respective nodes, and edges as well. Calchas tries to link the gap between research question and efficient dynamic biomedical resources discovery. Currently, Calchas supports more than 6000 tools from the bioinformatics domain, as managed by the Elixir tools and data services registry (https://bio.tools). Work is also in progress to include resources from the FAIRDOM repository (https://fair-dom.org) aiming to offer a FAIR-compliant resource discovery service in the life sciences domain.
Intelligent tutoring systems (ITS) incorporate techniques for transferring knowledge and skills to students. These systems use a combination of computer-aided instruction methods and artificial intelligence. In this paper we present a web-based intelligent tutoring system for learning Prolog. We present the architecture of our system and we provide details on each one of its modules. Each lesson includes the corresponding lecture with theory and exercises, a practice module where students can apply the corresponding theory and an assessment module to verify user’s understanding. The system can be used with or without a teacher enabling distant learning. Among the novelties of our system is its flexibility to adapt to individual student choices and profile, offering a wide range of alternatives and trying to continuously keep the interest of the final user. The preliminary evaluation performed confirms the usability of our system and the benefits of using it for learning Prolog.
Cancer research has led to more cancer patients being cured, and many more enabled to live with their cancer. As such, some cancers are now considered a chronic disease, where patients and their families face the challenge to take an active role in their own care and in some cases in their treatment. To this direction the iManageCancer project aims to provide a cancer specific self-management platform designed according to the needs of patient groups while focusing, in parallel, on the wellbeing of the cancer patient. A special emphasis was given to avoidance, early detection and management of adverse events of cancer therapy but also, importantly, on psycho-emotional evaluation and self-motivated goals. In this paper, we present the use-case requirements and then we explain the corresponding system architecture. We describe in detail the main technological components of the designed platform, and we discuss further directions of research.
This visual paper aims at proposing a framework for detecting depression in cancer patients using prosodic and statistical features extracted by speech, while chatting with an augmented reality virtual coach.
It is well documented that the diagnosis of cancer affects the wellbeing of the whole family adding overwhelming stresses and uncertainties. Having regard to that family education and enhancement of resilience is an important factor that should be promoted and facilitated in a holistic framework for addressing a severe and chronic condition such as cancer. In this paper we review shortly the notion of resilience in the literature identifying three tools that try to support family resilience. Then we focus in the cancer domain and we describe a unique tool implemented to this direction. To our knowledge, this is the first time such a tool will be used to complete patient profile with family resilience information eventually leading to patient and family engagement and empowerment.
As the number of clinical guidelines and rules for effective management of cancer therapy is rapidly increasing decision support systems are more and more required. To this direction, in this paper, we present a collaborative knowledge management system for cancer diseases leading to decision support and intelligent diagnosis. Clinicians can specify a variety of knowledge rules in a collaborative fashion. Then, those rules are applied on top of patient data collected within a personal health record. The generated knowledge is formulated as a free text and returned back to the clinicians to support them and enhance the communication with their patients.
Purpose. Biomedical research is being catalyzed by the vast amount of data rapidly collected through the application of information technologies (IT). Despite IT advances, the methods for involving patients and citizens in biomedical research remain static, paper-based and organized around national boundaries and anachronistic legal frameworks. The purpose of this paper is to study the current situation in the European Union and identify the requirements for building effective IT systems enabling the secondary use of data and biomaterial. Method. We review existing European legislation for secondary use of patient’s biomaterial and data for research, identify types and scopes of consent, formal requirements for consent, and consider their implications for implementing electronic consent tools. Results. In the light of our analysis of the European legislation requirements, we identified preconditions for building an effective tool enabling secondary usage of both patient’s data and biomaterial. We proceeded to develop a modular tool, named Donor’s Support Tool (DST), designed to connect researchers with participants, and to promote engagement, informed participation and individual decision making. As a proof of concept, we then deployed the tool in a real clinical environment; here we report our experiences showing the significant advantages and the potential of our approach.
Today we are witnessing an explosion in the size and the amount of the available RDF datasets. As such, conventional single node RDF management systems give their position to clustered ones. However most of the currently available clustered RDF database systems partition data using hash functions and/or vertical and horizontal partition algorithms with a significant impact on the number of nodes required for query answering, increasing the total cost of query evaluation. In this paper we present a novel semantic partitioning approach, exploiting both the structure and the semantics of an RDF Dataset, for producing vertical partitions that significantly reduce the number of nodes that should be visited for query answering. To construct these partitions, first we select the most important nodes in a dataset as centroids, using the notion of relevance. Then we use the notion of dependence to assign each remaining node to the appropriate centroid. We evaluate our approach using three real world datasets and demonstrate the nice properties that the constructed partitions possess showing that they significantly reduce the total number of nodes required for query answering while introducing minimal storage overhead.
Given the explosive growth in both data size and schema complexity, data sources are becoming increasingly difficult to use and comprehend. Summarization aspires to produce an abridged version of the original data source highlighting its most representative concepts. In this paper, we present an advanced version of the RDF Digest, a novel platform that automatically produces and visualizes high quality summaries of RDF/S Knowledge Bases (KBs). A summary is a valid RDFS graph that includes the most representative concepts of the schema, adapted to the corresponding instances. To construct this graph we designed and implemented two algorithms that exploit both the structure of the corresponding graph and the semantics of the KB. Initially we identify the most important nodes using the notion of relevance. Then we explore how to select the edges connecting these nodes by maximizing either locally or globally the importance of the selected edges. The extensive evalua-tion performed compares our system with two other systems and shows the benefits of our approach and the considerable advantages gained.
As the internet grows daily and millions of news articles are produced everyday worldwide by various sources, the need to store, index, search and explore news articles is more than prominent. In this paper we present an integrated platform dedicated to news articles, providing storage, indexing and searching functionalities, implemented using semantic web technologies and services. Besides using the developed APIs, the users through intuitive graphical user interfaces can save articles from RSS channels, import them through wrappers from external news sites or manually insert them using forms. A search engine on top allows the users to explore all registered information. All components have been implemented using semantic web technologies, using a novel ontology to model the news domain, a triple store for the management of data and web services exchanging JSON-DL messages. The registered articles become automatically part of the Linked Open Data cloud, enabling better data and knowledge sharing. Our preliminary evaluation shows the high-quality of the developed platform and the benefits of our approach.
As users struggle to navigate on the vast amount of information now available, methods and tools for enabling the quick exploration of the databases content is of paramount importance. To this direction we present Apantisis, a novel question answering system implemented for the Greek language ready to be attached to any external database/knowledge-base. An ingestion module enables the semi/automatic construction of the data dictionary that is used for question answering whereas the Greek Language Dictionary, the Syntactic and the Semantic Rules are also stored in an internal, extensible knowledge base. After the ingestion phase, the system is accepting questions in natural language, and automatically constructs the corresponding relational algebra query to be further evaluated by the external database. The results are then formulated as free text and returned to the user. We highlight the unique features of our system with respect to the Greek language and we present its implementation and a preliminary evaluation. Finally, we argue that our solution is flexible and modular and can be used for improving the usability of traditional database systems.
The objective of the INTEGRATE project (http://www.fp7-integrate.eu/) that has recently concluded successfully was the development of innovative biomedical applications focused on streamlining the execution of clinical research, on enabling multidisciplinary collaboration, on management and large-scale sharing of multi-level heterogeneous datasets, and on the development of new methodologies and of predictive multi-scale models in cancer. In this paper, we present the way the INTEGRATE consortium has approached important challenges such as the integration of multi-scale biomedical data in the context of post-genomic clinical trials, the development of predictive models and the implementation of tools to facilitate the efficient execution of postgenomic multi-centric clinical trials in breast cancer. Furthermore, we provide a number of key “lessons learned” during the process and give directions for further future research and development.
Background:
The European eHealthMonitor project (eHM) developed a user-sensitive and interactive
web portal for the dementia care setting: The eHM Dementia Portal (eHM-DP). It aims to
provide targeted support for informal caregivers of persons with dementia and
professionals.
Objective:
The objective of this study was to assess the usefulness and impact of the eHM-DP
service in the dementia care setting from two user perspectives: informal caregivers and
professionals.
Methods:
The evaluation study was conducted from June to September 2014 and followed a
before-after, user-participatory, mixed-method design with questionnaires and
interviews. The used intervention was the eHM-DP: an interactive web portal for
informal caregivers and professionals that was tested for a 12-week period. Primary
outcomes for caregivers included: empowerment, quality of life, caregiver burden,
decision aid as well as perceived usefulness and benefits of the eHM-DP. Primary
outcomes for professionals involved decision aid as well as perceived usefulness and
benefits of the eHM-DP.
Results:
A total number of 25 informal caregivers and 6 professionals used the eHM-DP over the
12 weeks study period. Both professionals and informal caregivers indicated perceived
benefits and support by the eHM-DP. In total 65% (n=16) of informal caregivers would
use the eHM-DP if they had access to it. Major perceived benefits (65%; n=16 of informal
caregivers) were individualized information acquisition, improved interaction between
informal caregivers and professionals, access to support from home and empowerment
in health related decisions (PrepDM Score: 67.9). Professionals highlighted the
improved treatment and care over the disease course (83%, n=5) and improved health
care access for people living in rural areas (67%; n=4). However, there was no
improvement in caregiver burden (BSFC) and quality of life (EQ-5D-5L) over the study
period.
Conclusions:
Our study provides insight into the different user perspectives on an eHealth support
service in the dementia treatment and care setting. The presented results are of
importance for future developments and the uptake of eHealth solutions in the dementia
domain and reinforce the importance of early user involvement. Turning to the primary
target of the eHM-DP service, the presented findings suggest that the eHM-DP service
proved to be a valuable post-diagnostic support service, in particular for the homebased
care setting. Further research on a larger scale is needed to enhance the
implementation in existing health care infrastructures.
Background:
The uniqueness of a patient as determined by the integration of clinical data and psychological aspects should be the aspired aim of a personalized medicine approach. Nevertheless, given the time constraints usually imposed by the clinical setting, it is not easy for the physicians to collect information about the patient’s unique mental dimensions and needs related to their illness. Such information may be useful to tailor patient-physician communication improving the patient’s understanding of provided information, her involvement in the treatment process and, in general, her empowerment during and after the therapeutic journey. The primary objective of this study is to evaluate the effect of an interactive empowerment tool (IEm) on enhancing the breast cancer patient-physician experience, in terms of empowerment increasing, by providing physicians a personalized patient’s profile, accompanied by specific recommendations to suggest them how to interact with each single patient on the basis of her personal profile.
Methods:
The study will be implemented as a 2-arm randomized controlled trial with 100 adult breast cancer patients who fill in the ALGA-BC questionnaire, a computerized validated instrument to evaluate the patient’s physical and psychological characteristics following a breast cancer diagnosis. The IEm tool will collect and analyze the patient’s answers in real time and send them, together with specific recommendations, to the physician’s computer immediately before his/her first encounter with the patient. Patients will be randomized to either the intervention group using the IEm tool or to a control group who will only fill the questionnaire without taking the advantages of the tool (physicians will not receive the patient’s profile). Our primary outcome, patient empowerment, will be assessed by t-test for independent groups. Effect sizes will be calculated as mean group differences with standard deviations.
Discussion:
The proposed approach is supposed to improve the patient-physician communication leading to an increased patient’s participation in the therapeutic process with a consequent improvement in patient empowerment and personalization of care.
The aggregation of heterogeneous data from different institutions in cultural heritage and e-science has the potential to create rich data resources useful for a range of different purposes, from research to education and public interests. In this paper, we present the X3ML framework, a framework for information integration that handles effectively and efficiently the steps involved in schema mapping, Unique Resource Identifier (URI) definition and generation, data transformation, provision and aggregation. The framework is based on the X3ML mapping definition language for describing both schema mappings and URI generation policies and has a lot of advantages when compared to other relevant frameworks. We describe the architecture of the framework as well as details on the various available components. Usability aspects are discussed and performance metrics are demonstrated. The high impact of our work is verified via the increasing number of international projects that adopt and use this framework.
The advent of new web technologies and the explosion of available information online led to an information overload. During this information revolution blogs have become considerably mainstream as a media of providing news. Although there are several arguments about their validity and credibility the large amount of blogs currently available require the usage of advanced techniques for the collection, analysis, mining and efficient querying of the available information. To this direction we present BlogSearch a novel platform allowing aggregating, indexing and searching blog articles. The information is modelled using a novel RDF/S Ontology named Blogs Ontology and published as Linked Open Data. In addition, two set of APIs are provided for inserting, updating and searching information whereas the platform provides also graphical user interfaces (GUIs) for searching and inserting information. To the best of our knowledge our platform is the only one currently available publishing blog articles as Linked Open Data and simultaneously providing API’s and GUIs for aggregating, inserting and searching articles.
Personalized healthcare systems aim at providing sufficient treatment information for patients. Especially for patients that receive multi-drug treatments a key issue is the minimization of the risk for presenting drug-drug interactions (DDIs). Apart from software platforms designed for assisting physicians for optimum prescription practices, DDIs may be the result of self-medication of conventional drugs, alternative medicines, food habits, alcohol or smoking. To this respect, it is crucial for personalized health systems to provide necessary information for users for drug-drug interactions or similar information regarding risks for modulation of the ensuing treatment. In this work we describe a DDI service including drug–food, drug–herb and lifestyle factors developed in the context of a personalized patient empowerment platform. The solution enables guidance to patients for their medication on how to reduce the risk of unwanted drug interactions and side effects in a seamless and transparent way. We present and analyze the implemented services and describe examples on using an alerting service to support potential DDIs in two different chronic diseases, congestive heart failure and osteoarthritis.
Today we are witnessing an explosion in the size and the amount of the available RDF datasets. As such, conventional single node RDF management systems give their position to clustered ones. However most of the currently available clustered RDF database systems partition data using hash functions and/or vertical and hori-zontal partitions with a significant impact on the number of nodes required for query answering, increasing the total cost of query evaluation. Although RDF datasets can be interpreted as simple graphs, besides their struc-tural information they have also attached rich semantics which could be exploited to improve the partition algorithms and dictate a different approach. As such, in this paper, we focus on effectively partitioning RDF datasets across multiple nodes exploiting all available information, both structural and semantic. To this direction we present RDFCluster, a novel platform that accepts as input an RDF dataset and the number of the available computational nodes and gener-ates the corresponding partitions, exploiting both the semantics of the dataset and the structure of the corresponding graph. We view an RDF dataset as two distinct and interconnected graphs, i.e. the schema and the instance graph. Since query formulation is usually based on the schema, we generate vertical partitions based on schema clusters. To do so we select first the most important schema nodes as centroids and assign the rest of the schema nodes to their closest centroid. Then individuals are instantiated under the corresponding schema nodes producing the final partitions of the dataset. To identify the most important nodes we reuse the notion of relevance based on the well-established measures of the relative cardinality and the in/out degree centrality of a node. Then to assign the rest of the schema nodes to a centroid we define the notion of dependence assigning each schema node to the cluster with the maximum dependence between that node and the corresponding centroid. Preliminary evaluation with three datasets, namely the CRMdig, the LUBM and the Etmo, accompanied with their corresponding queries show the nice properties of the produced partitions with respect to query answering, i.e. the high quality of the constructed partitions and the low storage overhead it introduces. Our partitioning scheme can be adopted for efficient storage of RDF data reduc-ing communication costs and enabling efficient query answering. Our approach is unique in the way that constructs data partitions, based on schema clusters, com-bining structural information with rich semantics.
Personalized medicine should target not only the genetic and clinical aspects of the individual patients but also the different cognitive, psychological, family and social factors involved in various clinical choices. To this direction, in this paper, we present instruments to assess the psycho-emotional status of cancer patients and to evaluate the resilience in their family constructing in such a way an augmented patient profile. Using this profile, a decision aid can increase patient’s participation in the consultation process with their physicians and improve their satisfaction and involvement in the decision-making process. The decision aid provides standardized sets of questions related to patient’s condition and choices and the patient can choose or create his own lost to further discuss with his doctor. Our preliminary evaluation of shows promising results and the potential benefits of the tools.
Information of the healthcare domain and especially the personal health records is characterized by its heterogeneity. Clinical, lifestyle, environmental data and personal preferences are stored and managed within such platforms. Data of such a diverse information space is difficult to be delivered, especially to non-IT users like physicians or managers. Another obstacle of the health data management and analysis is the volume, which increases more and more making the need for efficient visualization methods of data and data analysis results mandatory. The objective of this work is to present architectural design recommendations of an environment for the seamless integration and intelligent processing of distributed and heterogeneous clinical information. Such a system will ease health-care professionals to orient themselves in the disperse information space and enhance their decision-making capabilities.
Background. As cancer survival increases, a new partnership model that empowers patients and caregivers and promotes self-management of the disease is needed. The H2020 iManageCancer project aims to respond to the aforementioned needs with a cancer specific personalized platform. This abstract describes the development of two tools to: 1) monitor the psycho-emotional status of the patients, and 2) evaluate family resilience in cancer, as part of the activities for the project. Methods: To identify the requirements for the tools, a review on patient empowerment, psycho-cognitive constructs relevant in cancer treatments, and family resilience in (cancer) disease was performed, together with an analysis based on patients interviews. Additionally, an online survey was promoted to the cancer community. On this basis, two use case scenarios were created to develop the tools. Results: A complex structure emerged as relevant in the assessment of the psycho-emotional status of cancer patients: psychosocial aspects, cognitive aspects, perceived health state, decisional preferences and role, perceived control of the disease, emotional profile, and patient engagement. For the assessment of family resilience, communication and problem solving, social and economic resources, connectedness and spirituality, and reappraisal emerged as important aspects. Conclusions: To the best of our knowledge these tools within the broader iManageCancer platform represent the the first approach combining psycho-emotional and resilience monitoring of the individual patient and his family, to enable patient empowerment end strengthen patient-doctor communication.
The dynamic nature of the data on the Web gives rise to a multitude of problems related to the description and analysis of the evolution of such data. These problems become more and more important to a large number of users and domains when, for example, changes are constant and interrelated. In this paper, we focus on the problem of identifying, analyzing and understanding ontology evolution. The dynamic nature of the data on the Web gives rise to a multitude of problems related to the description and analysis of the evolution of such data. Traditional approaches for identifying and analyzing changes are descriptive, focusing on the provision of a ``delta'' that describes the changes and often overwhelming the user with loads of information. Here, we take an alternative approach which aims at giving a high-level overview of the change process and at identifying the most important changes in the ontology. For doing so, we consider different metrics of ``change intensity'', taking into account the changes that affected each class and its neighborhood, as well as ontological information related to the importance and connectivity of each class in the different versions. We argue that this approach will allow a better understanding of the intent (rather than the actions) of the editor, and a better focusing of the curator analyzing the changes; traditional delta-based approaches can subsequently be used for a more fine-grained analysis.
Over the last few years we witness an explosion on the development of data management solutions for big data applications. To this direction NoSQL databases provide new opportunities by enabling elastic scaling, fault tolerance, high availability and schema flexibility. Despite these benefits, their limitations in the flexibility of query mechanisms impose a real barrier for any application that has not predetermined access use-cases. One of the main reasons for this bottleneck is that NoSQL databases do not support joins. In this poster we present a solution that efficiently supports joins over such databases. More specifically, we present a query optimization and execution module placed on top of Cassandra clusters that is able to efficiently combine information stored in different columnfamilies. Our preliminary evaluation demonstrates the feasibility of our solution and the advantages gained when compared to a recent commercial solution by DataStax. To the best of our knowledge our approach is the first and the only available open source solution allowing joins over NoSQL Cassandra databases.
Mode analysis in logic programs has been used mainly for code optimization. The mode analysis in this paper supports the program construction process. It is applied to partially complete logic programs. The program construction process is based on schema refinements and refinements by data type operations. Refinements by data type operations are at the end of the refinement process. This mode analysis supports the proper application of refinements by data type operations. In addition, it checks that the declared modes as defined by the Data Type (DT) operations are consistent with the inferred runtime modes. We have implemented an algorithm for mode analysis based on minimal function graphs. An overview of our logic program development method and the denotational semantics of the analysis framework are presented in this paper.
An overview is presented on the data management architecture of iManageCancer.
The aggregation of heterogeneous data from different institutions in cultural heritage and e-science has the potential to create rich data resources useful for a range of different purposes, from research to education and public interests. In this paper, we present the architecture and functionality of X3ML data exchange framework, that handles effectively and efficiently the schema mapping, URI definition and generation, and data transformation steps of the provision and aggregation process. The X3ML framework is based on the X3ML mapping definition language that offers the building blocks for describing both schema mappings and URI generation policies, and the X3ML engine, that handles the URI generation and the data transformation. The X3ML framework supports the cognitive process of mapping and it has a lot of advantages compared to other existing tools including that the schema mappings are expressed in a declarative way, and are both human and machine readable allowing domain experts to understand them, the schema matching and the URI generation policies comprise different distinct steps in the exchange workflow, and follow different life cycles. Furthermore X3ML is symmetric and potentially invertible allowing bidirectional interaction between providers and aggregator and thus supporting not only a rich aggregators’ repository but also corrections and improvements in the providers’ data bases.
The advances in healthcare and information technology are shifting more and more the ownership of data from medical institutions and doctors to individual citizens. However, since the medical information of an individual is confidential, the only basis for sharing it, is through prior informed consent which will regulate access to his private healthcare data. This paper highlights challenges investigated in three EU research projects and presents a solution utilizing novel access control mechanisms to ensure the selective exposure of the patients' sensitive information thereby empowering them. Our solution can efficiently support the entire life- cycle of consent such as withdrawal, activation, deletion or update. Moreover it responds to complex and different scenarios in which the patient can define complicated and dynamic access control policies at different granularity levels. In this paper we propose a Personal Health Record (PHR) system, accessible through desktop and mobile devices, that explores the efficient access regulation to information according to the consent forms provided by the patients.
Word wide web has become the first choice of patients to inform them-selves about their disease, side effects and possible treatments. While patient’s knowledge from internet is widely regarded as having a positive influence on the treatment, a lot of criticism exists for the quality and the diversity of the available information. In this paper we demonstrate the Personal Medical Information Rec-ommender (PMIR), a semantically-enabled, intelligent platform that empowers patients to search in a high quality set of web documents for relevant medical knowledge. In addition, the platform automatically provides intelligent and per-sonalized recommendations, according to the individual preferences and medical conditions. To demonstrate the platform example patients will be used to show the functionality of the system. Then we will allow conference participants to directly interact with the system to test its capabilities.
Ontology summarization aspires to produce an abridged version of the original ontology that highlights its most representative concepts. In this paper, we present RDF Digest, a novel platform that automatically produces and visu-alizes summaries of RDF/S Knowledge Bases (KBs). A summary is a valid RDFS document/graph that includes the most representative concepts of the schema, adapted to the corresponding instances. To construct this graph our al-gorithm exploits the semantics and the structure of the schema and the distribu-tion of the corresponding data/instances. A novel feature of our platform is that it allows summary exploration through extensible summaries. The aim of this demonstration is to dive in the exploration of the sources using summaries and to enhance the understanding of the various algorithms used.
Η ραγδαία ανάπτυξη του διαδικτύου και η εκτεταμένη χρήση των τεχνολογιών του Σημασιολογικού Ιστού έφεραν στο προσκήνιο την ανάγκη για γρήγορη κατανόηση, ευέλικτη εξερεύνηση και επιλογή πολύπλοκων διαδικτυακών εγγράφων και σχημάτων. Προς αυτή την κατεύθυνση, η περιληπτική παρουσίαση των οντολογιών φιλοδοξεί να παράγει μια συντετμημένη εκδοχή της αρχικής οντολογίας, η οποία τονίζει τις πιο αντιπροσωπευτικές έννοιες της. Σε αυτή τη δουλειά, παρουσιάζουμε το RDF Digest, μια πρωτότυπη πλατφόρμα, που παράγει αυτόματα περιλήψεις μιας RDF/S βάσης γνώσης. Μια περίληψη είναι ένα έγκυρο RDFS έγγραφο/γράφος που περιλαμβάνει τις πιο αντιπροσωπευτικές έννοιες του σχήματος προσαρμοσμένη στα αντίστοιχα στιγμιότυπα (στα αντίστοιχα δεδομένα). Για την κατασκευή αυτού του γράφου, ο αλγόριθμος μας εκμεταλλεύεται τη σημασιολογία και τη δομή του σχήματος, καθώς και την κατανομή των αντίστοιχων δεδομένων/στιγμιότυπων. Η αξιολόγηση που πραγματοποιήθηκε καταδεικνύει τα οφέλη της προσέγγισής μας και τα σημαντικά πλεονεκτήματα που κερδίζονται.
The advancements in healthcare practice have brought to the fore the need for flexible access to health-related information and created an ever-growing demand for the design and the development of data management infrastructures for translational and personalized medicine. In this paper, we present the data management solution implemented for the MyHealthAvatar EU research project, a project that attempts to create a digital representation of a patient’s health status. The platform is capable of aggregating several knowledge sources relevant for the provision of individualized personal services. To this end, state of the art technologies are exploited, such as ontologies to model all available information, semantic integration to enable data and query translation and a variety of linking services to allow connecting to external sources. All original information is stored in a NoSQL database for reasons of efficiency and fault tolerance. Then it is semantically uplifted through a semantic warehouse which enables efficient access to it. All different technologies are combined to create a novel web-based platform allowing seamless user interaction through APIs that support personalized, granular and secure access to the relevant information.
Patients today have ample opportunities to inform themselves in the internet about their disease and possible treatments. While this type of patient empowerment is widely regarded as having a positive influence on the treatment, there exists the problem that the quality of information that can be found on online is very diverse. This paper presents a platform which empowers patients in two ways: First it allows searching in a high quality document repository, and secondly it automatically provides intelligent and personalized recommendations, according to the individual preferences and medical conditions. To the best of our knowledge our platform is the only one which combines a search engine with automatic recommendations exploiting individual patient profiles and preferences.
The exponential growth of the web and the extended use of semantic web technologies has brought to the fore the need for quick understanding, flexible exploration and selection of complex web documents and schemas. To this direction, ontology summarization aspires to produce an abridged version of the original ontology that highlights its most representative concepts. In this paper, we present RDF Digest, a novel platform that automatically produces summaries of RDF/S Knowledge Bases (KBs). A summary is a valid RDFS document/graph that includes the most representative concepts of the schema adapted to the corresponding instances. To construct this graph, our algorithm exploits the semantics and the structure of the schema and the distribution of the corresponding data/instances. The performed preliminary evaluation demonstrates the benefits of our approach and the considerable advantages gained.
Ontology evolution aims at maintaining an ontology up to date with respect to changes in the domain that it models or novel requirements of information systems that it enables. The recent industrial adoption of SemanticWeb techniques, which rely on ontologies, has led to the increased importance of the ontology evolution research. Typical approaches to ontology evolution are designed as multiple-stage processes combining techniques from a variety ofelds (e.g., natural language processing and reasoning). However, the few existing surveys on this topic lack an indepth analysis of the various stages of the ontology evolution process. This survey extends the literature by adopting a process-centric view of ontology evolution. Accordingly, werst provide an overall process model synthesized from an overview of the existing models in the literature. Then we survey the major approaches to each of the steps in this process and conclude on future challenges for techniques aiming to solve that particular stage.
Personal health record (PHR) systems are a rapidly expanding area in the field of health information technology which motivates an ongoing research towards their evaluation in several different aspects. In this direction, we present a systematic review of the currently available PHR systems. Initially, we define a clear and concise set of requirements for efficient PHR systems which is based on real-world implementation experiences of several European research projects and also on established and widely used formal standards. Subsequently, these requirements are used to perform a systematic evaluation of existing PHR system implementations. Our evaluation study provides a thorough requirement analysis and an insight on the current status of personal health record systems. The results of the present work can therefore be used as a basis for future evaluation studies which should be conducted periodically as technology evolves and requirements are revised.
Disclaimer: These documents are made available as a means to ensure timely dissemination of scholarly and technical work on a non-commercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each copyright holder. These works may not be reposted without the explicit permission of the copyright holder.
Haridimos Kondylakis from 2013 and on is a visiting lecturer at the Computer Science Department, University of Crete and at the Department of Informatics Engineering at Technological Educational Institute of Crete. Before that he was a laboratory assistant at the Department of Informatics Engineering at Technological Educational Institute of Crete and the Vocational Training Institute of Crete. While studying at the Computer Science Department, University of Crete he was also a laboratory assistant.
Ph.D Students
Georgia Toullinou - Graph Summarization (co-supervision with Dimitris Plexousakis)
Underdraduate Students
Raphael Papadakis - Caching layer for RDF Stream Processing
Michalis Lafazanis - Caching layer for RDF Stream Processing
George fotopoulos - Recommendations for photography equipment
Ioannis Xylouris - Automatic generation of PROM/PREMS
PHD Students
Georgia Trouli - Machine learning for summaries)
Galini Polychronidou - Graph based machine learning in Prostate Cancer
Giannis Vassiliou - Diverse RDF/S Summaries
Underdraduate Students
Manos Gemeliaris – Implementing a classification algorithm over SPARK
Master Students
Pantelis Georgiadis - Machine learning models for prostate cancer prognosis and metastasis biomarkers.
Aggelos Mertsmekis - Analysis and Visualization of Medical Data.
Athanasios Makris - Systematic review on machinle learning for breast cancer.
Despoina Tsatsaroni - Alignment algorithms for biological sequences.
Hellenic Open University, Greece
Department of Electrical and Computer Engineering, HMU, Greece
Master Students
M.Sc. Nikos Kardoulakis - Schema Discovery for RDF/S KBs
M.Sc. Fanis Alevizakis - Summaries for RDF/S KBs
M.Sc. Petros Zervoudakis - Incremental Evaluation of Continuous Analytic Queries on a High-Level Query Language (co-supervision with Dimitris Plexousakis and Nicolaos Spyratos)
M.Sc. Giannis Agathaggelos - Semantic Partitioning for large RDF/S datasets (co-supervision with Dimitris Plexousakis)
M.Sc. Stratigi - Recommendation & Preferences (co-supervision with Kostas Stefanidis & Dimitris Plexousakis)
M.Sc. Alexandros Pappas - Summaries of Graph Databases (co-supervision with Dimitris Plexousakis) (2017)
Underdraduate Students
Alkyoni Hatzidaki - Analysis of SPARQL queries for Yago (2023)
Ioanna Boufidou - Analysis of SPARQL queries for DBpedia (2023)
Konstantinos Zioutos - Recommendations of healthy recipes (2023)
Dafni Palmetaki - Recommendation algorithms (2023)
Diakakis aimilios - Developing a serious games for citations (2018/2019)
Alexandros Rasidakis - Exploring DBs using mobile devices (2018/2019)
Giannis Makropodis - Developing a psychoemotional ontology for cancer (2018/2019)
Antreas Nikitakis - Spark Indexing of Data Series(2018/2019)
Giorgos Karagianidis - Indexing of Data Series (2018/2019)
Kostas Melesanakis - CQL to SQL translator (2019)
Alexandros Aggourakis - Search engine for health documents (2019)
Elena Louka - Medical Chatbot (2019)
Giannikoy Vasiliki - SPARQL query answering demo using SPARK (2019)
Stavroula Stauripierakou - myPhotoGear (2019)
Kyriazis Lazaridis – A serious game for increasing citations through semantic technologies (2018/2019)
Alexandros Charisiadis - Query Optimization over NoSQL Databases(2018/2019)
Christos Dardamanis - Exploration of social networks evolution (2018/2019)
Thanos Georgiou - DNA join workflows base on SPARK (2018)
Stavros Zafeiris - Indexing of DNA sequencing data (2018)
Savvas Kalarhakis – Parallel Data Series Indexing over Spark (2018)
Christos Athinaiou - Exploring database evolution through evolution operators(2018)
Nikos Karamesinis - Implementing a search engine for health documents (2018)
Fanourios Zervakakis – Parallel Data Series Indexing Using Multithreading (2018)
Nikos Kardoulakis - DB Navigation using Mobile Devices (2018)
Mihalis Smyrlis - Evaluation of Question Answering systems for Medical Documents (2016)
Antonis Fountouris - Implementing Joings over NoSQL Databases (2015)
Giorgos Dimitriadis - iSAX Indexing
Thomas Triantafyllou - iSAX Indexing
Apostolos Planas - Benchmarking Joins over NoSQL Databases
Master Students
Alekos Eythymios - Developing a psychoemotional ontology for cancer (2018)
Alexandros Damianakis - Ontology exploration through mobile devices (2018/2019)
Georgia Trouli - Using machine learning for optimal node selection for summaries (2019)
Giannis Tsampos– Semantic Based NLP (2018)
Underdraduate Students
Marios Vardalahakis - Implementing a platform for eHealth Smart Analytics (2018)
Mihalis Giannoulis - Collaborative Knowledge Management System (2018)
Nikos Kosmadakis
Stelios Ninidakis
Master Students
Tsekas Konstantinos - Alignment of biological sequences through timeseries indexing algorithms (2022)
Vanis Anastasios - Popypods identification through deep learning (2022)
Vradis Konstantinos - Review on methods for CVD identification (2022)
Konstantinidou Dimitra - Summarizing KEGG graphs (2022)
David Chaim - Summarizing KEGG graphs (2022)
George Bougioukas - Deep Learning models for cancer (2022)
Christos Raspoptsis - Machine Learning for Clustering Cancer Patients (2022)
Computer Science Department, University of Crete
Department of Electric and Computer Engineering at Hellenic Mediteranean University, Greece
Biomedical Engineering MSc Program, UOC, TUC & FORTH
Biomedical Engineering MSc Program, UOC, TUC & FORTH
Hellenic Open University, Greece
Department of Electric and Computer Engineering at Hellenic Mediteranean University, Greece
Hellenic Open University, Greece
Computer Science Department, University of Crete
Department of Electric and Computer Engineering at Hellenic Mediteranean University, Greece
Aalborg University
Biomedical Engineering MSc Program, UOC, TUC & FORTH
Hellenic Open University, Greece
Department of Electric and Computer Engineering at Hellenic Mediteranean University, Greece
Department of Electric and Computer Engineering at Hellenic Mediteranean University, Greece
Computer Science Department, University of Crete
Department of Electric and Computer Engineering at Hellenic Mediteranean University, Greece
Hellenic Open University, Greece
Computer Science Department, University of Crete
Department of Electric and Computer Engineering at Hellenic Mediteranean University, Greece
Department of Electric and Computer Engineering at Hellenic Mediteranean University, Greece
Department of Electric and Computer Engineering at Hellenic Mediteranean University, Greece
Hellenic Open University, Greece
Department of Informatic Engineering, Technological Educational Institute of Crete, Greece
Department of Informatic Engineering, Technological Educational Institute of Crete, Greece
Department of Informatic Engineering, Technological Educational Institute of Crete, Greece
Computer Science Department, University of Crete
Department of Informatic Engineering, Technological Educational Institute of Crete, Greece
Department of Electrical Engineering, Technological Educational Institute of Crete, Greece
Computer Science Department, University of Crete
Department of Informatics Engineering, Technological Educational Institute of Crete, Greece
Department of Electrical Engineering, Technological Educational Institute of Crete, Greece
Computer Science Department, University of Crete
Department of Informatics Engineering, Technological Educational Institute of Crete, Greece
Computer Science Department, University of Crete
Department of Informatics Engineering, Technological Educational Institute of Crete, Greece
Department of Informatics Engineering, Technological Educational Institute of Crete
Department of Informatics Engineering, Technological Educational Institute of Crete
Computer Science Department, University of Crete
Department of Informatics Engineering, Technological Educational Institute of Crete
Computer Science Department, University of Crete
Department of Informatics Engineering, Technological Educational Institute of Crete
Department of Informatics Engineering, Technological Educational Institute of Crete
Department of Informatics Engineering, Technological Educational Institute of Crete
Teaching Assistant @ Computer Science Department, University of Crete
Logic
Database Management Systems
Knowledge Representation and Processing
Files and Databases
Laboratory Assistant @ Technological Educational Institute of Crete
Web Programming
Object-Oriented Programming
Programming
Programming in C I
Programming in C II
Programming in Programming in Visual Basic
Laboratory Assistant @ Vocational Training Institute of Crete
Oracle Databases
Database Systems
I would be happy to talk to you. You can use the following contact details to find me online
You can find me daily at my office located at Computational Medicine Laboratory at Institute of Computer Science,FORTH.
Institute of Computer Science,Foundation for Research and Technology-Hellas (FORTH), Science and Technology Park of Crete, N. Plasthra 100, Vassilika Vouton, ,GR 700 13 Heraklion, Crete, Greece.