Showing papers by "Nicos Maglaveras published in 2017"

Machine Learning and Data Mining Methods in Diabetes Research.

Abstract: The remarkable advances in biotechnology and health sciences have led to a significant production of data, such as high throughput genetic data and clinical information, generated from large Electronic Health Records (EHRs). To this end, application of machine learning and data mining methods in biosciences is presently, more than ever before, vital and indispensable in efforts to transform intelligently all available information into valuable knowledge. Diabetes mellitus (DM) is defined as a group of metabolic disorders exerting significant pressure on human health worldwide. Extensive research in all aspects of diabetes (diagnosis, etiopathophysiology, therapy, etc.) has led to the generation of huge amounts of data. The aim of the present study is to conduct a systematic review of the applications of machine learning, data mining techniques and tools in the field of diabetes research with respect to a) Prediction and Diagnosis, b) Diabetic Complications, c) Genetic Background and Environment, and e) Health Care and Management with the first category appearing to be the most popular. A wide range of machine learning algorithms were employed. In general, 85% of those used were characterized by supervised learning approaches and 15% by unsupervised ones, and more specifically, association rules. Support vector machines (SVM) arise as the most successful and widely used algorithm. Concerning the type of data, clinical datasets were mainly used. The title applications in the selected articles project the usefulness of extracting valuable knowledge leading to new hypotheses targeting deeper understanding and further investigation in DM.

Α Respiratory Sound Database for the Development of Automated Classification

Abstract: The automatic analysis of respiratory sounds has been a field of great research interest during the last decades. Automated classification of respiratory sounds has the potential to detect abnormalities in the early stages of a respiratory dysfunction and thus enhance the effectiveness of decision making. However, the existence of a publically available large database, in which new algorithms can be implemented, evaluated, and compared, is still lacking and is vital for further developments in the field. In the context of the International Conference on Biomedical and Health Informatics (ICBHI), the first scientific challenge was organized with the main goal of developing algorithms able to characterize respiratory sound recordings derived from clinical and non-clinical environments. The database was created by two research teams in Portugal and in Greece, and it includes 920 recordings acquired from 126 subjects. A total of 6898 respiration cycles were recorded. The cycles were annotated by respiratory experts as including crackles, wheezes, a combination of them, or no adventitious respiratory sounds. The recordings were collected using heterogeneous equipment and their duration ranged from 10 to 90 s. The chest locations from which the recordings were acquired was also provided. Noise levels in some respiration cycles were high, which simulated real life conditions and made the classification process more challenging.

Clusters of ineffective efforts during mechanical ventilation: impact on outcome

Abstract: The aim of this study was to investigate the role of ineffective efforts (IEs), specifically clusters of IEs, during mechanical ventilation on the outcome of critically ill patients. In a prospective observational study, 24-h recordings were obtained in 110 patients on the 1st day of assisted ventilation (pressure support or proportional assist), using a prototype monitor validated to identify IEs. Patients remaining on assisted ventilation were studied again on the 3rd day (n = 37) and on the 6th day (n = 13). To describe the clusters of IEs, the concept of an IEs event was developed, defined as a 3-min period of time containing more than 30 IEs. Along with all patient data, to minimize selection bias by time of recording, analysis was performed only on 1st day data of patients with ≥16 h of recording (1st day group). The analysis included 2931 h of assisted ventilation and 4,456,537 breaths. Neither the IEs index (IEs as a percentage of total breaths) in general nor a value above 10 % was correlated with patient outcome. Overall, IEs events were identified in 38 % of patients. In multivariate analysis, the presence of events in the 1st day group (n = 79) was associated with the risk of being on mechanical ventilation ≥8 days after first recording [odds ratio 6.4, 95 % confidence interval (1.1–38.30)] and hospital mortality [20 (2.3–175)]. Analysis of the data for all patients revealed similarly increased risks for prolonged ventilation [3.4 (1.1–10.7)] and mortality [4.9 (1.3–18)]. Clusters of IEs are often present in mechanically ventilated critically ill patients and are associated with prolonged mechanical ventilation and increased mortality. Studies to find ways of improving patient-ventilator interaction are warranted.

Restrictions in the T-cell repertoire of chronic lymphocytic leukemia: high-throughput immunoprofiling supports selection by shared antigenic elements

Abstract: Immunoglobulin (IG) gene repertoire restrictions strongly support antigen selection in the pathogenesis of chronic lymphocytic leukemia (CLL). Given the emerging multifarious interactions between CLL and bystander T cells, we sought to determine whether antigen(s) are also selecting T cells in CLL. We performed a large-scale, next-generation sequencing (NGS) study of the T-cell repertoire, focusing on major stereotyped subsets representing CLL subgroups with undisputed antigenic drive, but also included patients carrying non-subset IG rearrangements to seek for T-cell immunogenetic signatures ubiquitous in CLL. Considering the inherent limitations of NGS, we deployed bioinformatics algorithms for qualitative curation of T-cell receptor rearrangements, and included multiple types of controls. Overall, we document the clonal architecture of the T-cell repertoire in CLL. These T-cell clones persist and further expand overtime, and can be shared by different patients, most especially patients belonging to the same stereotyped subset. Notably, these shared clonotypes appear to be disease-specific, as they are found in neither public databases nor healthy controls. Altogether, these findings indicate that antigen drive likely underlies T-cell expansions in CLL and may be acting in a CLL subset-specific context. Whether these are the same antigens interacting with the malignant clone or tumor-derived antigens remains to be elucidated.

Self-gated fetal cardiac MRI with tiny golden angle iGRASP: A feasibility study

Abstract: PURPOSE: To develop and assess a technique for self-gated fetal cardiac cine magnetic resonance imaging (MRI) using tiny golden angle radial sampling combined with iGRASP (iterative Golden-angle RAdial Sparse Parallel) for accelerated acquisition based on parallel imaging and compressed sensing.MATERIALS AND METHODS: Fetal cardiac data were acquired from five volunteers in gestational week 29-37 at 1.5T using tiny golden angles for eddy currents reduction. The acquired multicoil radial projections were input to a principal component analysis-based compression stage. The cardiac self-gating (CSG) signal for cardiac gating was extracted from the acquired radial projections and the iGRASP reconstruction procedure was applied. In all acquisitions, a total of 4000 radial spokes were acquired within a breath-hold of less than 15 seconds using a balanced steady-state free precession pulse sequence. The images were qualitatively compared by two independent observers (on a scale of 1-4) to a single midventricular cine image from metric optimized gating (MOG) and real-time acquisitions.RESULTS: For iGRASP and MOG images, good overall image quality (2.8 ± 0.4 and 2.6 ± 1.3, respectively, for observer 1; 3.6 ± 0.5 and 3.4 ± 0.9, respectively, for observer 2) and cardiac diagnostic quality (3.8 ± 0.4 and 3.4 ± 0.9, respectively, for observer 1; 3.6 ± 0.5 and 3.6 ± 0.9, respectively, for observer 2) were obtained, with visualized myocardial thickening over the cardiac cycle and well-defined myocardial borders to ventricular lumen and liver/lung tissue. For iGRASP, MOG, and real time, left ventricular lumen diameter (14.1 ± 2.2 mm, 14.2 ± 1.9 mm, 14.7 ± 1.1 mm, respectively) and wall thickness (2.7 ± 0.3 mm, 2.6 ± 0.3 mm, 3.0 ± 0.4, respectively) showed agreement and no statistically significant difference was found (all P > 0.05). Images with iGRASP tended to have higher overall image quality scores compared with MOG and particularly real-time images, albeit not statistically significant in this feasibility study (P > 0.99 and P = 0.12, respectively).CONCLUSION: Fetal cardiac cine MRI can be performed with iGRASP using tiny golden angles and CSG. Comparison with other fetal cardiac cine MRI methods showed that the proposed method produces high-quality fetal cardiac reconstructions.LEVEL OF EVIDENCE: 2 J. Magn. Reson. Imaging 2017. (Less)

Development of an innovative mHealth platform for remote physical activity monitoring and health coaching of cardiac rehabilitation patients

Abstract: This article presents the main features and components of iCardia — an innovative mHealth platform designed to support remote monitoring and health coaching of cardiac rehabilitation (CR) patients, through Fitbit wearable sensor devices, smartphones, and personalized SMS textmessages. The design and development of iCardia were based on an iterative, user-centered design process and an open-service architecture to ensure rapid scalability and adherence to evidence-based guidelines for easier transition into clinical practice. iCardia has the potential to enable a paradigm shift towards a collaborative CR environment that utilizes mHealth technologies to engage patients to more effectively self-manage their cardiovascular disease.

Association of global and local low endothelial shear stress with high-risk plaque using intracoronary 3D optical coherence tomography: Introduction of ‘shear stress score’

Abstract: Aims The association of low endothelial shear stress (ESS) with high-risk plaque (HRP) has not been thoroughly investigated in humans. We investigated the local ESS and lumen remodelling patterns in HRPs using optical coherence tomography (OCT), developed the shear stress score , and explored its association with the prevalence of HRPs and clinical outcomes. Methods and results A total of 35 coronary arteries from 30 patients with stable angina or acute coronary syndrome (ACS) were reconstructed with three dimensional (3D) OCT. ESS was calculated using computational fluid dynamics and classified into low, moderate, and high in 3-mm-long subsegments. In each subsegment, (i) fibroatheromas (FAs) were classified into HRPs and non-HRPs based on fibrous cap (FC) thickness and lipid pool size, and (ii) lumen remodelling was classified into constrictive, compensatory, and expansive. In each artery the shear stress score was calculated as metric of the extent and severity of low ESS. FAs in low ESS subsegments had thinner FC compared with high ESS (89 ± 84 vs.138 ± 83 µm, P < 0.05). Low ESS subsegments predominantly co-localized with HRPs vs. non-HRPs (29 vs. 9%, P < 0.05) and high ESS subsegments predominantly with non-HRPs (9 vs. 24%, P < 0.05). Compensatory and expansive lumen remodelling were the predominant responses within subsegments with low ESS and HRPs. In non-stenotic FAs, low ESS was associated with HRPs vs. non-HRPs (29 vs. 3%, P < 0.05). Arteries with increased shear stress score had increased frequency of HRPs and were associated with ACS vs. stable angina. Conclusion Local low ESS and expansive lumen remodelling are associated with HRP. Arteries with increased shear stress score have increased frequency of HRPs and propensity to present with ACS.

PATHway I: Design and rationale for the investigation of the feasibility, clinical effectiveness and cost-effectiveness of a technology enabled cardiac rehabilitation platform.

Abstract: Introduction Exercise-based cardiac rehabilitation (CR) independently alters the clinical course of cardiovascular diseases resulting in a significant reduction in all-cause and cardiac mortality. However, only 15%–30% of all eligible patients participate in a phase 2 ambulatory programme. The uptake rate of community-based programmes following phase 2 CR and adherence to long-term exercise is extremely poor. Newer care models, involving telerehabilitation programmes that are delivered remotely, show considerable promise for increasing adherence. In this view, the PATHway (Physical Activity Towards Health) platform was developed and now needs to be evaluated in terms of its feasibility and clinical efficacy. Methods and analysis In a multicentre randomised controlled pilot trial, 120 participants (m/f, age 40–80 years) completing a phase 2 ambulatory CR programme will be randomised on a 1:1 basis to PATHway or usual care. PATHway involves a comprehensive, internet-enabled, sensor-based home CR platform and provides individualised heart rate monitored exercise programmes (exerclasses and exergames) as the basis on which to provide a personalised lifestyle intervention programme. The control group will receive usual care. Study outcomes will be assessed at baseline, 3 months and 6 months after completion of phase 2 of the CR programme. The primary outcome is the change in active energy expenditure. Secondary outcomes include cardiopulmonary endurance capacity, muscle strength, body composition, cardiovascular risk factors, peripheral endothelial vascular function, patient satisfaction, health-related quality of life (HRQoL), well-being, mediators of behaviour change and safety. HRQoL and healthcare costs will be taken into account in cost-effectiveness evaluation. Ethics and dissemination The study will be conducted in accordance with the Declaration of Helsinki. This protocol has been approved by the director and clinical director of the PATHway study and by the ethical committee of each participating site. Results will be disseminated via peer-reviewed scientific journals and presentations at congresses and events. Trial registration number NCT02717806. This trial is currently in the pre-results stage.

Wrist sensors — An application to acquire sensory data from Android Wear® smartwatches for connected health

Abstract: In our days where P4 (predictive, preventive, personalized, and participatory) medicine is recognized as a promising sustainable solution to the problems of healthcare systems, there is an evident demand for individual-level biological, physical, behavioral or environmental data to support the development of efficient P4 medicine interventions. The self-acquisition of such data with the help of wearable sensor technology is the primary objective of the trending Quantified Self (QS) movement. Smart devices (smartphones, smartwatches, wrist sensors, etc.), with dozens of built-sensors and great adoption by the public, emerge as the perfect technology tools for satisfying the data acquisition pursuits of QS. In this effort, smartwatches demonstrate certain sensory advantages when compared to portable smart devices (e.g., smartphones). In this work we introduce Wrist Sensors, a novel Android Wear app for presenting the available sensors of compatible smartwatches and recording measurements from any user-specified subset of them. The recordings are persistently stored and made available in the communicating Android smartphone. The app, which has been made freely available via the Google Play distribution service, can serve as a valuable data acquisition tool to facilitate the development of efficient P4 medicine interventions. In fact, the app has already been used by a published study for acquiring the input dataset for developing a novel real-time bite detection algorithm.

Identification of Barriers and Facilitators for eHealth Acceptance: The KONFIDO Study

Abstract: In this paper, we present one of the key KONFIDO project’s activities, the identification of key barriers and facilitators regarding eHealth solutions acceptance, focusing on security and interoperability. The methodology presented includes an end-user survey and an end-user workshop, engaging various stakeholders from Europe, in order to gain value out of their experience and insight in real-world healthcare settings. The analysis of the results provides a list of explicitly identified barriers and facilitators of adopting eHealth solutions in a Europe-wide scale, useful in the context of KONFIDO and beyond.

Technological Module for Unsupervised, Personalized Cardiac Rehabilitation Exercising

Abstract: Cardiac Rehabilitation (CR) can significantly improve mortality and morbidity rates from Cardiovascular Diseases (CVD). Nevertheless, traditional CR is diminished by low subsequent adherence rates. Thus, in this paper, an e-Health technological module for human motion analysis and user modelling is proposed, in order to address the requirements of unsupervised, tele-rehabilitation systems for CVD, by evaluating and personalizing prescribed physical CR programs. The proposed module consists of a) an exercise capturing and evaluation component, and b) a user modelling and decision support system for personalization of cardiac rehabilitation programs. In particular, the module monitors and analyses the body movements of the patient when exercising in real-time, while based on this analysis and the heart-rate measurements, it is capable of short-term and long-term CR session adaptation. The proposed module constitutes a significant tool for internet-enabled sensor-based home exercise platforms.

Predictive Modeling of Exercise Response in CVD Patients under Rehabilitation

Abstract: Exercise-based rehabilitation plays a key role for patients with cardiovascular disease (CVD) in improving their well-being and reducing their symptoms. Monitoring and assessing the exercise response at an individual level is critical toward achieving better health outcomes. 15 exercise sessions performed by 5 CVD patients and 9 sessions from 3 regularly active individuals were monitored, and heart rate (HR) data were acquired. A model based on the HR dynamics during exercising at different intensities was built, and simulations were performed to assess performance in different scenarios of exercise selection. Our results show that the application of simple rules in exercise selection, which consider both the HR and the beneficial HR zones of individuals, can lead to beneficial execution of exercise programs (%time spent in beneficial HR zones: 60.6±27.5 for CVD patients). Personalized guidance during exercise has the potential to significantly contribute in the beneficial execution of exercise-based cardiac rehabilitation programs.

Introducing a framework for reporting Behavioral Informatics interventions

Abstract: This work introduces a novel Behavioral Informatics Reporting Framework (BIIRF) which has been developed in the effort to review, study and compare the increasing number of Behavioral Informatics (BI) interventions for Connected Health (CH). The framework was developed through an iterative design cycle and it comprises of (i) a short-list of BI Intervention Eligibility Criteria for candidate interventions, and (ii) a BI Intervention Reporting Form. The eligibility criteria attempt to define the scope of the framework, while the reporting form - the heart of the framework - includes a hierarchical list of 39 features of interest for BI interventions, which are organized in 8 feature categories. The included features were carefully selected based on a number of predefined objectives and a successful BI intervention for the prevention of obesity and eating disorders (namely the SPLENDID intervention) has been employed as an example to facilitate the completion of the form. BIIRF has undergone preliminary evaluation and the evaluation outcomes are being exploited towards the improvement of the framework.

Evaluation of Linked, Open Data Sources for Mining Adverse Drug Reaction Signals

Abstract: Linked Data is an emerging paradigm of publishing data in the Internet, accompanied with semantic annotations in a machine understandable fashion. The Internet provides vast data, useful in identifying Public Health trends, e.g. concerning the use of drugs, or the spread of diseases. Current practice of exploiting such data includes their combination from different sources, in order to reinforce their exploitation potential, based on unstructured data management practices and the Linked Data paradigm. In this paper, we present the design, the challenges and an evaluation of a Linked Data model to be used in the context of a platform exploiting social media and bibliographic data sources (namely, Twitter and PubMed), focusing on the application of Adverse Drug Reaction (ADR) signal identification. More specifically, we present the challenges of exploiting Bio2RDF as a Linked Open Data source in this respect, focusing on collecting, updating and normalizing data with the ultimate goal of identifying ADR signals, and evaluate the presented model against three reference evaluation datasets.

Overview of Health Behavior Change Interventions to Promote Physical-activity-related Adherence in Patients with Heart Disease

Abstract: The ongoing increase in the incidence of cardiovascular disease is often associated with unhealthy lifestyle choices, while healthy lifestyle is one of the most important medical recommendations for patients with heart disease. Despite the importance of making healthy decisions daily, patient adherence to such changes is typically poor with the lowest level reported on physical activity regimen. To facilitate patient health behavior change towards a healthy lifestyle, use of health behavior interventions seems to be the most preferred tool in cardiac care routine. However, the question about use and development of successful interventions is remaining of relevance due the increasing rates of patient dropouts in rehabilitation programs and hospital readmissions. Researchers from different domains, including eHealth, are working on enabling intervention optimization. Thus, to analyze the design components that can be further used in patient adherence intervention development from the Health behavior informatics perspective, this paper provides an overview of interventions’ designs in the domain of physical-activity-related adherence in patients with heart disease. The analysis of the design approaches lead to the conclusion that the central elements for intervention design are the target patient population with its specific characteristics, and chosen type of physical-activity-related behavior. Additionally, we have found that study design and its quality should be considered when analyzing a specific intervention or intervention program effect.

Memorandum: A Mobile App for Efficient Note Keeping in Concurrent Multi-participant Human Subject Studies

Abstract: Note keeping is an indispensable ingredient of successful research. Although traditionally performed on paper, recently the task is increasingly facilitated by Electronic Lab Notebooks, i.e., ICT programs that allow their users to make electronic observations in laboratory settings. When it comes to human subject studies (HSR), i.e., the scientific investigation of human beings for medical, behavioral or social purposes, it is sometimes the case that multiple study participants perform a certain task concurrently. In such concurrent multi-participant experiments, efficient note keeping is critical as it can help assure the quality of the collected data and filter out compromised cases. The current paper presents Memorandum, a novel configurable Android application that allows the assistants of medical, behavioral or social HSR experiments to quickly and easily keep notes about the study participants. The app, which has already been employed in a behavioral study involving 40 participants, is freely available via Google Play.