- Diabetes Management and Research
- Diabetes and associated disorders
- Artificial Intelligence in Healthcare
- Nutritional Studies and Diet
- Pancreatic function and diabetes
- Cardiac Imaging and Diagnostics
- Mobile Health and mHealth Applications
- Advanced Chemical Sensor Technologies
- Control Systems and Identification
- Machine Learning in Healthcare
- Time Series Analysis and Forecasting
- Coronary Interventions and Diagnostics
- Context-Aware Activity Recognition Systems
- Stroke Rehabilitation and Recovery
- Balance, Gait, and Falls Prevention
- Advanced Adaptive Filtering Techniques
- Bioinformatics and Genomic Networks
- Cardiovascular Function and Risk Factors
- Digital Mental Health Interventions
- Diet and metabolism studies
- Obesity, Physical Activity, Diet
- Cardiovascular Health and Disease Prevention
- Spectroscopy Techniques in Biomedical and Chemical Research
- Radiomics and Machine Learning in Medical Imaging
- Diabetes Treatment and Management
University of Ioannina
2016-2025
Foundation for Research and Technology Hellas
2023-2024
FORTH Institute of Molecular Biology and Biotechnology
2019
Applied Multilayers (United Kingdom)
2013
University of Cyprus
2013
Chinese University of Hong Kong
2013
University of Patras
2010-2012
Data-driven techniques have recently drawn significant interest in the predictive modeling of subcutaneous (s.c.) glucose concentration type 1 diabetes. In this study, s.c. prediction is treated as a multivariate regression problem, which addressed using support vector (SVR). The proposed method based on variables concerning: (i) profile, (ii) plasma insulin concentration, (iii) appearance meal-derived systemic circulation, and (iv) energy expenditure during physical activities. Six cases...
Background: The prevention of hypoglycemic events is paramount importance in the daily management insulin-treated diabetes. use short-term prediction algorithms subcutaneous (s.c.) glucose concentration may contribute significantly toward this direction. literature suggests that, although recent profile a prominent predictor hypoglycemia, overall patient's context greatly impacts its accurate estimation. objective study to evaluate performance support vector for regression (SVR) s.c. method...
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Goal</i> : The modern way of living has significantly influenced the daily diet. ever-increasing number people with obesity, diabetes and cardiovascular diseases stresses need to find tools that could help in intake necessary nutrients. xmlns:xlink="http://www.w3.org/1999/xlink">Methods:</i> In this paper, we present an automated image-based dietary assessment system Mediterranean food, based...
METABO is a diabetes monitoring and management system which aims at recording interpreting patient's context, as well as, providing decision support to both the patient doctor. The consists of (a) Patient's Mobile Device (PMD), (b) different types unobtrusive biosensors, (c) Central Subsystem (CS) located remotely hospital (d) Control Panel (CP) from physicians can follow-up their patients gain also access CS. provides multi-parametric facilitates efficient systematic dietary, physical...
In this study, an individualized predictive model of the subcutaneous glucose concentration in type 1 diabetes is presented, which relies on Random Forests regression technique. A multivariate dataset utilized concerning s.c. profile, plasma insulin concentration, intestinal absorption meal-derived and daily energy expenditure. attempt to capture rhythms metabolism, we also introduce a time feature analysis. The comes from continuous multi-day recordings 27 patients free-living conditions....
Type 1 diabetes (T1D) is a chronic endocrine disorder characterized by high blood glucose levels, impacting millions of people globally. Its management requires intensive insulin therapy, frequent monitoring, and lifestyle adjustments. The accurate prediction the short-term course levels in subcutaneous space T1D people, as measured continuous monitoring (CGM) system, essential for improving control avoiding harmful hypoglycaemic hyperglycaemic swings, facilitating precise individualized...
We propose an online machine-learning solution to the problem of nonlinear glucose time series prediction in type 1 diabetes. Recently, extreme learning machine (ELM) has been proposed for training single hidden layer feed-forward neural networks. The high accuracy and fast speed ELM drive us investigate its applicability problem. Given that diabetes self-monitoring data are received sequentially, we focus on sequential (OS-ELM) kernels (KOS-ELM). A multivariate feature set is utilized...
The emergency of cloud computing and Generic Enablers (GEs) as the building blocks Future Internet (FI) applications highlights new requirements in area services. Though, due to current restrictions various certification standards related with privacy safety health data, utilization such has been many instances unlawful. Here, we focus on demonstrating a “software data” provisioning solution propose mapping FI application use case software specifications (using GEs). aim is establish...
We propose a non-linear recursive solution to the problem of short-term prediction glucose in type 1 diabetes. The Fixed Budget Quantized Kernel Least Mean Square (QKLMS-FB) algorithm is employed construct univariate model subcutaneous concentration, which: (i) handles nonlinearities by transforming input space into high-dimensional Reproducing Hilbert Space and, (ii) finds sparse retaining representative subset training vectors. dataset comes from continuous multi-day recordings 15 patients...
We present the architecture and usability testing of a novel cloud-based platform, which integrates cyber-physical systems interoperability standards enabling clinical decision support system for risk stratification, diagnosis, prognosis treatment CAD. In this work multi-disciplinary human data were used development machine learning computational biomechanics based predictive models. Two Lab-on-Chip devices have been integrated into cloud platform. A targeted RNA-panel provides mRNA gene...
We present a new dataset of food images that can be used to evaluate recognition systems and dietary assessment systems. The Mediterranean Greek -MedGRFood consists from the cuisine, mainly cuisine. contains 42,880 belonging 132 classes which have been collected web. Based on EfficientNet family convolutional neural networks, specifically EfficientNetB2, we propose deep learning schema achieves 83.4% top-1 accuracy 97.8% top-5 in MedGRFood for recognition. This includes use fine tuning,...
The prediction of obstructive atherosclerotic disease has significant clinical meaning for the decision making. In this study, a machine learning predictive model based on gradient boosting classifier is presented, aiming to identify patients high CAD risk and those low risk. methodology includes five steps: preprocessing input data, class imbalance handling applying Easy Ensemble algorithm, recursive feature elimination technique implementation, implementation classifier, finally...
Over the years and with help of technology, daily care type 1 diabetes has been improved significantly. The increased adoption continuous glucose monitoring, subcutaneous insulin injection accurate behavioral monitoring mHealth solutions have contributed to this phenomenon. In study we present a mobile application for automated dietary assessment Mediterranean food images as part GlucoseML system. Based on short-term predictive analysis trajectory, is type-1 self-management A computer vision...
Managing daily nutrition is a prominent concern among individuals in contemporary society. The advancement of dietary assessment systems and applications utilizing images has facilitated the effective management individuals' nutritional information habits over time. determination food weight or volume vital part these for assessing quantities information. This study presents novel methodology evaluating by extracted features from training them through advanced boosting regression algorithms....
Diabetic patients must adhere continually to a complex daily regime in order maintain the blood glucose levels within safe range. Many factors impact variations such as diet, medication and exercise. This work presents modeling methodology for prediction type 1 diabetic patients. The physiological processes related diabetes (i.e. insulin absorption, gut absorption) well effects of exercise on dynamics are quantified using compartmental models. Furthermore, method employs Support Vector...
In this work, we present a novel computational approach for the prediction of atherosclerotic plaque growth. particular, patient-specific coronary computed tomography angiography (CCTA) data were collected from 60 patients at two time points. Additionally, blood samples biochemical analysis. The CCTA used 3D reconstruction arteries, which then modeling model growth is based on multi-level approach: i) flow modeled in lumen and arterial wall, ii) low high density lipoprotein monocytes...
It is generally accepted that a healthy diet plays an important role in modern lifestyle and can prevent or reduce the effects of diseases, such as obesity, diabetes cardiovascular diseases. Technological advancement wide spread smartphones enable monitoring recording nutritional habits on daily basis, through mHealth solutions. The most difficult task dietary systems for calculating composition food to estimate its volume. In this study, we present volume estimation system based structure...