arXiv subjects

Dimitrios I. Fotiadis

Publications and source records attributed to Dimitrios I. Fotiadis.

10 recordsLinked to original sources

arXiv2026

OliveGemma: A 3 Billion Visual Language Model for Recognising the Mediterranean & European Diet

Image based dietary assessment offers a scalable alternative to self reported food diaries, yet fine-grained food recognition remains challenging due to high intra-class variability and visually similar dishes. This study presents OliveGemma, a vision language model for recognising and reasoning about Mediterranean and European cuisine. Built on the open-weight PaliGemma-2-3B architecture, OliveGemma is fine-tuned with LoRA on a unified corpus of 17,340 images from three European research project datasets (MedGR, ODIN, and VIPPSTAR), reconciled into a vocabulary of 216 composed dish categories and paired with 102,642 instruction style question-answer items covering dish recognition, likely and visible ingredients, class boundary discrimination, visual evidence and overall visual food understanding. Under a 3-fold cross-validation scheme, OliveGemma achieves a top-1 accuracy of 92.96% +/- 0.91%, exceeding the strongest CNN baseline (DenseNet-121) by 7.31% and outperforming zero-shot frontier models with exact instructions and bounded classes including Gemini Flash 3 and 3.5, GPT-5.4 Mini, and Claude Haiku 4.6 by 8%, 46%, and 64% respectively. Furthermore, OliveGemma demonstrates competitive performance on Top-3 and Top-5 accuracy, being second best across CNNs and frontier models, surpassed only by DenseNet-121. In addition, OliveGemma achieves 90.79% +/- 1.3% Exact-Set on the likely ingredients of the food categories. These results demonstrate that PEFT adaptation of a small VLM can surpass substantially larger proprietary models on specialised food recognition. The model is publicly available at https://huggingface.co/JamesZar/OliveGemma-3B and the experiments and results can be found at https://github.com/tsiokris/OliveGemma.

cs.CV↗

arXiv2026

TDGT: A Tabular Data Generation Toolkit supporting adaptive GPU-accelerated Bayesian mixture models, diffusion-based models, and latent-space generative modeling

The growing demand for privacy-preserving data sharing has positioned synthetic data generation as a critical component of responsible AI workflows. Despite notable advances in generative modeling, existing solutions often lack integration of adaptive generation strategies, multi-metric evaluation, and accessible end-to-end generators within a unified web-based toolkit. In this work, we introduce TDGT (Tabular Data Generation Toolkit), a web-based toolkit for synthetic tabular data generation and fidelity assessment. TDGT introduces the Adaptive Bayesian Mixture Synthesizer (ABMS), a novel algorithm that autonomously determines the optimal number of mixture components through iterative cluster quality optimization, eliminating the need for manual hyperparameter configuration. Building upon ABMS, we further propose VAE-ABMS, a hybrid architecture that couples Variational Autoencoder-based latent space learning with adaptive Bayesian mixture synthesis, enabling high-fidelity generation of complex, nonlinear tabular distributions. For large-scale scenarios, TDGT provides a GPU-accelerated variant of ABMS leveraging CUDA-based k-means clustering and Gaussian mixture fitting. Synthetic data fidelity is assessed through eleven statistical fidelity metrics spanning distributional divergence, structural correlation, and sample-level similarity, complemented by privacy risk indicators including k-anonymity scoring and disclosure rate estimation. The web-based toolkit supports a real-time streaming interface with interactive Plotly-based visualizations. TDGT is assessed across datasets from healthcare, socioeconomic modeling, and cybersecurity domains, demonstrating consistent generation fidelity and statistical coherence across heterogeneous feature types and data scales.

cs.LG↗

arXiv2026

Frailty Estimation in Elderly Oncology Patients Using Multimodal Wearable Data and Multi-Instance Learning

Frailty and functional decline strongly influence treatment tolerance and outcomes in older patients with cancer, yet assessment is typically limited to infrequent clinic visits. We propose a multimodal wearable framework to estimate frailty-related functional change between visits in elderly breast cancer patients enrolled in the multicenter CARDIOCARE study. Free-living smartwatch physical activity and sleep features are combined with ECG-derived heart rate variability (HRV) features from a chest strap and organized into patient-horizon bags aligned to month 3 (M3) and month 6 (M6) follow-ups. Our innovation is an attention-based multiple instance learning (MIL) formulation that fuses irregular, multimodal wearable instances under real-world missingness and weak supervision. An attention-based MIL model with modality-specific multilayer perceptron (MLP) encoders with embedding dimension 128 aggregates variable-length and partially missing longitudinal instances to predict discretized change-from-baseline classes (worsened, stable, improved) for FACIT-F and handgrip strength. Under subject-independent leave-one-subject-out (LOSO) evaluation, the full multimodal model achieved balanced accuracy/F1 of 0.68 +/- 0.08/0.67 +/- 0.09 at M3 and 0.70 +/- 0.10/0.69 +/- 0.08 at M6 for handgrip, and 0.59 +/- 0.04/0.58 +/- 0.06 at M3 and 0.64 +/- 0.05/0.63 +/- 0.07 at M6 for FACIT-F. Ablation results indicated that smartwatch activity and sleep provide the strongest predictive information for frailty-related functional changes, while HRV contributes complementary information when fused with smartwatch streams.

cs.LG↗

arXiv2026

Stress Estimation in Elderly Oncology Patients Using Visual Wearable Representations and Multi-Instance Learning

Psychological stress is clinically relevant in cardio-oncology, yet it is typically assessed only through patient-reported outcome measures (PROMs) and is rarely integrated into continuous cardiotoxicity surveillance. We estimate perceived stress in an elderly, multicenter breast cancer cohort (CARDIOCARE) using multimodal wearable data from a smartwatch (physical activity and sleep) and a chest-worn ECG sensor. Wearable streams are transformed into heterogeneous visual representations, yielding a weakly supervised setting in which a single Perceived Stress Scale (PSS) score corresponds to many unlabeled windows. A lightweight pretrained mixture-of-experts backbone (Tiny-BioMoE) embeds each representation into 192-dimensional vectors, which are aggregated via attention-based multiple instance learning (MIL) to predict PSS at month 3 (M3) and month 6 (M6). Under leave-one-subject-out (LOSO) evaluation, predictions showed moderate agreement with questionnaire scores (M3: R^2=0.24, Pearson r=0.42, Spearman rho=0.48; M6: R^2=0.28, Pearson r=0.49, Spearman rho=0.52), with global RMSE/MAE of 6.62/6.07 at M3 and 6.13/5.54 at M6.

cs.LG↗

arXiv2025

Synthetic Data Blueprint (SDB): A modular framework for the statistical, structural, and graph-based evaluation of synthetic tabular data

In the rapidly evolving era of Artificial Intelligence (AI), synthetic data are widely used to accelerate innovation while preserving privacy and enabling broader data accessibility. However, the evaluation of synthetic data remains fragmented across heterogeneous metrics, ad-hoc scripts, and incomplete reporting practices. To address this gap, we introduce Synthetic Data Blueprint (SDB), a modular Pythonic based library to quantitatively and visually assess the fidelity of synthetic tabular data. SDB supports: (i) automated feature-type detection, (ii) distributional and dependency-level fidelity metrics, (iii) graph- and embedding-based structure preservation scores, and (iv) a rich suite of data visualization schemas. To demonstrate the breadth, robustness, and domain-agnostic applicability of the SDB, we evaluated the framework across three real-world use cases that differ substantially in scale, feature composition, statistical complexity, and downstream analytical requirements. These include: (i) healthcare diagnostics, (ii) socioeconomic and financial modelling, and (iii) cybersecurity and network traffic analysis. These use cases reveal how SDB can address diverse data fidelity assessment challenges, varying from mixed-type clinical variables to high-cardinality categorical attributes and high-dimensional telemetry signals, while at the same time offering a consistent, transparent, and reproducible benchmarking across heterogeneous domains.

cs.LG↗

arXiv2025

AI Model Passport: Data and System Traceability Framework for Transparent AI in Health

The increasing integration of Artificial Intelligence (AI) into health and biomedical systems necessitates robust frameworks for transparency, accountability, and ethical compliance. Existing frameworks often rely on human-readable, manual documentation which limits scalability, comparability, and machine interpretability across projects and platforms. They also fail to provide a unique, verifiable identity for AI models to ensure their provenance and authenticity across systems and use cases, limiting reproducibility and stakeholder trust. This paper introduces the concept of the AI Model Passport, a structured and standardized documentation framework that acts as a digital identity and verification tool for AI models. It captures essential metadata to uniquely identify, verify, trace and monitor AI models across their lifecycle - from data acquisition and preprocessing to model design, development and deployment. In addition, an implementation of this framework is presented through AIPassport, an MLOps tool developed within the ProCAncer-I EU project for medical imaging applications. AIPassport automates metadata collection, ensures proper versioning, decouples results from source scripts, and integrates with various development environments. Its effectiveness is showcased through a lesion segmentation use case using data from the ProCAncer-I dataset, illustrating how the AI Model Passport enhances transparency, reproducibility, and regulatory readiness while reducing manual effort. This approach aims to set a new standard for fostering trust and accountability in AI-driven healthcare solutions, aspiring to serve as the basis for developing transparent and regulation compliant AI systems across domains.

cs.AI↗

arXiv2025

A Deep Learning approach for Depressive Symptoms assessment in Parkinson's disease patients using facial videos

Parkinson's disease (PD) is a neurodegenerative disorder, manifesting with motor and non-motor symptoms. Depressive symptoms are prevalent in PD, affecting up to 45% of patients. They are often underdiagnosed due to overlapping motor features, such as hypomimia. This study explores deep learning (DL) models-ViViT, Video Swin Tiny, and 3D CNN-LSTM with attention layers-to assess the presence and severity of depressive symptoms, as detected by the Geriatric Depression Scale (GDS), in PD patients through facial video analysis. The same parameters were assessed in a secondary analysis taking into account whether patients were one hour after (ON-medication state) or 12 hours without (OFF-medication state) dopaminergic medication. Using a dataset of 1,875 videos from 178 patients, the Video Swin Tiny model achieved the highest performance, with up to 94% accuracy and 93.7% F1-score in binary classification (presence of absence of depressive symptoms), and 87.1% accuracy with an 85.4% F1-score in multiclass tasks (absence or mild or severe depressive symptoms).

eess.IV↗

arXiv2023

3-D Registration on Carotid Artery imaging data: MRI for different timesteps

A common problem which is faced by the researchers when dealing with arterial carotid imaging data is the registration of the geometrical structures between different imaging modalities or different timesteps. The use of the "Patient Position" DICOM field is not adequate to achieve accurate results due to the fact that the carotid artery is a relatively small structure and even imperceptible changes in patient position and/or direction make it difficult. While there is a wide range of simple/advanced registration techniques in the literature, there is a considerable number of studies which address the geometrical structure of the carotid artery without using any registration technique. On the other hand the existence of various registration techniques prohibits an objective comparison of the results using different registration techniques. In this paper we present a method for estimating the statistical significance that the choice of the registration technique has on the carotid geometry. One-Way Analysis of Variance(ANOVA) showed that the p-values were <0.0001 for the distances of the lumen from the centerline for both right and left carotids of the patient case that was studied.

physics.med-ph↗

arXiv2021

A new smart-cropping pipeline for prostate segmentation using deep learning networks

Prostate segmentation from magnetic resonance imaging (MRI) is a challenging task. In recent years, several network architectures have been proposed to automate this process and alleviate the burden of manual annotation. Although the performance of these models has achieved promising results, there is still room for improvement before these models can be used safely and effectively in clinical practice. One of the major challenges in prostate MR image segmentation is the presence of class imbalance in the image labels where the background pixels dominate over the prostate. In the present work we propose a DL-based pipeline for cropping the region around the prostate from MRI images to produce a more balanced distribution of the foreground pixels (prostate) and the background pixels and improve segmentation accuracy. The effect of DL-cropping for improving the segmentation performance compared to standard center-cropping is assessed using five popular DL networks for prostate segmentation, namely U-net, U-net+, Res Unet++, Bridge U-net and Dense U-net. The proposed smart-cropping outperformed the standard center cropping in terms of segmentation accuracy for all the evaluated prostate segmentation networks. In terms of Dice score, the highest improvement was achieved for the U-net+ and ResU-net++ architectures corresponding to 8.9% and 8%, respectively.

eess.IV↗

arXiv2019

3D Reconstruction of Coronary Arteries and Atherosclerotic Plaques based on Computed Tomography Angiography Images

The purpose of this study is to present a new semi-automated methodology for three-dimensional (3D) reconstruction of coronary arteries and their plaque morphology using Computed Tomography Angiography (CTA) images. The methodology is summarized in seven stages: pre-processing of the acquired CTA images, extraction of the vessel tree centerline, estimation of a weight function for lumen, outer wall and calcified plaque, lumen segmentation, outer wall segmentation, plaque detection, and finally 3D surfaces construction. The methodology was evaluated using both expert manual annotations and estimations of a recently presented Intravascular Ultrasound (IVUS) reconstruction method. As far as the manual annotation validation process is concerned, the mean value of the comparison metrics for the 3D segmentation were 0.749 and 1.746 for the Dice coefficient and Hausdorff distance, respectively. On the other hand, the correlation coefficients for the degree of stenosis 1, the degree of stenosis 2, the plaque burden, the minimal lumen area and the minimal lumen diameter, when comparing the derived from the proposed methodology 3D models with the IVUS reconstructed models, were 0.79, 0.77, 0.75, 0.85, 0.81, respectively. The proposed methodology is an innovative approach for reconstruction of coronary arteries, since it provides 3D models of the lumen, the outer wall and the CP plaques, using the minimal user interaction. Its first implementation demonstrated that it provides an accurate reconstruction of coronary arteries and thus, it may have a wide clinical applicability

eess.IV↗