ABSTRACT. Hospital crises rarely emerge as isolated events; they evolve as progressive trajectories of fragilization in which pressure accumulates, instability increases, and decision windows gradually close. Yet most operational AI remains event-centric, optimizing forecasts or alarms rather than supporting the sequential, constrained decisions that govern real crisis response. This paper proposes a decision-oriented neuro-symbolic approach to crisis anticipation, where the goal is not merely to predict failure but to preserve controllable time through governance-ready outputs: an interpretable operational state, a near-horizon escalation outlook, and an auditable Evidence-to-State-to-Trajectory-to-Action justification chain.

We instantiate the framework on 169 weeks of NHS weekly A&E sitreps published by NHS England, leveraging routinely reported indicators (demand, 4-hour performance, and severe delay markers) to extract weak-signal trajectory primitives capturing sustained strain and loss of stability. A symbolic layer then maps these primitives into a compact four-state fragilization timeline (LOW/STRAIN/UNSTABLE/CRITICAL) with explicit semantics and constrained escalation logic, ensuring stable and defensible transitions rather than noisy alerting. In time-respecting evaluation, the neuro component achieves approximately 94% discrimination (AUC ≈ 0.94) for next-week performance breach, while the neuro-symbolic fusion converts this predictive power into actionable, explainable-by-design escalation states suitable for operational governance.

The contribution is a deployable anticipation layer that reframes crisis AI around trajectories and decision windows: not “will a crisis happen,” but how resilience is eroding and what can still be changed before escalation becomes inevitable.

ABSTRACT. Antimicrobial resistance poses a fundamental threat to global health, necessitating rapid identification of resistance potential in novel gene sequences. Current machine learning approaches often rely on transductive learning, performing well on known genes but failing to generalize to unseen sequences. We present BugIQ, a Neurosymbolic Graph Neural Network that integrates pre-trained protein language models (ESM-2) with a structured knowledge graph of drug chemistry and gene ontology. By enforcing a symbolic consistency loss within the embedding space, BugIQ learns generalizable biological rules rather than memorizing graph topology, isolating genuine biological signals from dataset biases. Under simulated clinical sequencing noise (sigma=0.5), BugIQ retains a 72.66% AUC, significantly outperforming traditional gradient boosting baselines that suffer catastrophic collapse (-48.5%). Our results show that symbolic grounding acts as a critical regularizer, enabling robust deployment in noisy clinical environments while providing white-box interpretability for uncatalogued resistance mechanisms.

ABSTRACT. Abstract—Drug-induced liver injury (DILI) significantly contributes to drug failures, prompting computational predictive modeling efforts. Classical machine-learning (ML) approaches, including support vector machines (SVM), achieve strong performance but have limited interpretability and struggle with complex interactions. This study compares a support vector quantum kernel (SVQK) model with classical ML models (SVM, logistic regression, k-nearest neighbors, ensembles) on a benchmark FDA dataset of 475 drugs for DILI prediction. Quantum-enhanced SVMs, implemented via quantum kernel estimation on IBM quantum hardware, achieved comparable or slightly superior performance (AUC-ROC ∼0.85–0.90) versus outperforming the best classical baseline (AUC-ROC = 0.82 ± 0.04) under the same 16-feature constraint. Quantum kernels demonstrated robustness to noise and maintained consistent performance with fewer features. However, interpretability remains challenging, and quantum models face translational hurdles due to hardware limitations and regulatory acceptance. While quantum-enhanced models exhibit promising potential to improve DILI prediction, realizing a significant quantum advantage requires further advancements in quantum hardware and algorithm

ABSTRACT. Hand resting tremor is a hallmark manifestation of Parkinson’s disease (PD). Its clinical assessment predominantly depends on visual inspection and rating scales, which are inherently subjective and qualitative. This study introduces an automated and explainable framework for the quantitative evaluation of hand resting tremor using routine clinical videos captured in unconstrained settings. The proposed pipeline combines automatic hand detection, dense optical flow for motion analysis, and principal component analysis to generate a concise motion representation. Nine quantitative descriptors spanning spectral, temporal, and image-based domains were extracted to characterize tremor dynamics. Binary classifiers were trained to distinguish tremor from non-tremor cases through nested, subject-independent cross-validation. Nonlinear ensemble models, particularly Extra Trees, achieved balanced accuracies of up to 0.78 and AUC scores of up to 0.76, demonstrating robust subject-independent discrimination. Statistical analysis complemented by SHAP-based explainability identified dominant tremor frequency, spectral concentration within the 4–6 Hz band, tremor-band amplitude, motion periodicity, and inter-limb asymmetry as the most discriminative features, consistent with established clinical characteristics of Parkinsonian resting tremor. Spatial motion descriptors contributed comparatively less to classification performance. This framework offers a fully vision-based, low-cost, and interpretable method for objective tremor assessment, facilitating the development of scalable digital biomarkers for PD evaluation.

ABSTRACT. Biochemical recurrence (BCR) after curative-intent therapy for prostate cancer is assessed from heterogeneous evidence spanning histopathology, multiparametric MRI (mpMRI), and structured clinical variables. Learning robust multimodal predictors remains challenging due to small cohort sizes, modality-specific noise, and missingness. In this work, we study intermediate fusion for BCR risk prediction on the MICCAI CHIMERA benchmark under a controlled setup. We train attention-based multiple instance learning (ABMIL) aggregators to obtain patient-level embeddings for whole-slide histology patches and for sequence-wise MRI-CORE slice embeddings (ADC/HBV/T2w), and then compare two fusion operators with matched modality projections, modality-type embeddings, and an identical prediction head: (i) a modality-token Transformer encoder with CLS readout (MEFT) and (ii) seeded cross-attention pooling with learned fusion tokens (MEST). All models are trained with the Cox objective and evaluated using a fixed 5-fold cross-validation with out-of-fold embedding generation to avoid leakage. Both fusion approaches achieve comparable performance (C-index ≈ 0.85), with MEST showing slightly lower fold-to-fold variance. Ablations indicate that histology and clinical variables dominate performance, while mpMRI provides complementary gains when fused despite weaker monomodal performance. These results suggest that, on CHIMERA, the fusion operator has a modest effect under matched embeddings, and that preserving sequence-wise mpMRI representations can improve multimodal risk stratification.

ABSTRACT. Magnetic Resonance Imaging (MRI) is highly susceptible to acquisition artifacts, which degrade image quality and compromise medical diagnosis. In clinical settings, the absence of robust recovery methods forces technologists to repeat corrupted scans, unnecessarily increasing operational costs and patient discomfort. Although Deep Learning methods have shown potential for correcting MRI artifacts, their development is hindered by the limited availability of artifact-corrupted images. When available, data remain dispersed across hospitals and cannot be centralized due to privacy and regulatory constraints. Compounding this, scanner heterogeneity requires models to be trained on multi-institutional data to achieve efficient correction.

To address these challenges, we propose OHANA, an end-to-end solution for synthetic artifact generation and correction in multi-contrast brain MRI. We first introduce a synthetic data generator that simulates multiple artifact types, creating artifact-corrupted training datasets. Using this augmented data, we extend MC2-Net to perform artifact correction across four MRI contrasts. OHANA also integrates Federated Learning, allowing multiple institutions to collaboratively train models without sharing data.

Experiments demonstrate that OHANA outperforms state-of-the-art artifact-correction approaches, achieving a 17.2% improvement in the Structural Similarity Index Measure. A radiologist assessed the realism of the generated artifacts and the diagnostic fidelity of the corrected images. These results highlight the potential of OHANA to improve medical diagnosis.

ABSTRACT. Electroencephalography (EEG) and functional Magnetic Resonance Imaging (fMRI) are two major functional brain imaging modalities. EEG and fMRI can be recorded simultaneously to measure brain activity and take advantage of both modalities, providing good temporal resolution from EEG signals and high spatial resolution from fMRI. Indeed, the spatial resolution of EEG signals is poor due to the ill-posed inverse problem of source localisation. Previous work has enabled EEG electrode detection using Ultra-Short TE MR images. Building upon a previously introduced method, we adapt and validate it for EEG electrode localization on T1-weighted MRI, thereby extending its applicability to a new imaging modality. By relying solely on T1-weighted images, which is commonly acquired in fMRI protocols, this approach is both simple and easily applicable to existing EEG-fMRI datasets. Although detections are slightly less accurate than those obtained on ultra-short TE sequences, the results remain excellent, with an average detection accuracy of 99.27%, an average positioning error of 2.59 mm, and perfect accuracy in electrode labeling.

ABSTRACT. Current neuroimaging studies face methodological limitations in neural-phenotypic integration for biomarker validation. We present a multiscale computational framework integrating graph-theoretic connectivity analysis with dual-threshold statistical validation and unsupervised phenotypic stratification. Our approach combines Welch's t-test, FDR correction, and effect size thresholds to prioritize biological relevance over statistical artifacts. Applied to task-based fMRI from 86 participants across the schizophrenia spectrum, we extracted nodal and network-level metrics from sparse weighted functional graphs. Familial validation through a four-group design dissociated state-dependent from trait-related markers, establishing cerebellar-cortical hypoconnectivity as illness-specific rather than genetic risk. Consensus clustering of cognitive and symptomatic features identified three stable phenotypic subtypes with distinct neuropsychological profiles, highlighting the potential utility of phenotype stratification. This framework demonstrates that integrating multiscale graph analysis with dual-threshold validation and unsupervised stratification establishes a reproducible computational standard for biomarker discovery, significantly enhancing biological interpretability beyond traditional neurological analysis.

ABSTRACT. Alzheimer’s disease (AD) and frontotemporal dementia (FTD) are two of the most prevalent neurodegenerative disorders, yet their overlapping clinical presentations pose substantial diagnostic challenges. Current diagnostic methods rely on costly neuroimaging and subjective clinical assessments, creating an urgent need for accessible, objective screening tools. This paper presents a novel approach using a Bidirectional Temporal Graph Convolutional Network (GCN)-Transformer framework that incorporates comprehensive fractal pattern analysis to automatically classify AD, FTD, and cognitively normal (CN) individuals from EEG signals. 14 different features from each EEG channel were extracted to build a complete picture: 3 time domain features (mean, variance, standard deviation), 4 fractal dimension measures (Higuchi, Petrosian, Katz Fractal Dimensions, and Detrended Fluctuation Analysis), and 7 frequency domain features (spectral entropy, band powers across delta,theta, alpha, beta, gamma bands, and peak frequency). The aim of these multi-scale features is to effectively capture the self-similar patterns and nonlinear dynamics characteristic of neurodegenerative brain activity across all 19 EEG channels. The architecture combines graph convolutional layers with graph constrained multi-head attention mechanisms operating on dynamic temporal adjacency matrices, with bidirectional processing strengthened by positional encoding. Ten-fold cross-validation of the dataset yielded training accuracy of 98.65±0.38% and test accuracy of 97.00±0.45%, while leave-one-subject-out validation achieved 85.00% accuracy along with 86.23% precision, 85.00% recall, 85.34% F1-score, 85.00% sensitivity, and 92.50% specificity. Our ablation studies demonstrated clear advantages over standard Graphical Convolutional Network (94.12±0.78%) and Graphical Attention Transformer variants (95.45±0.68%). The proposed framework achieves competitive performance while utilizing only 800k parameters, enabling efficient deployment in resource-constrained clinical settings for early-stage dementia screening.

ABSTRACT. We present a novel multi-point calibration and correction framework for compensating nonlinear channel bleed-through (crosstalk) in epifluorescence microscopy, enabling improved quantification in liquid biopsy assays. Singly labeled control slides spanning a broad fluorescence dynamic range are used to sample intensity-dependent crosstalk at multiple levels. For an imaging system with M detection channels, each fluorophore’s bleed-through behavior is quantified using N calibration 1×M vectors corresponding to discrete intensity states. The resulting N^M precomputed correction matrices collectively model nonlinear crosstalk interactions.

In contrast to conventional Newton-based or general iterative solvers, which require explicit modeling of continuous intensity-dependent crosstalk and may incur substantial computational cost or convergence uncertainty, the proposed method employs a fixed set of precomputed matrices combined with an efficient positional indexing of numerous crosstalk states obtained via multi-level image segmentation in the correction process. In a 2K × 2K four-channel fluorescence system, a tri-point implementation reduced processing time to 1.8 seconds, compared with 374 seconds for a two-iteration Newton solver. The method achieved correction improvement factors of 1.42 fold relative to a general iterative solver and 1.91 fold relative to a Newton solver. These results demonstrate that the multi-point framework provides both substantial computational acceleration and improved correction accuracy for quantitative fluorescence imaging.

ABSTRACT. Longitudinal magnetic resonance imaging (MRI) is essential for monitoring the progression of Multiple Sclerosis (MS) disease. This work presents a lightweight end to end integrated software platform for longitudinal brain MRI analysis and progression assessment in MS. The proposed platform integrates visualization, annotation management, pre processing (Original (O), Histogram Normalized (HN), Gaussian Smoothed (GS), Gaussian smoothed following histogram normalization (GSHN)), rigid registration and disease evolution quantification within a unified and persistent workflow. Annotation propagation across different time points (T1-T4) of the disease progression is supported through rigid transformation reuse. This enables spatially aligned initialization of segmented lesions and structured comparison of stable, projected and newly appearing MS lesions. Experimental evaluation on five longitudinal clinical cases, at T1-T4, demonstrated consistent improvement in intra subject alignment across all preprocessing variants using correlation based rigid registration, with median MSE reductions of up to 0.57 and correlation coefficient, ρ, increasing up to 0.93. The proposed system enables reproducible longitudinal studies, lessens the burden of manual (M) delineations and annotations and lays the groundwork for future research into lesions evolution and the characterization of pre-lesional tissue as well as automated lesions segmentation.

ABSTRACT. Accurate prediction of ambulance response times is critical for emergency medical services (EMS) dispatch. Existing Machine Learning (ML) models mostly depend on the quality of the training data provided, and often fail to communicate the complex uncertainty inherent in EMS data. This paper presents a process-aware conformal prediction framework that provides distribution-free prediction intervals with finite-sample coverage guarantees for EMS response times. The response process is decomposed into four operational phases: Call processing, dispatch, crew mobilization, and travel. Conformal Quantile Regression (CQR) is applied independently to each stage to compose intervals that maintain coverage guarantee based on the provided data. Four conformal variants are evaluated on a synthetic dataset of 75,000 incidents calibrated and adjusted to reflect the Cyprus operational model. Findings suggest that the CQR produces the tightest intervals out of the four methods and would be the preferred conformal variant for real-time dispatch application when the objective is to minimize the interval width while maintaining coverage. Per-stage decomposition produces wider intervals but provides diagnostic information useful for quality assurance and process improvement.

ABSTRACT. The assignment of International Classification of Diseases (ICD) codes to clinical encounters is a labor-intensive process that is both error-prone and costly. While automated coding models have made substantial progress on benchmark datasets, they typically operate as batch-oriented black boxes that produce ranked code lists without supporting evidence. Clinical coders, however, require interactive tools that explain predictions in terms they can verify against the source document. We present ClinCode Copilot, a web-based clinical coding assistant that integrates a hybrid machine learning model with an interactive workspace designed for real-time ICD code review. The system combines an end-to-end label attention classifier with chunk-level K-nearest neighbor retrieval to provide dual interpretability: per-code attention highlighting identifies which sections of a discharge summary support each prediction, while similar patient retrieval surfaces training cases that received the same code. Built with a FastAPI backend serving a Bio_ClinicalBERT encoder and FAISS index, and a Next.js frontend with resizable multi-panel layout, ClinCode Copilot enables clinical coders to review predictions, inspect evidence, and verify codes within a single interface. The underlying model achieves Micro-F1 of 0.482 on 1275 ICD-9 codes from MIMIC-III. The tool is open-source, and the code is publicly available at https://github.com/ieeta-mith/clincode-copilot.

ABSTRACT. Chemotherapy is essential in cancer treatment but often causes serious toxicities such as cardiotoxicity and neutropenia. In current practice, toxicity monitoring relies mainly on population guidelines and periodic visits, so adverse events are often detected late. ChemoTwin is a clinical decision support prototype that uses a digital twin approach to help clinicians monitor chemotherapy toxicity. The system combines continuous monitoring of patient data, toxicity risk prediction, and access to clinical guidelines so that recommendations stay evidence-based. The prototype brings together more than 15 clinical parameters and estimates risks for key adverse events. Preliminary evaluation shows promising prediction performance and positive feedback from oncology practitioners, supporting the feasibility of this approach for future clinical validation.

ABSTRACT. Chronic kidney disease requires long-term clinical management, and many patients undergo hemodialysis several times per week. In some cases, particularly among older patients, an arteriovenous graft is implanted to enable vascular access. Although effective, graft implantation carries risks, including infections, occlusions, and graft failure, which can be life-threatening. TeleGraft is a European project that aims to develop an intelligent arteriovenous graft for hemodialysis patients and to enable regular monitoring of blood flow anomalies and infections, through embedded sensors and Raman spectroscopy. This paper presents a telemonitoring system designed to monitor biosignals acquired from arteriovenous grafts and to detect potential complications. The system is also designed to be modular, adaptable by enabling monitoring of other types of biosignals, and support patients with distinct medical conditions. As a result, we developed a web application to allow healthcare professionals to monitor the status of the graft and patient data, but also a mobile application that supports the acquisition and submission of various health biosignals to the telemonitoring platform.

ABSTRACT. The digital transformation of healthcare in Europe is currently undergoing a paradigm shift, moving from provider-centric systems toward a patient-oriented ecosystem mandated by the European Health Data Space regulation. While the digitalization of hospital infrastructures has been largely successful, citizens often remain passive subjects with limited control over their clinical data. This paper presents the design and implementation of a National Mobile Health Application developed as a Minimum Viable Product to bridge the gap between legislative requirements and practical citizen empowerment. The proposed solution integrates an Electronic Health Record viewer, secure medical sharing via the Verifiable Health Links protocol, and teleconsultation services within a unified, cross-platform interface. A key contribution of this work is the integration of the European Digital Identity Wallet for high-assurance authentication and the OpenNCP framework to facilitate cross-border semantic interoperability through the MyHealth@EU infrastructure. By utilizing open standards such as HL7 FHIR and CDA, the MVP demonstrates a scalable framework that enables citizens to exercise their rights to data access, portability, and insertion. Our results highlight how the synergy of these components fosters health literacy and therapeutic equity, providing a technical blueprint for National Health Authorities to comply with the EHDS and ensure that clinical data remains actionable across borders.

ABSTRACT. Falls are a leading cause of injury and loss of independence among older adults. While physical factors such as balance and muscle strength are known contributors to fall risk, increasing evidence suggests that cognitive function, particularly executive function, attention, and processing speed, also plays a significant role. Understanding the relationship between cognitive performance and fall risk may help improve fall prevention strategies. This study aimed to examine whether cognitive function predicts fall risk in older adults participating in a combined cognitive and physical training program. A sample of community-dwelling older adults aged 60 years and above participated in a structured cognitive and physical training program titled LLM Care. Cognitive function was assessed using standardized neuropsychological tests evaluating global cognition, executive function, and attention. Fall risk was measured using established functional mobility and balance assessments. Participants completed a multi-week intervention consisting of cognitive training exercises and physical activities focusing on balance, strength, and coordination. Statistical analyses were conducted to examine associations between cognitive performance and fall risk outcomes.

ABSTRACT. Antimicrobial peptides (AMPs) have emerged as strong candidates to replace traditional antibiotics in response to the escalating problem of antimicrobial resistance. With the rapid progress of computational biology and the structural complexity of AMPs, reliable machine learning–based classification approaches have become increasingly essential. In this work, we introduce a novel statistical framework designed for systematic feature grouping, scoring, and selection to improve AMP prediction performance. The proposed approach applies k-means clustering to partition peptide features into coherent groups and prioritizes them using q-values obtained from z-score–based statistical analysis. The framework was validated on two large-scale AMP benchmark datasets comprising 3,556 and 12,022 peptide sequences. Feature subsets derived from the ranked groups were used to train seven machine learning classifiers along with their ensemble variants. The results indicate that the proposed method surpasses existing grouping-based strategies on Dataset 1, achieving an accuracy of 0.9238 ± 0.0092, and delivers competitive performance on Dataset 2 with an accuracy of 0.8667 ± 0.0083.

ABSTRACT. Alzheimer’s disease is a multifactorial neurodegenerative disorder for which single-target pharmacological strategies have shown limited clinical success. The complexity of its underlying molecular mechanisms suggests that effective treatment may require the simultaneous modulation of multiple pathological pathways. In this context, drug combination therapies represent a promising alternative, although the experimental exploration of all possible combinations is unfeasible due to the vast combinatorial space. Network medicine provides a framework to address this challenge by modeling diseases and drugs within the human protein–protein interaction network and quantifying their topological relationships. In this study, we present a network-based pipeline to systematically evaluate potential drug combinations for Alzheimer by integrating drug–disease proximity and drug–drug separation metrics. We apply this framework to a large set of approved and investigational drugs in order to identify combinations that are both proximal to the disease module and topologically complementary. We specifically prioritize the combination of memantine and glimepiride, which exhibits an ideal complementary exposure pattern (sAB = 1.1). Our results reveal that while direct interaction with the disease module is limited (1.4%), this pair influences 356 unique proteins (13.2% of the total Alzheimer’s disease module) through 2-hop propagation paths. Overall, this framework identifies drug pairs that target complementary disease pathways, providing a clear biological rationale for developing new combination therapies for Alzheimer’s.

ABSTRACT. Μuch progress has been noted in the prediction of drug-drug interactions with machine learning. Usually, this is performed in the context of closed world assumption, whereby all training and test data stem from the same distribution. In reality, the test data can be a mixture of in and and out of distribution (OoD) instances. However, the predictor would misclassify the out of OoD instances, leading to erroneous predictions and thus misleading clinicians.

In the current work we apply supervised contrastive learning (SCL) to detect OoD instances, and distinguish them from in- distribution instances. In particular, the OoD instances are un- known interactions, i.e. interactions that have not been observed in the training set. The method applies SCL on embeddings of drug pairs, but also on the penultimate layer, and the on the output of an neural network that was trained to predict in-distribution instances. We compare the there aforementioned approaches to a baseline k-NN based method that does not use SCL. The role of SCL is to enforce a strong separation of the in-distribution and OoD instances. We also, study the role of negative samples representing lack of any interaction among drug pairs. Moreover, there are experiments with different mixtures of in-distribution (observed) and OoD (unobserved) interactions. Results are presented on a public data extracted from Twosides data set, and by and large SCL performs better than the baseline method.

ABSTRACT. Predicting drug-disease interactions is essential for pharmacological discovery, yet most computational models underperform in cold-start scenarios where entities have limited interaction history. Standard Graph Neural Networks (GNNs) typically require dense connectivity, leaving sparse nodes as topological informationally isolated within biological networks. This study addresses such sparsity by constructing a unified heterogeneous knowledge graph using drug-gene (ChG-Miner) and drug-disease (DCh-Miner) associations from the BioSNAP database. We demonstrate that gene-target interactions serve as a latent bridge, allowing the model to infer context for drugs with minimal disease history through their genomic profiles. We evaluate two GNN architectures — GraphSAGE and Graph Attention Networks (GAT) — under severe sparsity, where over 50% of nodes exhibit a degree d ≤ 1.Results reveal a significant architectural trade-off: while GAT is effective in data-rich regions, its performance drops to a ROC-AUC of 0.79 for degree-1 nodes. In contrast, GraphSAGE maintains a ROC-AUC of 0.97 in the same sparse regions. This suggests that attention mechanisms become a liability when local neighborhoods are insufficient for statistical weighting. Our findings indicate that while attention-based models excel in well-annotated contexts, mean-pooling aggregators like GraphSAGE provide superior robustness for rare diseases and novel compound association prediction.

ABSTRACT. This study evaluates biomedical language models for sentence-level adverse drug event detection in regulated pharmacovigilance settings. Using the deduplicated ADE-Corpus-V2 dataset, models from 110M to 7B parameters were compared under a fixed single-GPU setup with five random seeds. The evaluation considered regulatory needs such as reproducibility, traceability, validation, and change control. The results show that mid-sized biomedical pretrained language models, especially 300–400M parameter models, achieved strong and stable performance. Larger models did not clearly outperform them. These findings suggest that reliable ADE classification can be achieved with practical, auditable models without relying on very large proprietary LLMs.

ABSTRACT. The high failure rate of Alzheimer's disease trials is partly attributable to the lack of valid surrogate outcomes that reliably predict patient-relevant benefit. This study introduces multibiomarker putative surrogate outcomes, integrating amyloid, tau and neurodegeneration features to address this gap. Using a comprehensive simulation framework, we evaluated unweighted and variance-weighted multibiomarker models against the standard reduction in amyloid load benchmark under realistic trial conditions, including informative censoring and heterogeneous progression rates. Results demonstrate that while the single-marker benchmark failed to predict cognitive decline, the variance-weighted multibiomarker outcome significantly optimised the signal-to-noise ratio, increasing statistical power from 44 percent to 77 percent in small-sample trials. Furthermore, the model established a valid surrogate threshold effect for the Alzheimer's Disease Assessment Scale–Cognitive Subscale 13 but no such validation was established for the Mini-Mental State Examination. These findings confirm that variance-based feature engineering is essential for constructing robust multivariable surrogate endpoints, ultimately enabling faster and more reliable evaluation of new Alzheimer’s disease treatments.

ABSTRACT. Type 2 diabetes mellitus (T2DM) is a chronic metabolic disease associated with hyperglycemia and vascular complications, including diabetic retinopathy (DR), which re- flects microvascular alterations that may also be associated with neurodegenerative processes. Cognitive decline is commonly evaluated using the Mini-Mental State Examination (MMSE), and retinal imaging has been investigated as a potential non- invasive source of biomarkers for brain-related disorders. This study proposes a transfer learning pipeline for cognitive status classification using fundus photographs from 54 patients with DR. Feature representations were extracted from pre-trained models and subsequently used in a classification framework to distinguish between cognitive normal (CN) and mild cognitive impairment (MCI) groups based on an MMSE-derived threshold. Patient-level nested cross-validation was employed for perfor- mance evaluation. The proposed framework achieved a balanced accuracy of 0.6191 and an area under the ROC curve (AUC) of 0.6191. Matthews Correlation Coefficient (MCC) reached 0.2530, indicating a weak-to-moderate association between predicted and true labels under the studied conditions. Although classifica- tion performance was moderate, the results provide prelimi- nary evidence that retinal embeddings may contain information potentially related to cognitive status. Further validation on larger and externally independent cohorts is required to confirm generalization capability.

ABSTRACT. Effective chronic disease management requires continuous patient monitoring, structured risk evaluation, and the secure embedding of predictive services within routine clinical workflows. This study introduces a dual-interface digital health platform designed to facilitate coordinated interaction between patients and clinicians in the context of chronic disease monitoring. The system comprises a patient-oriented mobile application, called CURA, for structured collection of symptoms and clinical parameters, a clinician-facing web dashboard enabling longitudinal visualization and supervised evaluation, and a centralized orchestration layer responsible for secure data governance, traceable processing, and modular scalability. Data acquisition follows predefined schemas to ensure semantic consistency and interoperability, thereby reducing downstream ambiguity and supporting reliable longitudinal assessment. Embedded within this architecture, an AI-driven risk stratification engine functions as a transversal processing module, producing probabilistic risk indicators that are directly integrated into the clinical workflow. Crucially, these outputs are presented as decision-support signals rather than autonomous determinations and require explicit physician validation. This human-in-the-loop design preserves clinical authority, mitigates automation bias, and guarantees meaningful oversight in alignment with high-risk medical software standards.

ABSTRACT. Telemedicine systems deployed in long-term care environments must operate under constrained clinical review capacity. Traditional evaluation based on global discrimination metrics does not adequately reflect real-world triage settings, where only a fraction of daily cases can be reviewed. We propose a deployment-oriented evaluation framework for telemedicine-based risk prioritization under fixed review budgets. Episodes are ranked by predicted mandatory referral risk, and performance is assessed using Recall@k, ranking quality, subgroup fairness, and robustness analyses. Across 10-fold cross-validation on real telemedicine data, calibrated gradient boosting models achieved the highest recall of mandatory cases under constrained review capacity, while maintaining lower subgroup disparity across age strata. Graph-based modelling demonstrated stable performance but did not surpass strong tabular baselines under the current feature configuration. Our findings reframe telemedicine AI evaluation around operational prioritization, equity and robustness, providing a practical pathway toward safer AI-assisted triage systems.

ABSTRACT. Falls during nocturnal bed-exit transitions represent a major safety concern in institutionalized older adults. While numerous fall detection systems have been validated under controlled laboratory conditions, real-world residential deployments remain limited, particularly for early-risk prediction under extreme class imbalance. This work presents the deployment and validation of a pressure-sensing sheet system (1056 sensors, 48×22 matrix) installed in a residential care facility to identify pre-fall movement patterns up to 10 minutes before fall events. From 21 recorded falls, 10 contained sufficient historical data for predictive modelling. To mitigate initial methodological bias caused by positive-only training, a balanced dataset was constructed using 10 pre-fall windows and 20 normal-activity windows. Several machine learning models were evaluated, with Decision Tree selected for its interpretability and performance. Despite extreme data scarcity, results demonstrate discrimination above random baseline and validate the technical feasibility of early bed-exit risk detection in real-world residential environments. The study highlights methodological challenges inherent to rare-event prediction in assistive healthcare systems.

ABSTRACT. As ALS progresses, patients lose motor skills and become reliant on eye movements for communication. Effective tools for severe ALS must be minimal-effort, error-tolerant, and accessible. We present an eye-tracking system that maps left/right gaze movements to binary codewords encoding 24 essential daily keywords via Huffman or max-heap trees. Keywords are partitioned into two groups of 12, yielding short codewords (3--4 bits for Huffman, 1--3 bits for max-heap) that reduce gaze actions by 52--70 % compared to conventional QWERTY-based dwell selection. Unlike existing systems requiring expensive IR-based trackers, our approach classifies gaze into only two directions, enabling standard webcam use. Error-tolerant communication is supported through adaptive thresholds, FIR filtering, and Huffman-based error detection (91.7 % and 95.8 % for single-bit deletion and insertion errors). When an error is detected, a top-k candidate correction mechanism achieves up to 97.2 % correction accuracy with one additional gaze action. Testing with 11 participants aged 10 to 80 achieved a character error rate of 1.3 % and an average detection time of 1.77 seconds per character.

ABSTRACT. Improving the representation of underserved populations in clinical studies (CS) is key to advancing more equitable and generalizable medical research. A primary barrier to accessibility is the complexity of eligibility criteria, which are typically stored as unstructured free text. We present an artificial intelligence (AI) based pipeline that automates transformation of these narratives into structured, standardized, and both machine- and human-readable formats. The system employs a multi-stage architecture that uses large language models (LLMs) to segment eligibility criteria and perform biomedical concept extraction with metadata (operators, units, and temporal constraints). To ensure high-fidelity semantic normalization without the need for extensive annotated datasets, we implement a Retrieval Augmented Generation (RAG) framework. This framework pairs dense vector embeddings for high-recall candidate retrieval with LLM-based neural re-ranking for dual-ontology linking to SNOMED CT and the CDISC Unified Study Definition Model (USDM).

ABSTRACT. Wearable data streams play a critical role in remote health monitoring applications. However, they are prone to missing values, which may result in biased findings and inappropriate treatment decisions. This challenge underscores the importance of real-time imputation for streaming data. Our study evaluates the effectiveness of online imputation methods for enhancing anomaly detection in patients' wearable data streams. Experiments were conducted under various missing data mechanisms and rates, utilizing four online imputation methods: Linear Regression, Hoeffding Tree regressor, Hoeffding Adaptive Tree regressor, and Multi-layer Perceptron for regression, in addition to a baseline mean imputation. Results demonstrate that online imputation methods, particularly the Hoeffding Adaptive Tree, achieved superior imputation performance and improved anomaly detection in most scenarios. Nevertheless, each missing mechanism produced distinct result patterns that vary in both performance and stability, highlighting the necessity for further research to advance online imputation approaches for critical healthcare scenarios.

ABSTRACT. This paper investigates the cognitive and emotional processes underlying medical palpation to inform the design of a remote palpation system. Traditional telepalpation systems often focus solely on force feedback, yet effective palpation integrates tactile sensations with clinical assessment, prior experience, and patient responses, both physical and emotional.

We analyze two complementary pathways of tactile perception: active touch by doctors, involving kinesthetic and tactile feedback, and passive touch by patients, eliciting subjective and emotional responses conveyed verbally and nonverbally. Building on this framework, we propose a remote palpation system that leverages multimodal interaction to replicate these processes.

Empirical evaluation demonstrated reliable spatial touch perception and reasonably consistent force perception across participants, validating the feasibility of the system. These findings lay a foundation for further development and testing of remote and mixed-reality palpation systems.

ABSTRACT. Robotic surgery has transitioned from a novel surgical tool to an increasingly integral component of modern surgical practice. Advances in robotics, artificial intelligence (AI), augmented reality (AR), and image-guided navigation have collectively shifted the surgical paradigm towards more precise, personalized, and data-driven procedures. AI-assisted robotic platforms have demonstrated measurable improvements in surgical accuracy, reduced intraoperative complications, and shorter operative times compared with conventional methods. Augmented reality and fusion imaging have been successfully used for enhanced visual guidance in complex procedures. Despite these advancements, technical and ethical challenges persist, including system costs, training requirements, integration complexity, data privacy, and regulatory oversight. This paper reviews recent progress in robotic surgery, evaluates emerging trends, and discusses practical challenges.

ABSTRACT. Drawing-based assessments are widely used to evaluate cognitive and motor function, yet most computational approaches emphasize geometric and kinematic features while neglecting the sequential organization of drawing behavior. As drawing involves structured spatial decomposition and ordered stroke execution, sequence differences may reflect learned strategies rather than neurological impairment, particularly across heterogeneous educational backgrounds. Additionally, datasets including individuals with no formal education are typically small, and enriched symbolic representations increase token cardinality, limiting conventional sequence analysis methods. This paper proposes a strategy-aware framework integrating YOLOv8-based stroke classification with two novel descriptors: the Hierarchical Sequence Signature (HSS) and the Hybrid Sequential Entropy Descriptor (HSED). These descriptors capture multi-level structural and entropy-based properties of drawing sequences while producing fixed-dimensional per-subject feature vectors without population-level parameter estimation. Evaluated on a balanced dataset of 78 participants, the framework achieves strong discriminative performance (HSED AUC = 0.933) and demonstrates robust generalization.

ABSTRACT. This paper presents the development and pilot evaluation of the HelpInMind mobile application. HelpInMind is designed to complement the treatment of generalized anxiety disorder by enhancing patients’ adherence to therapy. The application delivers brief motivational interviewing interventions through an embodied conversational agent, grounded in the transtheoretical model of change. The intervention protocol comprises 23 sessions for anxiety treatment, designed to map motivational interviewing techniques onto the five stages of behavioral change. An initial version of HelpInMind was evaluated with psychology students (N = 11), focusing on both interaction with the virtual agent and overall app usability. The results indicate high levels of usability, quality of interaction, and perceived usefulness. In addition, participant feedback identified several functional improvements, which will inform the development of a subsequent version to be evaluated with patients undergoing anxiety treatment.

ABSTRACT. Open-source PACS platforms such as Dicoogle have demonstrated that pluggable, distributed architectures can support scalable medical imaging infrastructures across research and clinical environments. However, the evolution of their web interfaces has not always kept pace with modern frontend architectural practices, leading to limitations in extensibility models, integration patterns, and workflow continuity. This paper presents Dicoogle-Next, an architectural evolution of the Dicoogle Web layer that modernizes its frontend architecture while preserving the platform’s core plugin-based backend design. Rather than focusing solely on interface redesign, this work introduces a typed, build-time extensibility model for Web UI plugins, a decoupled REST-based integration layer, and an integrated web-based DICOM viewing component. The proposed architecture replaces dynamic runtime JavaScript injection with a structured extension framework based on React and TypeScript, providing stronger type safety, clearer extension points, improved maintainability, and better alignment with contemporary development tooling. The resulting system transforms the Dicoogle Web interface from a functionality-oriented administration console into a modular, workflow-oriented platform capable of supporting archive management, image review, and future AI-driven extensions within a unified architecture. This work demonstrates a practical modernization strategy for legacy PACS web consoles, balancing extensibility, maintainability, and usability in open-source medical imaging systems.

ABSTRACT. Wearable respiratory inductance plethysmography enablescontinuous monitoring of breathing during daily activities and provides a key sensing modality for a personalized digital twin applications. However, reliable machine learning models for respiratory signal quality assessment typically requires extensive manual annotations, which are costly and limit scalable deploy- ment across individuals. This work investigates how many labeled samples are required for semi-supervised and self-supervised learning methods to achieve performance comparable to fully supervised training in wearable office monitoring scenarios. Us- ing multimodal Respiratory Inductance Plethysmography record- ings, we systematically compare hierarchical supervised learning, co-training, MixMatch-based semi-supervised learning, and self- supervised feature learning under varying labeling budgets. Experimental results show that modern semi-supervised learning approaches achieve nearly supervised performance using less labeled data. These findings demonstrate that reliable respiratory monitoring models can be trained with substantially reduced annotation effort, enabling more scalable and individualized model generation. The proposed data-efficient learning frame- work supports the practical creation of personalized digital twins by leveraging large amounts of unlabeled wearable data while minimizing manual labeling requirements

ABSTRACT. Early identification of handwriting difficulties is crucial to prevent long-term academic and emotional consequences in children. Traditional assessments focus on the written product and are time-consuming, subjective, and unsuitable for large-scale screening. This study proposes a machine-learning approach for the ecological identification of handwriting difficulties in primary school children using a sensorized ink pen that records kinematic and dynamic signals during paper-and-pencil tasks. Graphomotor fluency tasks from a gold standard handwriting test were administered to 530 Italian pupils. A large set of process-related handwriting indicators was extracted and used to train binary classification models distinguishing children at risk from not-at-risk peers. Models were developed separately by grade and task using nested cross-validation and imbalance-aware training strategies. The best models achieved f1-scores up to 0.91. Model explainability revealed heterogeneous handwriting patterns, potentially enabling personalized training. These results support the feasibility of ecological, explainable machine learning tools for early, school-based screening of handwriting difficulties.

ABSTRACT. The COVID-19 pandemic has underscored the need for robust predictive frameworks to assess mortality risk from heterogeneous clinical data, where identifying shared attributes and severity patterns across patient groups is critical for improving outcomes. This study develops a data-driven and interpretable clustering methodology to identify patient groups associated with mortality using a comprehensive COVID-19 dataset from AIIMS Rishikesh, aiming to uncover risk patterns driven by demographic, clinical, and temporal features. A total of 2,985 records from 1,416 unique patients were analysed, and to effectively model heterogeneous feature types, a composite similarity matrix was constructed by integrating cosine similarity for numerical features with Jaccard similarity for categorical features, enabling the formation of clusters representing patients with shared attributes. Cluster characteristics were systematically evaluated to identify key determinants of mortality, including comorbidities, age, and respiratory indications. The proposed approach identified 180 distinct patient groups exhibiting unique demographic and clinical patterns, with high-mortality clusters predominantly characterised by Type 2 diabetes, hypertension, advanced age, and severe respiratory symptoms. Several clusters reflected critical illness trajectories requiring aggressive interventions, while a small subset demonstrated uniformly fatal outcomes, indicating extremely high-risk profiles. Overall, this study demonstrates that scalable, interpretable clustering techniques based on an integrated cosine–Jaccard similarity framework effectively capture complex clinical signals and provide actionable insights to support targeted care strategies and inform predictive risk modeling for public health management.

ABSTRACT. Understanding COVID-19 transmission patterns is critical for guiding targeted interventions and supporting proactive decision-making in pandemic control. This study constructs infection networks from COVID-19 epidemiological data provided by the Cyprus Ministry of Health, covering the period from March 2020 to May 2021 and representing the first four pandemic waves. Spatial transmission patterns are derived by aggregating postcode-to-postcode infections between confirmed cases. The aggregated spatial networks show stronger transmission between nearby areas and weaker transmission across distant ones. These results demonstrate that simple network analysis can support the understanding of spatial transmission dynamics and provide actionable information for epidemic control.

ABSTRACT. This study introduces two composite biomarkers for cardiovascular disease risk prediction, requiring only four non- invasively obtainable inputs: biological sex, maximum heart rate, ST-segment depression, and exercise-induced angina. No blood tests or imaging are required. On the UCI Cleveland dataset (n=297), logistic regression achieved an AUC of 0.910 with all 13 clinical features and 0.833 with the two biomarkers alone. Multi- site validation for two biomarkers across five hospital cohorts (n=905) yielded an AUC of 0.858, while cross-site validation pro- duced 0.857, substantiating generalizability. Since these measure- ments are obtainable via consumer smartwatch sensors [5], the biomarkers hold potential for at-home cardiovascular screening.

ABSTRACT. Artificial Intelligence (AI) is increasingly embedded in medical equipment that analyzes structured clinical data, yet its internal operation often remains insufficiently transparent to healthcare professionals responsible for its evaluation and adoption. This article presents a human-centered explainability pipeline designed to translate system-level behavior of AI-enabled medical equipment into clinically accessible narrative reports. The proposed framework integrates metaheuristic feature selection, supervised classification, and constrained large language model (LLM) generation to characterize parameter prioritization patterns, their stability across validation folds, and predictive performance under controlled conditions. To ensure methodological realism and reproducibility, the experimental framework emulates representative configurations of AI-enabled medical equipment using widely adopted feature selection and classification methods evaluated on publicly available clinical datasets. The pipeline generates structured explanations grounded in quantitative system artifacts while preventing unsupported diagnostic claims. Experimental results demonstrate robust predictive performance and stable parameter prioritization, while a human-centered evaluation involving healthcare professionals confirms the clarity, coherence, and practical interpretability of the generated reports. These findings support the proposed framework as a transparent and reproducible approach for improving human understanding of AI-assisted medical equipment.

ABSTRACT. This paper presents the findings of a case study exploring the implementation of Virtual Reality (VR) as an educational intervention for teaching empathy to medical students. Despite the growing use of Virtual reality (VR) in medical education, there is limited research examining its specific role in teaching empathy competencies. We conducted a study with 20 medical students at Aristotle University of Thessaloniki who participated in VR-based interventions implementing scenarios of patient-doctor interactions. Using qualitative methodologies (semi-structured interviews and observations), we identified 24 major themes demonstrating how VR creates embodied learning experiences that teach and develop empathy and communication skills. Our findings suggest VR is a feasible, useful, and effective adjunct to the traditional medical curricula for teaching soft skills while maintaining humanistic values in increasingly technological medical training. This work contributes to the growing evidence base for VR applications in medical education

ABSTRACT. Rapid Eye Movement (REM) sleep Behaviour Disorder (RBD) represents a critical clinical window into neurodegenerative processes, specifically alpha-synucleinopathies. However, the qualitative analysis of motor events in video-polysomnography (vPSG) remains highly subjective and time-consuming. This work or study presents a novel human-AI collaborative framework for typifying RBD movements, combining Recurrent All-Pairs Field Transforms (RAFT) for optical flow estimation and reconstruction-based Autoencoders for anomaly detection. The demonstration aims to evaluate how AI-driven motion visualisations can assist clinicians in categorising movement intensity, topography, and complexity. By engaging the scientific community in a structured labelling workshop, we aim to refine a robust ground-truth dataset while exploring the interpretability of deep learning models in sleep medicine.

ABSTRACT. While deep learning-based auto-segmentation has significantly reduced the manual workload in radiotherapy planning, the heavy computational requirements of state-of-the-art frameworks like nnU-Net remain a barrier to clinical adoption in resource-constrained environments. This paper addresses the challenge of high VRAM consumption and long training latencies by integrating a lightweight backbone, FFLUNet, into an established end-to-end prostate cancer segmentation for RT planning (PCS-RT) workflow. FFLUNet utilizes a multi-view feature fusion strategy and depth wise separable convolutions to maintain high-dimensional feature representation with a fraction of the parameters found in standard U-Net backbones. We evaluated the system using a dual-dataset approach: the public MSD Hippocampus dataset and a private TSCns dataset containing planning CT scans for high-grade gliomas. Our results demonstrate that replacing the standard nnU-Net backbone with FFLUNet reduces peak training memory from ≈7.9 GB to ≈3.6 GB and training time from ≈72 hours to ≈20 hours per fold on a single NVIDIA RTX 2080 Ti. Statistical analysis confirms that the slight variations in Dice performance do not represent a meaningful loss in segmentation quality. This work provides a scalable, cost-effective framework for deploying high-performance auto-segmentation in standard clinical settings without the need for enterprise-grade GPU clusters.

ABSTRACT. Early and reliable identification of retinal diseases is critical for preventing irreversible vision loss; however, automated diagnosis remains challenging due to subtle pathological variations and heterogeneous imaging conditions. In this work, we propose the Hierarchical Cross Attention Fusion Network (HCAF-Net), a clinically inspired dual backbone framework that enables structured interaction between global contextual reasoning and fine-grained lesion analysis. Unlike conventional fusion strategies based on simple concatenation or late integration, HCAF-Net performs stage-wise bidirectional cross-attention between a Swin Transformer and a MobileViT-v2 backbone, ensuring symmetric multi-level feature exchange across hierarchical representations. To further stabilize optimization and enforce representational consistency, we introduce auxiliary supervision with KL-divergence-based mutual distillation, allowing both branches to act as soft teachers for each other. Extensive experiments on RetinaMNIST, APTOS2019, and ODIR-5K demonstrate consistent improvements over state-of-the-art baselines, achieving accuracies of 0.683, 0.875, and 0.764, and AUC scores of 0.893, 0.949, and 0.932, respectively. The results validate that hierarchical bidirectional fusion, combined with distributional consistency regularization, provides a robust, dataset-agnostic framework for automated retinal disease classification.

ABSTRACT. Medical image classification often struggles with data scarcity and high-dimensional features, yet quantum kernels offer untapped potential for efficient feature extraction in such tasks. This study addresses these challenges by integrating digital-analog quantum kernels, rooted in the Digital-Analog Quantum Computing paradigm, into classical machine learning pipelines as non-parameterized feature extractors paired with lightweight classifiers. Using two binary benchmarks from the MedMNIST collection, we systematically vary quantum circuit architectures through iterative block repetitions and quantum pooling techniques to evaluate their influence on performance. Our analysis quantifies trade-offs between circuit depth and classification accuracy, alongside robustness to simulated bit-flip noise. Results reveal that moderate circuit depths enhance accuracy on larger datasets, although data scarcity limits gains on smaller cohorts, while quantum pooling effectively reduces dimensionality and cuts classical training costs without sacrificing pattern integrity. Despite the datasets' limited scale, which tempers generalizability claims, these findings strongly encourage further exploration of hybrid quantum-classical approaches for medical imaging.

ABSTRACT. Falls in institutionalized populations represent a major clinical and operational challenge, motivating decision-support tools for risk stratification and outcome prediction. This study presents a machine learning pipeline to predict post-fall mortality and related fall outcomes in nursing home residents using routinely collected structured variables, including demographic, mobility, and functional assessment data. Data from 2009 to 2022 were organized into task-specific datasets comprising a falls-only cohort of 1,368 records and a mixed cohort of 1,445 cases. Preprocessing involved one-hot encoding, MinMax scaling, and stratified 80/20 splits. To address severe class imbalance in mortality prediction, autoencoder-based data augmentation was applied, followed by grid-search model selection and SHAP-based interpretability. Fall occurrence prediction achieved the highest performance, with Gradient Boosting reaching a Macro-F1 of 0.89. Post-fall mortality prediction remained challenging due to extreme imbalance, yielding a Macro-F1 of 0.48, while the multiclass fall-count severity task achieved a Macro-F1 of 0.46. These findings demonstrate the potential of explainable machine learning for fall-related risk stratification in nursing homes while emphasizing the need for larger datasets, richer clinical variables, and external validation to improve rare-event prediction and support real-world implementation in long-term care settings.

ABSTRACT. Visual Acuity (VA) serves as the cornerstone of functional visual assessment and is critical for the diagnosis and monitoring of ophthalmic pathologies. While traditional tools like the Snellen and ETDRS charts are widely used, they face inherent limitations in balancing testing efficiency with measurement precision. Furthermore, current adaptive digital systems predominantly utilize Latin letter optotypes, creating significant cognitive barriers for specific Chinese demographics, particularly pediatric populations and illiterate adults. To bridge this gap, this study introduces Q2VA, a novel digital visual acuity testing module tailored specifically for the Chinese population, employing the “Tumbling E” optotype to eliminate linguistic bias and integrating a Bayesian adaptive algorithm with an active learning strategy to optimize assessment. The system dynamically estimates both the visual acuity threshold and the selection of stimuli. Validation through Monte Carlo simulations and psychophysical experimental results demonstrates that Q2VA can achieve high accuracy and precision of measurement within just 15 trials. The system maintains a low standard deviation (0.02 LogMAR) and high sensitivity to 0.03 LogMAR changes. Q2VA has the potential to provide a robust, efficient, and population-inclusive solution for high-precision visual acuity monitoring in eye clinics.

ABSTRACT. Counterfactual explanations are a promising approach for interpreting predictive models in healthcare. However, most existing counterfactual methods produce explanations with limited diversity and weak population-level coherence. This paper addresses this limitation by investigating whether organizing past clinical cases into clinically coherent clusters can improve the diversity of counterfactual explanations. We propose a clustering-aware, case-based framework that integrates contrastive retrieval through Nearest Unlike Neighbors with cluster-level organization and representative case selection to guide counterfactual generation. Feature modifications are driven by SHAP-based attributions and constrained to preserve clinical plausibility and validity. Experimental results on a Chronic Kidney Disease dataset show that clustering-driven case selection increases counterfactual diversity compared to other methods. These findings suggest that population-aware, case and cluster-based organization constitutes a relevant mechanism for enhancing the expressiveness of counterfactual explanations in healthcare predictive models.

ABSTRACT. This study investigates the factors influencing the application and intensity of coercive practices in psychiatric inpatient care, focusing on two French hospitals: CHU de Saint-Etienne and CH Drome Vivarais. Moving beyond the traditional binary prediction of “coercion vs. no coercion,” we introduce a novel coercion intensity score based on the frequency and type of restrictive measures during hospitalization. Twenty classification and several regression algorithms were evaluated, with XGBoost achieving the best classification performance and excels in intensity prediction. The models, while showing moderate performance comparable to the literature, identified key predictors including legal care modality, number of diagnoses, referral source, and patient age. Younger age, aggression, specific psychiatric diagnoses, and suicidal ideation were associated with higher coercion intensity.

ABSTRACT. Deep learning has achieved expert-level performance in automated electrocardiogram (ECG) diagnosis, yet the ``black-box'' nature of these models hinders their clinical deployment. Trust in medical AI requires not just high accuracy but also transparency regarding the specific physiological features driving predictions. Existing explainability methods for ECGs typically rely on post-hoc approximations (e.g., Grad-CAM and SHAP), which can be unstable, computationally expensive, and unfaithful to the model's actual decision-making process. In this work, we propose the ECG-IMN, an Interpretable Mesomorphic Neural Network tailored for high-resolution 12-lead ECG classification. Unlike standard classifiers, the ECG-IMN functions as a hypernetwork: a deep convolutional backbone generates the parameters of a strictly linear model specific to each input sample. This architecture enforces intrinsic interpretability, as the decision logic is mathematically transparent and the generated weights (W) serve as exact, high-resolution feature attribution maps. We introduce a transition decoder that effectively maps latent features to sample-wise weights, enabling precise localization of pathological evidence (e.g., ST-elevation, T-wave inversion) in both time and lead dimensions. We evaluate our approach on the PTB-XL dataset for classification tasks, demonstrating that the ECG-IMN achieves competitive predictive performance (AUROC comparable to black-box baselines) while providing faithful, instance-specific explanations. By explicitly decoupling parameter generation from prediction execution, our framework bridges the gap between deep learning capability and clinical trustworthiness, offering a principled path toward ``white-box'' cardiac diagnostics.

ABSTRACT. Intracranial electroencephalographic (iEEG) signals are typically captured using matrices with a high density of electrodes. This makes compressing these signals a valid objective for reducing memory usage in storage and transmission. Among the available compression methods, generative adversarial networks (GANs) have not been used. GANs are primarily associated with image generation and image-related applications. They can capture the probability distribution of variables in a dataset. Compression algorithms have minimally explored this characteristic, especially when applied to time series. We present an algorithm for multichannel signal compression based on a GAN, and we prove its efficacy on a dataset of intracranial electroencephalography (iEEG) signals. The compression method modifies the Backpropagation GAN (BPGAN) algorithm to work with signals instead of images. To get around the difficulties introduced by the peculiarities of the specific application, we propose an algorithm that separates channels of the iEEG signal into groups with high correlation and synchronization. The algorithm compresses groups separately by selecting a latent representation for each signal window among a finite set of possible representations. To validate the results, we decided to compare the results of an epilepsy seizure recognition experiment on the original and reconstructed signal. This approach presents encouraging results regarding compression ratio despite the introduction of some artifacts in the reconstruction. This could open for future improvements in applying GANs in compressing signals by increasing the reconstruction quality while keeping the high compression ratio.

ABSTRACT. Cross-subject generalization in EEG-based brain-computer interfaces (BCIs) remains challenging due to individual variability in neural signals. We investigate whether spectral representations offer more stable features for cross-subject transfer than temporal waveforms. Through correlation analyses across three EEG paradigms (SSVEP, P300, and Motor Imagery), we find that spectral features exhibit consistently higher cross-subject similarity than temporal signals. Motivated by this observation, we introduce ASPEN, a hybrid architecture that combines spectral and temporal feature streams via multiplicative fusion, requiring cross-modal agreement for features to propagate. Experiments across six benchmark datasets reveal that ASPEN is able to dynamically achieve the optimal spectral-temporal balance depending on the paradigm. ASPEN achieves the best unseen-subject accuracy on three of six datasets and competitive performance on others, demonstrating that multiplicative multimodal fusion enables effective cross-subject generalization.

ABSTRACT. This study examines the application of Lyapunov exponent analysis (LE) to characterize local field potentials (LFPs) from hippocampal brain slices in animal models, with a focus on distinguishing between basal and active states of hippocampal activity. We used LFP recordings obtained from hippocampal slices treated with kainic acid to induce active states, capturing transitions and sustained periods of activity. The signals were pre-processed to standardize their length and filtered using a 4th-order Butterworth bandpass filter to isolate gamma oscillations. LE analysis was used to assess the dynamical behavior of these signals, revealing that positive LE values indicate chaotic dynamics, which were prevalent in the active state recordings. Further analysis using time-series clustering distinguished patterns in progression from the basal to active states, suggesting that LE could serve as a biomarker for neurophysiological and pathological conditions, including Alzheimer's disease. Our findings suggest that LE analysis provides a novel approach to understanding the complex dynamics of the hippocampus, potentially contributing to the early diagnosis of neurodegenerative diseases.

ABSTRACT. Functional magnetic resonance imaging (fMRI) is a key tool for investigating spontaneous brain activity and functional brain organization. However, the reliability of downstream analyses critically depends on preprocessing quality, particularly in high-field acquisitions where susceptibility distortions, motion artifacts, and spatial misalignment may affect data interpretation.

This study evaluates the robustness of the fMRIPrep pipeline for preprocessing 7 Tesla fMRI data, with a particular focus on the reliability of its outputs. We propose a systematic validation framework specifically designed for ultra-high-field imaging, enabling a comprehensive and reproducible assessment of preprocessing performance. The framework is based on a stepwise analysis of pipeline outputs, including final derivatives, intermediate data, and quantitative metrics. It provides a structured evaluation of key preprocessing stages—motion correction, susceptibility distortion correction, anatomical–functional alignment, and spatial normalization to the MNI152 standard space—combining visual inspection, quantitative criteria, and expert assessment. This approach ensures a detailed validation of anatomical integrity, spatial consistency, and BOLD signal quality.

Results demonstrate that fMRIPrep effectively processes 7 Tesla fMRI data, producing reliable outputs with accurate functional–anatomical alignment and robust spatial normalization. The proposed framework highlights the consistency and quality of preprocessing across datasets. This work provides a transparent and reproducible evaluation strategy and establishes a methodological foundation for ultra-high-field fMRI analyses.

ABSTRACT. This paper presents an exploratory chirplet-based signal-processing study of wearable EEG recorded during outdoor cold-plunges in snow. It is based on the Muse EEG headbands with inertial measurements from the IMU (Inertial Measurement Unit) that is built into the Muse product. Because mobile EEG is strongly affected by motion artifacts, tri-axial gyroscope magnitude was used to reject high-motion intervals before chirplet-based time–frequency analysis. We analyze device-derived beta-band EEG estimates using Gaussian-chirplets and compare the resulting structure with conventional stationary Fourier-style representations. The aim is not to infer cognitive state or physiological cold response, but to evaluate whether the chirplet transform provides a compact representation of non-stationary wearable EEG segments recorded in a challenging outdoor environment. Results suggest that chirplet-based analysis reveals localized time–frequency structure.

ABSTRACT. Speech-based sideline screening for mild traumatic brain injury (mTBI), known as concussion, in sports-field environments is often contaminated by intense background noise, such as crowd cheering, whistles, and stadium announcements. This noise could compromise the reliability of speech-based screening. In this study, we present a controlled feature-level analysis of acoustic distortion induced by real-world sports-field noise and its variation across various enhancement strategies. Using clean audio-visual speech datasets collected from healthy speakers, we synthesise noisy speech by mixing several types of real sports-field noise at multiple signal-to-noise ratios (SNRs). From clean, noisy, and enhanced signals, we extract a set of evocative features commonly used in the field research and quantify noise-induced and enhancement-induced distortions at the individual-feature and feature-vector levels. The proposed analysis provides a principled basis and offers methodological foundations for assessing the robustness of speech analysis in noisy clinical and field-deployment scenarios.

ABSTRACT. Current visualization methods of cardiac electrical activity in the human ventricles depend on multiplanar reformats (short-axis and long-axis views) and three-dimensional volume rendering, making spatial navigation and orientation inherently complex, and increasing the potential of hindering critical anatomical and temporal patterns, such as those of re-entry. This study introduces a novel approach to flatten the geometry of the left ventricle, using disk harmonic mapping, to improve the exploration and analysis of such data, by creating tailored projections of the cardiac transmembrane potential onto a common reference domain inspired by the standardized polar reference model proposed by the American Heart Association. The proposed mapping technique is evaluated using angle and area distortion metrics and proved to be quasiconformal. Disk harmonic mapping presents a valuable addition to the visualization of the cardiac electrical activity patterns of the left ventricle, and it enables comparisons of datasets derived across different studies.

ABSTRACT. Aging leads to physiological decline and increased disease susceptibility, including cancer. The fruit fly Drosophila melanogaster serves as a powerful model to assess aging due to its short lifespan and conserved molecular pathways. The objective of this study was to implement interpretable convolutional neural networks (CNNs) and use transfer learning to assess the chronological age of Drosophila intestinal tissue from confocal microscopy images. A dataset of 557 flies was acquired, and images were preprocessed, trained, and evaluated using (a) a custom CNN architecture trained from scratch, and (b) three well-established CNN architectures, ResNet50V2, MobileNetV2, and InceptionV3, pre-trained on the ImageNet dataset. Single- (Y) and multi-channel (RGB) images were used as input to these models. Grad-CAM++ heatmaps were generated to interpret the decision-making process of the best-performing models. The results showed that CNNs can distinguish young (≤10 days) from old (≥40 days) intestinal tissues, with MobileNetV2 achieving the highest accuracy on multi-channel images (0.93) and both ResNet50V2 and MobileNetV2 performing best on single-channel images (0.92). The interpretability findings suggested that aging is driven by structural deterioration and changes in cell density and size. This study underscores the value of integrating advanced computational analysis into biological research.

ABSTRACT. Motor Units (MUs) compose the muscular structure and are the functional units of the motor command. For a MU, spikes correspond to motor neuron discharges that generate MU action potentials (MUAPs) captured by electromyography (EMG). Identifying MU activity from EMG is challenging because it requires both accurately assigning detected spikes to their corresponding MU and determining the number of active MUs. This process produces sequences of interspike intervals (ISIs)— the time between two successive spikes of the same MU— which are worth studying with dedicated complexity measures.

This paper proposes an entropy-based complexity measure, termed Dispersion Belief Permutation Entropy (DBPE), designed to capture subtle temporal patterns and uncertainty in ISI sequences and EMG. DBPE integrates dispersion-based symbolic dynamics with belief function theory to enhance sensitivity to hidden structural variations and robustness to noise. We investigate the sensitivity of DBPE to variations in MU recruitment and contraction-related changes in discharge patterns on a dataset comprising 73 MUs. % The results demonstrate that DBPE provides superior sensitivity to changes in discharge dynamics wrt aleardy established Dispersion Entropy (DE) and Belief Permutation Entropy (BPE). The proposed DBPE also offers greater sensitivity in discriminating between an increasing number of MUs, thus providing a useful tool for estimating the number of active MUs when fewer than five MUs are involved. These findings highlight DBPE as a powerful tool for quantitative characterization of neural discharge dynamics.

ABSTRACT. Stroke remains a major contributor to long-term disability and has prompted growing interest in serious games as engaging tools to support rehabilitation and, consequently, improve patient outcomes. Hence, this work presents the design and development of a therapist-driven rehabilitation platform that integrates motion tracking, games with progressive difficulty levels, a clinically inspired assessment test, and accessible interfaces. Combining principles of rehabilitation science and serious game design, the system enables therapists to monitor patient performance in real-time. Built in Unity3D and interfaced with the Microsoft Kinect V2 sensor, the platform employs visual and auditory feedback to reinforce correct execution and sustain motivation. A clinically relevant assessment test, inspired by the Fugl-Meyer Assessment (FMA), was developed to evaluate the motor function of post-stroke patients using clinically meaningful tasks rather than kinematic metrics extracted during gameplay, providing a reliable method for tracking patient progression over time. Also, three interactive prototypes: Window Cleaning, Object Catch, and Capsule were developed to target upper limb mobility, coordination, and reaction time. To evaluate the game quality, 14 participants performed the FMA-inspired assessment test, played all three prototypes and completed a questionnaire about their gameplay experience. The results showed that although the participants felt that the assessment test was an interesting tool to quantitatively track the person's motor evolution, the inaccuracies of the pose estimation method led to unreliable and inconsistent metric results. Regarding the games, the Capsule was the most preferred; however, in general, the participants felt very motivated while playing all the games. Nevertheless, there is still room for improvement, particularly in enhancing the pose tracking accuracy and implementing adaptive game progression in which level difficulty dynamically adjusts according to the player’s real-time performance.

ABSTRACT. Context and Motivation: Videogame addiction (VA), or Gaming Disorder, is recognized by the WHO in the ICD-11 and the APA in the DSM-5 as a significant mental health condition. Despite its clinical recognition, current diagnosis relies heavily on subjective self-report scales and questionnaires, such as the Game Addiction Scale (GAS), which necessitates direct intervention from health professionals. There is a critical need for innovative digital tools that provide objective scientific evidence to complement traditional psychological assessments. Purpose and Hypothesis: The ADICVIDEO project investigates the relationship between VA and objective patterns of emotional state, sleep quality, and fatigue. The central hypothesis posits that the degree of addiction can be quantified using real-time physiological and physical activity sensors, integrated with Machine Learning (ML) classifiers to identify risk patterns and biomarkers. Methodology: The research targets emerging adults (aged 18–30) within the university community. The study utilizes a multimodal approach across six sensorized stations equipped with Empatica E4 (physiological signals), Tobii Pro Glasses 3 (eye tracking), and Logitech Brio 4K cameras (facial recognition/FaceReader). Two core research protocols have been successfully implemented: the Transversal Study (Protocol PEIBA 1868-N-23), which focuses on initial usage habits and psychological well-being (n=440), and the Experimental Sessions (Protocol SICEIA 2024-1858), which involve synchronized biometric recording during gaming and sleep. Current Status and Progress: To date, the transversal phase is 100% complete, while experimental data collection stands at 75% (n=101). Key outcomes include the publication of player profiles in Computers & Education Open (2025) and the release of the PaGER-Sync ADICVIDEO dataset in Data in Brief (2025). Preliminary ML models (SVM, Random Forest) for emotion and fatigue classification are currently under development. Furthermore, clinical validation through a Delphi Expert Panel with psychiatrists and psychologists has been conducted to validate clinical dimensions of the ADICVIDEO model (Protocol SICEIA 2025-3504). Expected Impact and Future Horizon: The project aims to deliver diagnostic support tools for practitioners and early-warning systems for users. Future efforts are oriented toward P4 Medicine (Predictive, Preventive, Personalized, Participatory), utilizing explainable AI and biofeedback interfaces to enhance digital well-being and emotional self-management.

ABSTRACT. Abstract — The digital transformation of healthcare systems in Europe requires professionals who can design, manage, and evaluate interoperable health information infrastructures. This paper analyzes how healthcare professionals enrolled in the European master’s program ManagiDiTH (Managing the Digital Transformation in Healthcare) perceive interoperability implementation in their countries. Drawing on 72 student case studies from the course “Technologies in Interoperable Ecosystems,” we conduct a qualitative thematic analysis of interoperability experiences across Greece, Finland, Portugal, and cross-border settings. Using the General Conceptual Framework of eHealth Profiles and Competences, as the analytical lens, we map student-identified themes to the framework’s three profile domains (Health, Non-Health, ICT), the six phases of the eHealth service lifecycle (Plan, Build, Run, Enable, Manage, Use), and the competency decomposition into knowledge, skills, and attitudes. The findings reveal that students consistently identify competency needs spanning all three domains and all lifecycle phases, with particular emphasis on the Enable and Manage phases. Challenges reported by students—including legacy system integration, digital skills gaps, and data privacy concerns—map directly to specific competency areas defined in the framework.

ABSTRACT. Traditional centralized PACS architectures exhibit significant limitations under WAN deployments, particularly regarding latency and scalability. Although NoSQL databases have been explored to improve horizontal scalability, prior approaches typically separate metadata indexing from binary object storage, relying on distinct technologies and persistence models. Chunking mechanisms have also been used to handle large medical imaging objects, but mostly as technical workarounds rather than architectural design parameters. In contrast, this work proposes a unified, chunk-aware distributed persistence architecture where both DICOM metadata and image objects are managed within the same NoSQL infrastructure, and assesses its retrieval under realistic WAN conditions.

ABSTRACT. Standard Operating Procedures are essential instruments for ensuring consistency, regulatory compliance, and data quality in clinical research. Despite their recognized importance, SOP authoring remains a largely manual, unstructured process with limited tooling for systematic, multi-centre creation. This paper proposes WizarDoc, a modular, AI-assisted questionnaire-driven web application designed to streamline the production of SOP-aligned clinical study documentation. WizarDoc combines reusable template pools (governed question sets plus DOCX layouts) with a step-by-step wizard that dynamically generates contextual questions and AI-driven answer suggestions via a Retrieval-Augmented Generation (RAG) pipeline grounded in authoritative clinical standards. Captured identifier-keyed responses are exported as populated DOCX documents. The tool is publicly available at https://github.com/ieeta-mith/wizardoc.

ABSTRACT. The automated generation of medical discharge summaries using large language models (LLMs) promises efficiency gains for clinical documentation. However, factual inaccuracies, omissions of critical information, and confabulated content pose significant risks to patient safety and clinical uptake. Therefore, robust validation strategies are required before such systems can be safely deployed in routine care. However, many traditional evaluation metrics either rely on human-generated reference summaries or do not align well with human preferences. This paper aims at closing this gap and presents a multi-layered validation framework for AI-generated discharge summaries. We combine concept-based, metric-based, classifier-based, and LLM-based evaluation methods within an ensemble learning approach. The framework considers multilingual clinical environments and supports both real-time validation of individual summaries and longitudinal quality assurance for prompt and model updates. We report results from a proof-of-concept implementation on 21 real-world discharge summaries including human corrections. Our findings highlight the limited reliability of single automated metrics and demonstrate the value of ensemble-based validation aligned with clinician judgment.

ABSTRACT. The ankle-brachial index (ABI) is a crucial non- invasive diagnostic parameter in the diagnosis of peripheral arterial disease (PAD). The classic techniques used to obtain this index have several limitations, such as the need to use Doppler equipment, patient discomfort, and the time required to perform the test. In recent years, several studies have been conducted with the aim of mitigating these restrictions, including obtaining this index based on pulse wave velocity with biomedical signals such as the combination of ECG and PPG. These systems have been developed for clinical environments and use wired connections. This paper proposes an improved version of this previous methodology, incorporating real-time wireless data transmission to a mobile device and subsequent storage in a remote database, from which ABI values are estimated, with the aim of increasing the portability of the system and improving its applicability in domestic environments. The importance of home monitoring is justified by continuous tracking and data collection for the future development of predictive models. The new system was evaluated through a comparative study between the traditional method, the previous wired approach, and the new wireless solution in a clinical setting with 25 diabetic patients susceptible to PAD complications. The results show good accuracy, confirming its usefulness as a home-based alternative for measuring ABI, although further optimization is required.

ABSTRACT. Postural instability is a major contributor to fall risk in Parkinson’s disease (PD), yet early balance impairments are often not detected by standard clinical assessments. This study examines whether simple standing balance tasks, assessed using a single wearable inertial measurement unit (IMU), can effectively differentiate individuals with PD from age-matched healthy controls (HC). Trunk acceleration data were collected during four standing conditions: quiet standing on firm ground, standing on foam, and both conditions combined with a cognitive dual task. A broad set of time-domain, frequency-domain, and nonlinear postural sway features were extracted. Statistical analysis revealed significant group differences primarily during quiet standing on firm ground, where participants with PD showed increased sway path length, higher sway velocity, and greater jerk, indicating reduced smoothness of postural control. Machine learning classification using k-nearest neighbours (56% accuracy), support vector machines (62% accuracy), and decision trees (74% accuracy) demonstrated moderate baseline performance with the full feature set. After applying Joint Mutual Information feature selection, classification performance improved, with the decision tree achieving 82% accuracy. These results indicate that a brief, low-burden standing test using a single IMU, combined with targeted feature selection and machine learning, can sensitively detect PD-related postural instability and may serve as a practical screening or monitoring tool.

ABSTRACT. Understanding why tumors evade immune attack remains a central challenge in cancer research. We present a cross-cancer computational framework for discovering immune-dysregulated “traitor” cell states and linking them to actionable therapeutic targets. The framework integrates batch-aware representation learning with unsupervised manifold refinement to stabilize neighborhood geometry and identify ecosystem-specific resistance programs across single-cell cohorts from lung, colon, pancreatic, and melanoma tumors. Across cancers, refinement improved embedding compactness and preserved biologically coherent structure relative to baseline embeddings. Distinct resistance ecosystems emerged, including inflammatory myeloid programs in lung and colon, desmoplastic fibroblast hubs in pancreas, and cytotoxic interferon-enriched T-cell states in melanoma. Ecosystem-informed therapeutic prioritization highlighted actionable intervention points, including CXCR4-mediated stromal exclusion, CSF1R-driven macrophage survival, and IL6-associated fibro-inflammatory signaling. Expression-matched permutation testing and ligand–receptor interaction analysis supported the biological specificity of the discovered programs. These results demonstrate a generalizable, data-driven strategy for mapping dysfunctional immune ecosystems and accelerating cross-cancer therapeutic design.

ABSTRACT. Traditional speech-based anxiety detection has focused on acoustic prosody, such as pitch, energy, and voice quality. However, psycholinguistic research indicates that anxiety more directly affects cognitive processes involved in speech planning and lexical retrieval. To examine this discrepancy, we compared linguistic disfluency features, such as filled pauses and silence patterns, with acoustic prosodic features. The method can support practitioners in their clinical interviews as a decision-support tool for more objective measures of anxiety. Using 163 participants from the DAIC-WOZ dataset, we trained Random Forest classifiers under stratified 5-fold cross-validation. We evaluated the model’s performance across three settings: acoustic-only (n=5), disfluency-only (n=9), and combined fusion (n=14). The disfluency-only model significantly outperformed the acoustic baseline (F1=0.78 versus 0.67, p=0.008, Cohen’s d=2.18). Filled pause rate emerged as the strongest predictor (28.7% importance), with anxious participants exhibiting 74% higher rate. This research provides evidence that linguistic disfluency features are better speech biomarkers for detecting anxiety.

ABSTRACT. Induced pluripotent stem cell (iPSC)-derived natu ral killer (NK) cells offer a promising, scalable platform for next generation immunotherapy manufacturing. However, variability in differentiation efficiency across biological batches makes early quality assessment challenging. Because NK maturation requires several weeks, failures detected at late stages can lead to substan tial losses of time, cost, and experimental resources. To address this, we propose a deep learning (DL) framework for automated, early-stage, and non-destructive assessment of NK maturation status using bright-field microscopy images of Day-10 cultures. Our approach leverages a domain-specific preprocessing pipeline and modern DL architectures to capture subtle morphological indicators of hematopoietic commitment. The best-performing model (ResNet18) achieved an accuracy of 96.25%, with t-SNE and Grad-CAM analyses confirming that the network learns biologically relevant structural patterns. By providing early predictive insights through a label-free workflow, this framework offers a scalable strategy to standardize quality control in iPSC NK cell manufacturing.

ABSTRACT. Metagenomics has transformed microbial community analysis, yet challenges persist in computational efficiency, tool accessibility, and benchmarking standards. We present a metagenomic analysis pipeline that performs taxonomic identification in two phases. First, it optionally processes raw sequencing data through a preprocessing cascade including quality control, host DNA depletion, metagenomic assembly and binning. For the core analysis, it combines k-mer screening with an alignment system that switches between high-precision and fast tools depending on the genomic characteristics of the sample, followed by taxonomic assignment using exact matches and weighted lowest common ancestor strategies. Results show F1 scores above 0.80 across all biological domains and taxonomic levels, maintaining robust performance up to 30% sequence divergence. Validation on 30 Human Microbiome Project samples confirms accurate profiling of complex microbial communities (80.3% sensitivity, 20.5% MAPE) while detecting previously unreported taxa. The pipeline processes most datasets in under 60 minutes with a memory footprint of 2.82 GB.

ABSTRACT. Tremor is a characteristic sign of multiple neurological conditions and plays a critical role in assessment of disease severity and prognosis. Recent computer vision-based approaches using optical flow have demonstrated promising performance for video-based tremor evaluation. However, conventional optical flow representations encode directional components separately, which may introduce redundant computations and directional noise. In this study, we propose a motion magnitude-based framework to replace the conventional two-channel optical flow representation for modeling tremor dynamics in video sequences. The motion magnitude, derived from optical flow, provides a direction-invariant representation that better aligns with tremor amplitude characteristics while reducing computational complexity. To effectively exploit this representation, we design a dedicated feature extraction architecture tailored for magnitude-based inputs.

Models were trained and tested on a cohort of 65 patients with pathological resting tremor, from which 130 videos were obtained. Each video was graded by a neurology consultant on a score of 0-4 using the relevant part of the Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS). Our experiments demonstrate that the proposed method outperformed previous optical flow-based approaches. The framework achieved an overall accuracy of (71.76\%) and a within-one-grade accuracy of (98.35\%) compared to the human rater. These results indicate that the proposed motion magnitude representation provides an efficient and effective solution for automated tremor severity evaluation. The source code is publicly available at \url{https://github.com/lwp-1995/Motion_Magnitude_PD_Tremor}.

ABSTRACT. High internal AUCs (>0.95) are frequently reported for Alzheimer’s disease (AD) classifiers, yet such models often fail under clinical deployment. We attribute this gap to two compounding issues: (i) data leakage that inflates internal metrics and (ii) untested domain shift across cohorts. We introduce a leak-safe validation protocol with five mandatory safeguards and evaluate it on 3,238 subjects from OASIS-3 (population-representative; 23% AD; natural age distribution) and ADNI (trial-enriched; 72% AD; age-matched). Using FastSurfer-derived MRI morphometric features (and MRI+PET within-cohort analysis in OASIS), leak-safe models achieve realistic within-cohort performance (AUC 0.77–0.94) with limited age confounding (1.3–4.2% AUC contribution). We then perform strict bidirectional external validation without retraining. Training on ADNI and testing on OASIS-3 yields severe collapse (AUC 0.894→0.535; PR-AUC 0.32), whereas training on OASIS-3 and testing on ADNI shows upward generalization (AUC 0.838→0.851; PR-AUC 0.94). A sample-size learning curve in the ADNI→OASIS direction (N=142–571; six stratified subsamples) shows degradation does not decrease with more data and instead correlates positively with training size (Pearson r=0.847, p=0.033), suggesting larger trial-enriched sets increasingly internalize source-specific biases. Feature stability analysis identifies hippocampus/ICV ratio as the most cross-cohort stable biomarker. Overall, distributional mismatch—not data insufficiency—dominates cross-cohort generalization limits and is critical for deployment readiness.

ABSTRACT. Postural instability is a major source of disability in Parkinson’s disease, as impaired balance increases fall risk and accelerates loss of independence in daily life. In clinical practice, instability is commonly assessed through the pull test during outpatient visits. However, this task is difficult to replicate in telemedicine settings, where patients must perform evaluations at home, often without direct clinical supervision. Therefore, extracting quantitative markers of instability from simple standing tasks may provide a practical proxy for conventional assessment. This approach requires technologies capable of detecting subtle center-of-mass oscillations through non-invasive, easy-to-use systems. In this study, we evaluate a deep learning–based human pose estimation framework using an RGB-D camera to monitor patients during two 60-seconds standing tasks (eyes open and eyes closed). Kinematic features related to postural sway are estimated from body pose and correlated with reference clinical scales. Data from 40 patients recruited across two clinical centres show moderate agreement with clinical scores (Spearman ρ = 0.49), confirming the efficacy of the acquisition technology. Moreover, results suggest that 20 s recording with eye closed may be sufficient to capture clinically-relevant stability parameters.

ABSTRACT. Automated pathology image analysis is central to clinical diagnosis, but clinicians still ask which slide features drive a model’s decision and why. Vision–language models can produce natural language explanations, but these are often correlational and lack verifiable evidence. In this paper, we introduce an SQL-centered agentic framework that enables both feature measurement and reasoning to be auditable. Specifically, after extracting human-interpretable cellular features, Feature Reasoning Agents compose and execute SQL queries over feature tables to aggregate visual evidence into quantitative findings. A Knowledge Comparison Agent then evaluates these findings against established pathological knowledge, mirroring how pathologists justify diagnoses from measurable observations. Extensive experiments evaluated on two pathology visual question answering datasets demonstrate our method improves interpretability and decision traceability while producing executable SQL traces that link cellular measurements to diagnostic conclusions.

ABSTRACT. Vision-Language Models (VLMs) have recently demonstrated strong multimodal reasoning capabilities in medical imaging. However, most state-of-the-art systems are large-scale, proprietary, and predominantly trained on English corpora, limiting their applicability in low-resource clinical language settings such as Brazilian Portuguese. This work introduces three lightweight Vision-Language Models in Brazilian Portuguese for chest X-ray analysis, namely BodeVLM-RX, InternBodeVL-RX, and BodeMedRX, developed under a progressive learning paradigm and fine-tuned using the proposed Bode-CXR-VQA dataset, a translated and curated Portuguese adaptation of a large-scale radiology visual question-answering benchmark. The models are built upon distinct architectural foundations, including general-purpose and domain-specialized medical VLMs, enabling a controlled comparison of adaptation efficiency across different pretraining distributions. Experimental results evaluate the impact of scaling domain-specific supervision from 1k to 50k samples on closed-ended and open-ended clinical VQA tasks using Exact Match, BLEU, and ROUGE-L metrics, demonstrating consistent performance gains across all architectures, with markedly different scaling behaviors.

ABSTRACT. Background and Objective: Medical Visual Question Answering (Med-VQA) demands accurate multimodal interpretation of images and clinical questions, yet is hindered by data scarcity and high annotation costs. Vision-Language Models (VLMs) like CLIP offer strong zero-shot capabilities, but suffer from poor calibration and hallucinations in medical domains. This study introduces Active-CLIP, a hybrid framework combining zero-shot active learning with semi-supervised pseudo-label propagation to maximize annotation efficiency and performance in low-data regimes without fine-tuning the foundation model. Methods: Using frozen CLIP (ViT-B/32) as a zero-shot feature extractor and uncertainty estimator, Active-CLIP selects informative samples via Shannon entropy and propagates high-confidence pseudo-labels to the remaining pool using visual similarity in CLIP embedding space. A multimodal transformer is trained from scratch on the enriched set and evaluated on four public Med-VQA datasets (VQA-RAD, Path-VQA, Omni-Med-VQA-Mini, SLAKE) across labeling budgets of 10%–90%, compared against a real-labels-only baseline. Results: Active-CLIP consistently outperforms the baseline, with the largest gains at low ratios (10%–30%): up to +13% in BLEU-4 (Path-VQA) and +41% in ROUGE-L (SLAKE), alongside improved semantic alignment (higher AS, lower AE). Conclusions: Active-CLIP offers a promising, annotation-efficient solution for Med-VQA in data-scarce settings by leveraging zero-shot exploration and reliable pseudo-label exploitation.

ABSTRACT. Accurate segmentation of multiple sclerosis (MS) lesions from magnetic resonance imaging (MRI) is essential for assessing disease activity and progression. Among MRI sequences, FLAIR provides high lesion contrast, while T1-weighted images, though routinely acquired, encode lesion-related information in a less explicit manner. The difference in how lesion information is represented across these modalities makes MS lesion segmentation a well-established multimodal task to study cross-modal knowledge transfer. In this work, we investigate whether knowledge distillation inspired by the Learning Using Privileged Information (LUPI) paradigm can effectively transfer lesion-related information from a more informative modality (FLAIR) to a less informative one (T1-weighted). Our objective is to systematically assess the effectiveness of distillation in a controlled and clinically meaningful setting. We propose a teacher–student framework in which a teacher network is trained to segment lesions from FLAIR and T1-weighted images and guides a student trained exclusively on T1-weighted data. This setup enables the student to internalize lesion-specific representations that are not directly apparent in T1 contrast alone, providing a principled framework to evaluate cross-modal knowledge transfer in MS imaging. The proposed method is evaluated on two datasets, a public benchmark (ISBI 2015) and a clinical dataset from San Martino Hospital (Genoa, Italy), serving as a controlled experimental framework to quantify the benefits and limitations of LUPI-based distillation strategies in MS imaging.

ABSTRACT. Generative AI framework-based modeling and prediction of longitudinal human brain images offer an efficient mechanism to track neurodegenerative progression essential for the assessment of diseases like Alzheimer’s. Among the existing generative approaches, recent diffusion-based models have emerged as an effective alternative to generate disease progression images. Incorporating multi-modal and non-imaging attributes as conditional information into diffusion frameworks has been shown to improve controllability during such generations. However, existing methods do not explicitly ensure that information from non-imaging conditioning modalities is meaningfully aligned with image features to introduce desirable changes in the generated images, such as modulation of progression-specific regions. Further, more precise control over the generation process can be achieved by introducing progression-relevant structure into the internal representations of the model, lacking in the existing approaches. To address these limitations, we propose a diffusion auto-encoder-based framework for disease progression modeling that explicitly enforces alignment between different modalities. The alignment is enforced by introducing an explicit objective function that enables the model to focus on the regions exhibiting progression-related changes. Further, we devise a mechanism to better structure the latent representational space of the diffusion auto-encoding framework. Specifically, we assign separate latent subspaces for integrating progression-related conditions and retaining subject-specific identity information, allowing better controlled image generation. We have experimentally validated the performance of our model by evaluating on Alzheimer’s disease progression generation through various image similarity metrics and region-wise volumetric assessments. These results demonstrate that enforcing alignment and better structuring of the latent representational space of diffusion auto-encoding framework leads to more anatomically precise modeling of Alzheimer’s disease progression.

ABSTRACT. Learning from Demonstration (LfD) facilitates the transfer of human skills to robots through the analysis of motion-based demonstrations, but most approaches rely only on kinematics, limiting their use in surgical applications where interaction forces are critical. This work proposes a multimodal LfD framework that integrates position, velocity, orientation, force and torque data into a Hidden Markov Model (HMM). Thirty-three demonstrations of a puncture task on a deformable surface were collected from 11 participants and used to train the model. The UR3e robot reproduced the learned trajectory with sub-millimeter accuracy (RMSE = 0.1127 mm) and replicated key force patterns observed in human demonstrations. Results demonstrate the feasibility of incorporating force information into LfD, enhancing trajectory learning and contributing to the development of intelligent robotic systems for intraoperative assistance.

ABSTRACT. In endoscopic image analysis, conventional multi-task learning frameworks based on hard parameter sharing struggle to simultaneously satisfy the divergent requirements of lesion localization and qualitative characterization for feature representation. To address this, we propose GD-HMoE (Geometric-Decoupled Hierarchical Mixture of Experts), a framework driven by general structural prior distillation for joint lesion detection and classification. The proposed method constructs two complementary sub-modules—semantic and geometric experts—on a unified backbone, achieving feature decoupling through a dynamic routing mechanism. Furthermore, a task-oriented hierarchical feature aggregation strategy is designed, where structural features are primarily utilized for the localization task, while fused features support classification decisions. Experimental results on multi-center Inflammatory Bowel Disease (IBD) and iodine-stained Esophageal Cancer (EC) datasets demonstrate that GD-HMoE achieves superior performance in both classification and detection. The framework effectively mitigates negative transfer in multi-task learning and offers a novel structured modeling approach for medical image analysis.

ABSTRACT. Abstract—Knowledge Distillation (KD) is widely used to compress deep neural networks while maintaining high performance. Loss functions help in choosing the best strategy for image classification. This paper investigates hybrid distillation loss applied in KD for lung nodule malignancy classification on the unbalanced LIDC-IDRI dataset. We propose three hybrid distillation loss functions that combine MSE, Focal Loss, and MSE+Focal Loss for comparison with the baseline loss, using softened teacher outputs to mitigate class imbalance inherent in medical scenarios. Five teacher models (ResNet-50, DenseNet-121, ViT-Small, Swin-Transformer, MaxViT-Tiny) and two lightweight students (EfficientNet-B0, MobileNetV2) were evaluated across three temperatures (τ ) in 120 experiments, reducing the students’ parameters by 42 97% relative to their respective teachers. MSE-augmented distillation consistently outperformed the traditional distillation loss function. On the test set, MobileNetV2 achieved the highest gain (+4% AUC) with traditional loss and MSE at τ = 10, while EfficientNet-B0 improved by +1.3% with traditional loss and MSE at τ = 10. These findings indicate that symmetric penalization in probability space improves robustness under class imbalance. The source Code is available at https://github.com/graciellafavoreto/hybrid-kd-losses.

ABSTRACT. Continuous time generative models, including diffu- sion and flow-matching models, have shown strong performance in medical image synthesis, yet they produce outputs without any indication of how reliable their predictions are, a critical limitation for clinical applications. We propose an uncertainty-aware approach that augments these models with the ability to estimate predictive uncertainty by design, enabling them to jointly predict the generative update and its associated confidence. Uncertainty is integrated into the training objective and propagated along the sampling trajectory at inference time, yielding spatially resolved, output-aligned uncertainty maps without modifying the underlying generative formulation. We evaluate our approach on two medical image synthesis tasks, low dose CT denoising and T1-to-T2 MRI translation, across both diffusion and flow- matching models. Our method improves generation quality while producing meaningful uncertainty estimates, offering a practical path toward more transparent and trustworthy medical image synthesis.

ABSTRACT. The performance of deep learning models for surgical scene understanding is strongly limited by the availability of large-scale, diverse, and well-annotated datasets. In medical imaging, data acquisition is costly and constrained by privacy regulations, motivating the use of synthetic data generation. In this work, we investigate semantic mask-driven image synthesis for cataract surgery using both conditional generative adversarial networks and diffusion-based models. We introduce two complementary semantic augmentation strategies that operate directly at the mask level, enabling the generation of anatomically consistent yet diverse surgical scenes and expanding the effective semantic training distribution. Experimental results on the Cataract-1K dataset demonstrate that the proposed augmentation strategies significantly improve image quality and diversity, allowing generative models to overcome early performance saturation. Our findings highlight the importance of procedural semantic augmentation for scalable synthetic data generation in surgical imaging.

ABSTRACT. Cone-Beam Computed Tomography (CBCT) is widely used for real-time intraoperative imaging due to its low radiation dose and high acquisition speed. Despite its high resolution, CBCT suffers from more artifacts and thereby lower visual quality compared to conventional Computed Tomography (CT). A recent approach to mitigate these artifacts is synthetic CT (sCT) generation, translating CBCT volumes into the CT domain. We enhance sCT generation through multimodal learning, integrating intraoperative CBCT with preoperative CT. Beyond validation on two real-world datasets, we use a versatile synthetic dataset, to analyze how CBCT-CT alignment and CBCT quality affect resulting sCT quality. The results show that multimodal sCT consistently outperform unimodal baselines, with the greatest gains observed in well-aligned, low-quality CBCT-CT cases. We also show that these findings are reproducible in real-world clinical datasets.

ABSTRACT. Retinal prosthetic devices like Argus II help patients to partially restore vision by mapping the visual scenes to a spare grid of electrode phosphenes. For effective mapping on such highly sparse grids, image preprocessing becomes essential. This paper presents an object-centric hybrid preprocessing pipeline that addresses the limitations of existing methods through four contributions: (1) a single-object focus strategy combining spatial centrality, detection confidence, object area, and semantic class priority; (2) a GrabCut refinement stack with area; (3) a full perceptual simulation with CLAHE contrast enhancement and configurable electrode dropout, and (4) an open mobile and server deployment architecture to help participants see the expected outputs via virtual reality simulations. Our experiments with five scenarios and 16 participants show that the proposed pipeline leads to an increase in object recognition accuracy from 0% to 62.5%, and that the high-capacity YOLO11x object detector achieves an 80.77% accuracy without post hoc refinement. These results offer pragmatic guidance for deploying preprocessing systems on mobile hardware and establishing a reproducible framework for future vision algorithms.

ABSTRACT. Rapid Eye Movement (REM) sleep behaviour disorder (RBD) is an early biomarker of neurodegeneration but requires resource-intensive video-polysomnography for diagnosis. This study presents an interpretable pipeline for automated detection of idiopathic RBD (iRBD) using heart rate variability (HRV) features extracted from ECG during sleep. HRV metrics are computed separately for each sleep stage to capture stage-specific autonomic signatures. Selected features are used to train an Explainable Boosting Machine (EBM). The approach is evaluated on an age-matched cohort of 67 subjects (33 healthy controls and 34 iRBD) using nested Leave-One-Subject-Out cross-validation. Across ten random seeds, the REM stage yielded the best performance (AUC = 0.82 ± 0.02, accuracy = 0.75 ± 0.04). Four HRV features are consistently selected across folds, indicating stable autonomic markers of iRBD and highlighting the potential of single-lead ECG as a scalable screening tool.

ABSTRACT. Accurate ICD coding of clinical notes is essential for healthcare operations, supporting reimbursement, epidemiological surveillance, and quality measurement. Manual coding is labor-intensive, error-prone, and costly, motivating decades of research into automated coding. Recent neural approaches achieve strong performance but often function as black boxes, limiting their practical value in clinical settings. We propose a framework that achieves near-SOTA performance while providing two complementary forms of explanation. A BERT encoder is first pretrained with Jaccard-weighted supervised contrastive learning on multi-label clinical documents, then fine-tuned end-to-end with a label-aware chunk attention head that attends over 128-token segments rather than individual tokens. This chunk-level attention produces section-level explanations indicating which parts of a clinical note support each code prediction. The label attention classifier is ensembled with a chunk-level K-nearest neighbor retrieval system, providing a second form of interpretability through similar-case retrieval. On MIMIC-III with 1,275 ICD-9 codes, our ensemble achieves Micro-F1 of 0.482, rising to 0.538 on the 374 codes covering 80% of diagnoses. The code for reproducing our experiments is publicly available at https://github.com/ieeta-mith/ClinMap.

ABSTRACT. Leg Agility (LA) is a standard motor task for motor assessment in Parkinson's disease (PD), and recent work has explored its gamification to sustain patient engagement in unsupervised rehabilitation scenarios. However, the effect of this gamification on autonomic arousal during task execution has not been objectively characterised. This paper investigates electrodermal activity (EDA) arousal patterns in 33 PD patients performing traditional LA and two difficulty levels of a gamified Bouncing Ball (BB) exercise. A JMIM-based feature selection pipeline fed a Leave-One-Subject-Out XGBoost classifier with SHAP explainability, complemented by Mann-Whitney U testing with Benjamini-Hochberg FDR correction and per-leg-session monotonic trend analysis. The transition from LA to the game (BB1) is robustly detectable, with several spectral features at p < .001 and showing a monotonically increasing EDA trend in 70–80% of sessions. In contrast, discriminating between difficulty levels (BB1 vs BB2) yields only 61% accuracy, despite the statistical analysis results, revealing that difficulty-induced arousal changes are visible in the group, but individually heterogeneous. Inter-subject SHAP variance analysis identifies a rank divergence between feature importance and patient-to-patient variability in the BB1 vs BB2 comparison, a direct XAI signature of idiosyncratic responses, and pinpoints cvxPr_f_max as the most population-consistent biomarker. These findings suggest that gamification reliably increases autonomic arousal in PD patients, while difficulty modulation requires personalised rather than universal calibration, with implications for real-time adaptive game design in motor rehabilitation.

ABSTRACT. Voxel-level modeling of resting-state fMRI (rs-fMRI) for Parkinson’s disease (PD) diagnostic classification is challenging due to high dimensionality, scanner variability, and limited sample sizes. We propose a spatio-temporal graph convolutional network (ST-GCN) operating directly on voxel-wise BOLD time series, combining voxel-level harmonization with population-derived functional adjacency to preserve temporal dynamics and spatial structure. The framework was evaluated on a clinical cohort from the Biobizkaia Health Research Institute, augmented with healthy controls from the Parkinson’s Progression Markers Initiative (PPMI), achieving an AUC-ROC of 0.86 for PD vs. control classification. Edge-level interpretability analysis revealed stable discriminative patterns involving basal ganglia–thalamo–cortical circuits and associative cortical regions.

ABSTRACT. Learning-based 3D deformable registration is central to many clinical workflows for a plethora of applications. However, state-of-the-art models are memory and compute intensive, which limits deployment on local hospital workstations and other resource-constrained environments where larger memory footprint translates to slower inference. While post-training quantization has been widely explored in image classification and segmentation, it has not been systematically studied for medical image registration. We present, to our knowledge, the first systematic evaluation of low bit weight quantization for modern 3D registration across four architectures (VoxelMorph, NICE-Net, TransMorph, NICE-Trans) and three CT benchmarks (CrossTemporal, NLST, UCLA5DCT), comparing our novel mixed precision angle based quantizaton against the standard MSE based approaches. Our main contribution is an angle-aware mixed-precision scheme that assigns non-uniform bit-widths to convolutional layers using a novel importance score that combines network topology with the angular deviation between full-precision and quantized weight vectors. Inspired by our correlation study that shows that angle error correlates more strongly with registration loss than conventional weight MSE, establishing angular distortion as a more faithful proxy for performance degradation and the central signal driving our bit-assignment policy. Across all models and datasets, the proposed angle-aware mixed-precision strategy matches or improves target registration error and multi-organ Dice relative to RTN and DFQ at comparable or lower average bit-widths, while achieving substantial reductions in model size compared with FP32. These results indicate that carefully designed angle-aware quantization enables aggressive compression of 3D registration networks with minimal performance loss, making high-quality registration more practical in resource-limited clinical settings.

ABSTRACT. This paper presents a multimodal deep learning (DL) model for diabetic retinopathy (DR) referral that integrates retinal fundus images with clinically relevant data selected through an explainable process. Using Shapley Additive exPlanations (SHAP) across five machine learning (ML) models, we identified urinary albumin excretion, diabetes duration, insulin use, HbA1c, and systolic blood pressure as the most informative clinical features. We integrated these variables into an InceptionV3-based convolutional neural network (CNN) through late fusion and evaluated the model on two independent datasets from Hospital de Cl´ınicas de Porto Alegre (HCPA-2019: 2,522 images; HCPA-2021: 1,555 images). Compared with an image-only baseline, the multimodal model increased specificity from 56.7% to 64.7% in HCPA-2019 and from 72.4% to 77.5% in HCPA-2021, while maintaining sensitivity above 95% and an AUC above 0.93. These findings indicate that incorporating clinically interpretable metadata can reduce false-positive referrals and improve the clinical relevance of Artificial Intelligence (AI)- based DR screening.

ABSTRACT. Automated cephalometric landmark detection still constitutes a critical component of orthodontic diagnosis and treatment planning. Nevertheless, the current state of the art in heatmap regression and coordinate prediction is challenged by the presence of low-contrast images, highly variable anatomy, and the need to model complex spatial dependencies between landmarks. To overcome these difficulties, we propose Self Evolving Graph Neural Fields (SE-GNF), a new framework that combines implicit neural models with graph-based spatial reasoning to support accurate, continuous, and sub-pixel landmark localization. SE-GNF consists of three main components: a neural-field module that maps sampled coordinates to smooth spatial representations, a Graph Attention Network encoder that encodes anatomical dependencies between craniofacial structures, and a cross-attention-based detector with learnable landmark specific queries. Unlike discrete heatmap-based methods, SE-GNF provides differentiable coordinate predictions while maintaining anatomical relationships. The method generalizes to different cephalometric protocols without requiring any changes to the architecture. Experimentally, SE-GNF achieves mean radial errors of 0.98 mm (Test1) and 1.27 mm (Test2) on the ISBI 2015 challenge dataset (19 landmarks) and 0.70 mm on the ISBI 2023 challenge dataset (29 landmarks), thus providing a clinically sufficient level of precision for orthodontic landmark detection and treatment planning.

ABSTRACT. Accurate cephalometric landmark detection is fun damental to orthodontic diagnosis and treatment planning yet manual annotation is time-consuming and subject to inter observer variability. We propose a topology-aware framework for automatic localization of 19 anatomical landmarks on lateral cephalometric radiographs.Our approach leverages a ResNet-50 based feature extractor with a conditional diffusion model that produces anatomically consistent two-channel distance transform maps of bone and soft tissue anatomy. Global craniofacial topol ogy is represented by curve-level topology tokens of the mandible, maxilla, and cranial base, which direct the diffusion model to preserve anatomical consistency. Finally, anatomically consistent distance transforms are utilized by a diffeomorphic atlas-flow module that predicts a stationary velocity field and applies canonical atlas landmarks to patient anatomy using topology preserving diffeomorphisms. Our approach was evaluated using the ISBI 2015 benchmark dataset. Our method achieves a mean radial error of 1.09 ± 0.59 mm for Test1 and 1.22 ± 0.76 mm for Test2, with success detection rates of over 97% at 2.5 mm for Test1 and 93% at 2.5 mm for Test2. Our results show that topology-conditioned diffusion with diffeomorphic atlas warping provides a robust and accurate solution for landmark detection in cephalometric analysis

ABSTRACT. Dynamic assessment of diaphragmatic function is essential for monitoring respiratory status in preterm infants. Ultrasound enables bedside, real-time, non-invasive measurement of indices such as diaphragmatic thickness (DT), diaphragmatic thickening fraction (DTF), and diaphragmatic excursion (DE). However, frame-by-frame annotation of neonatal ultrasound data is costly and subject to operator variability, limiting the development of automated analysis methods. This study presents an automated system for simulated neonatal diaphragm ultrasound analysis and DT quantification. Using a Field II–based simulation framework, 20 B-mode ultrasound videos were generated (10 normal and 10 abnormal). Motion parameters including baseline DT, thickness modulation, breathing frequency, displacement amplitude, and cycle-wise gain were controlled to simulate physiological variation. Frame-wise diaphragm segmentation was performed using LightM-UNet to produce binary masks for automated DT measurement. Manual measurements were compared with automated measurements. Results showed strong agreement between manual and automated assessments, with high Spearman correlations (ρ = 0.95–0.99, p < 0.001), low relative error, and small standard error across both normal and abnormal cases. A significant reduction in DTF was observed in abnormal cases (37.1% in normal vs. 17.6% in abnormal; p = 0.021, AUC = 0.81), demonstrating the system’s potential to detect clinically relevant differences.

ABSTRACT. Synthetic data represented in the OMOP common data model (CDM) offers a safe foundation for research because it preserves the structure and clinical relationships of real-world datasets without exposing identifiable patient information. Synthetic data enables reproducible research, where datasets can be shared and used for benchmarking algorithms and analytics solutions can be prototyped before deployment with real data. However, there are limitations with using synthetic data such as a lack of clinical complexity, including atypical care pathways and rare comorbidities. Published case reports provide detailed descriptions of anonymised patient presentations, diagnoses, and outcomes, addressing the limitations arising from using synthetic data alone. This study generates OMOP data from published clinical case reports using Natural Language Processing (NLP) within an Extract, Transform, Load (ETL) pipeline. We created a novel case report corpus consisting of 118,653 open-access case report articles and associated supplementary files, and optimised for text processing. The MedCAT clinical NLP framework is used to extract and link medical concepts from free-text in the case reports to standard clinical terminologies. Using a purpose-built Python ETL package to map clinical terms to OMOP concepts and populate the clinical tables of the CDM, we generated an OMOP dataset of 83,290 patients with 1,490,318 condition records, 670,551 procedure records and 384,061 drug exposure records. Since the OMOP dataset is based on anonymised patients, it can be reused without restrictions. Future work will involve improving the accuracy of extracting clinical data and temporal events from case reports.

ABSTRACT. Cohort definition within the OHDSI ecosystem requires navigating the ATLAS interface or writing code against the WebAPI REST endpoints, imposing a steep technical barrier for clinical researchers without programming experience. This paper presents an MCP server that exposes the OHDSI WebAPI cohort definition subsystem as callable tools, covering vocabulary search, cohort authoring, execution, and reporting, enabling language model clients to construct and persist OHDSI-native cohort definitions through natural language interaction. To evaluate the approach, a small language model running locally on consumer-grade hardware was tasked with replicating a published cohort definition from the public OHDSI ATLAS instance using only natural language instructions. The model successfully reproduced the cohort structure and the majority of its clinical concepts, producing a valid definition that was accepted and stored on a live OHDSI instance in under five minutes, without any prior training on OHDSI workflows. These results demonstrate that well-designed MCP tool interfaces can offset model scale, enabling lightweight local models to execute OHDSI phenotyping workflows that would otherwise demand both domain expertise and significant computational resources.

ABSTRACT. Automated Characterization of Health Information at Large-scale Longitudinal Evidence Systems (ACHILLES) is a profiling tool developed by the OHDSI community to provide descriptive statistics of databases standardized to the OMOP Common Data Model. These characteristics are presented graphically in the ATLAS tool. However, this solution does not allow for database comparison across the data network. The proposed solution aggregates ACHILLES results files from databases in the network and displays the descriptive statistics through graphical dashboards. The tool is helpful to gain insight into the growth of the data network and is useful for the selection of databases for specific research questions. In the software demonstration, we show a first version of this tool that will be further developed in GREG in close collaboration with all our stakeholders, including OHDSI.

ABSTRACT. Federated techniques such as federated learning and federated analysis have emerged as a powerful paradigm for enabling multi-center research on sensitive clinical data while preserving patient privacy. In this study, we introduce a simulation framework that emulates a real-world federated research project focused on the analysis of multiple sclerosis (MS) patient data. The project comprises two components: an image segmentation task and a clinical data analysis task, where federated variants of survival analysis and Principal Component Analysis (PCA) are employed. To capture the complexity and heterogeneity of real clinical datasets, we construct a federation of high-fidelity synthetic cohorts designed to mirror MS-related clinical and demographic characteristics, while the imaging component leverages publicly available real-world datasets.

Our simulation replicates key elements of authentic federated workflows, including distributed data governance, site-specific preprocessing, model training across isolated nodes, and the secure aggregation of analytical outputs. This framework provides a realistic testbed for developing, evaluating, and benchmarking federated learning methods in the context of MS research.