Katerina D. Tzimourta

Brain–Computer Interface (BCI) systems are rapidly evolving and increasingly integrated into interactive environments such as gaming and Virtual/Augmented Reality. In such applications, user adaptability and engagement are critical. This study applies deep learning to predict user performance in a 3D BCI-controlled game using pre-game Motor Imagery (MI) electroencephalographic (EEG) recordings. A total of 72 EEG recordings were collected from 36 participants, 17 using the Muse 2 headset and 19 using the Emotiv Insight device, during left and right hand MI tasks. The signals were preprocessed and transformed into time–frequency spectrograms, which served as inputs to a custom convolutional neural network (CNN) designed to classify users into three performance levels: low, medium, and high. The model achieved classification accuracies of 83% and 95% on Muse 2 and Emotiv Insight data, respectively, at the epoch level, and 75% and 84% at the subject level, using LOSO-CV. These findings demonstrate the feasibility of using deep learning on MI EEG data to forecast user performance in BCI gaming, enabling adaptive systems that enhance both usability and user experience.

Abdominal Aortic Aneurysm (AAA) is a vascular disease characterized by abnormal dilation of the lower segment of the aorta (abdominal aorta), resulting from structural weakening and deformation of the aortic wall. Advanced imaging techniques, such as Computed Tomography (CT) and Magnetic Resonance Imaging (MRI), provide rich representations of a subject’s clinical condition, while recent studies have explored automated decision support using Deep Learning (DL). This review aims to summarize the current research efforts applying DL to AAA volumetric imaging, focusing on core tasks such as classification, detection, segmentation, and quantification. The methodology entails a systematic analysis of the relevant literature to focus on the types of DL architectures in use and preprocessing techniques commonly employed in decision support pipeline. In total, 67studies are reviewed. The reviewed studies report a wide variety of DL architectures, with segmentation tasks achieving high performance often exceeding 90% in evaluation metrics such as the Dice Similarity Coefficient. This review provides a comprehensive overview of present-day capabilities and limitations as the base for further work on DL-based AAA decision support systems.

The increasing complexity of data ecosystems, especially healthcare, highlights the urgent need for dynamic, user-centric consent management solutions. Traditional static consent models struggle to adapt to evolving privacy regulations, organizational needs, and user expectations. The CONSENT project introduces an innovative Consent Management Platform (CMP) that leverages Artificial Intelligence (AI) and Blockchain technologies to enable secure, transparent, and flexible management of consent in complex data workflows. By combining intelligent consent recommendation mechanisms with tamper-proof decentralized storage, the CONSENT platform aims to empower users with greater control over their data while facilitating organizational compliance with frameworks such as GDPR and CCPA. This paper presents the platforms vision, core technological pillars, and the planned evaluation strategy, including the anticipated implementation in healthcare and digital services, providing insights into how AI and Blockchain can reshape consent management for the healthcare digital age and beyond.

Introduction: This prospective descriptive study explored the disparities in perceptions and experiences regarding healthcare services between providers and users during the COVID-19 pandemic, with a specific focus on the impact of the digital divide on access to and quality of care. The study revealed significant inconsistencies in the experiences of healthcare providers and patients, particularly regarding the effectiveness of digital health interventions. Methods: This study was a prospective descriptive analysis conducted to evaluate and compare the use of electronic healthcare services between healthcare employees (HΕs) (N = 290) and consumers (Cs) (N = 263) from December 2024 to May 2025, utilizing an electronic survey after the COVID-19 pandemic. To ensure the statistical validity of the sample size, a power analysis was performed using G*Power 3.1.9.2 software. A questionnaire was developed to evaluate the readiness of healthcare employees and consumers for electronic healthcare services. It was validated to ensure reliability within this population and comprised 49 questions. Results: The response rate of the participants was 89.19%, and the Cronbach’s alpha for the questionnaire was 0.738. The study revealed notable differences in perceptions regarding health-related information and digital health technologies across genders and age groups. Specifically, 28.8% of females and 27.3% of males considered it important to be well-informed about health issues (χ² = 8.83, df = 3, p = 0.032). Conclusions: This research contributes to filling a gap in comparative analyses of provider and user perspectives, offering a comprehensive view of how digital health was adopted and experienced during a global crisis. Practically, it provides an evidence base to guide future interventions aimed at fostering more equitable, resilient, and user-friendly digital healthcare systems.

Human-centered design (HCD) has emerged as a critical approach for developing digital health technologies that are usable, acceptable, and effective within complex healthcare environments. Rooted in systems theory, ergonomics, and information systems research, HCD prioritizes the needs, capabilities, and limitations of diverse user groups - including patients, clinicians, and caregivers - throughout the design and implementation process. This review synthesizes current applications of HCD in four key domains: brain-computer interfaces (BCIs), augmented and virtual reality (AR/VR), artificial intelligence (AI)-based clinical decision support systems AI-CDSS, and mobile health (mHealth) technologies. It explores design frameworks, usability strategies, and models of human-technology collaboration that contribute to successful adoption and sustained use. Ethical and legal considerations - such as data privacy, informed consent, and algorithmic fairness - are also addressed, particularly in contexts involving biometric and neurophysiological data. While HCD practices have shown substantial potential to improve digital health outcomes, their implementation remains uneven across technologies and settings. Emerging trends, including adaptive personalization, explainable AI, and participatory co-design, are identified as promising directions for the development of more inclusive, trustworthy, and sustainable digital health innovations.

Kleefstra syndrome (KS) is a rare neurodevelopmental disorder associated with disruptions in the EHMT1 gene, often leading to intellectual disability, autism spectrum behaviors and epilepsy. The electroencephalogram (EEG) serves as a non-invasive tool to explore brain function in KS; yet, systematic characterizations of EEG features remain extremely limited. This review synthesizes current evidence on EEG findings in KS, highlighting the high prevalence of nonspecific abnormalities and seizures, but the absence of a consistent electrophysiological biomarker. Given the growing role of machine learning (ML) in extracting patterns from EEG data in related disorders—such as Angelman, Rett and Fragile X syndromes—this review explores how similar approaches could be adapted for KS. Despite promising perspectives, a lack of large-scale, publicly available EEG datasets hinders the application of ML methodologies in KS research. Future directions are proposed to address these gaps, including standardized EEG data collection, adoption of quantitative EEG analyses and integration of ML techniques adapted for small datasets. This multidisciplinary strategy holds potential for improving early diagnosis, monitoring and personalized interventions in Kleefstra syndrome.

Research interest in the application of electroencephalogram (EEG) as a non-invasive diagnostic tool for the automated detection of neurodegenerative diseases is growing. Open-access datasets have become crucial for researchers developing such methodologies. Our previously published open-access dataset of resting-state (eyes-closed) EEG recordings from patients with Alzheimer’s disease (AD), frontotemporal dementia (FTD), and cognitively normal (CN) controls has attracted significant attention. In this paper, we present a complementary dataset consisting of eyes-open photic stimulation recordings from the same cohort. The dataset includes recordings from 88 participants (36 AD, 23 FTD, and 29 CN) and is provided in Brain Imaging Data Structure (BIDS) format, promoting consistency and ease of use across research groups. Additionally, a fully preprocessed version is included, using EEGLAB-based pipelines that involve filtering, artifact removal, and Independent Component Analysis, preparing the data for machine learning applications. This new dataset enables the study of brain responses to visual stimulation across different cognitive states and supports the development and validation of automated classification algorithms for dementia detection. It offers a valuable benchmark for both methodological comparisons and biological investigations, and it is expected to significantly contribute to the fields of neurodegenerative disease research, biomarker discovery, and EEG-based diagnostics.

Background/Objectives: Maternal health risks remain one of the critical challenges in the world, contributing much to maternal and infant morbidity and mortality, especially in the most vulnerable populations. In the modern era, with the recent progress in the area of artificial intelligence and machine learning, much promise has emerged with regard to achieving the goal of early risk detection and its management. This research is set out to relate high-risk, low-risk, and mid-risk maternal health using machine learning algorithms based on physiological data. Materials and Methods: The applied dataset contains 1014 instances (i.e., cases) with seven attributes (i.e., variables), namely, Age, SystolicBP, DiastolicBP, BS, BodyTemp, HeartRate, and RiskLevel. The preprocessed dataset used was then trained and tested with six classifiers using 10-fold cross-validation. Finally, the performance metrics of the models erre compared using metrics like Accuracy, Precision, and the True Positive Rate. Results: The best performance was found for the Random Forest, also reaching the highest values for Accuracy (88.03%), TP Rate (88%), and Precision (88.10%), showing its robustness in handling maternal health risk classification. The mid-risk category was the most challenging across all the models, characterized by lowered Recall and Precision scores, hence underlining class imbalance as one of the bottlenecks in performance. Conclusions: Machine learning algorithms hold strong potential for improving maternal health risk prediction. The findings underline the place of machine learning in advancing maternal healthcare by driving more data-driven and personalized approaches.

Background: Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that poses critical challenges in global healthcare due to its increasing prevalence and severity. Diagnosing AD and other dementias, such as frontotemporal dementia (FTD), is slow and resource-intensive, underscoring the need for automated approaches. Methods: To address this gap, this study proposes a novel deep learning methodology for EEG classification of AD, FTD, and control (CN) signals. The approach incorporates advanced preprocessing techniques and CNN classification of FFT-based spectrograms and is evaluated using the leave-N-subjects-out validation, ensuring robust cross-subject generalizability. Results: The results indicate that the proposed methodology outperforms state-of-the-art machine learning and EEG-specific neural network models, achieving an accuracy of 79.45% for AD/CN classification and 80.69% for AD+FTD/CN classification. Conclusions: These results highlight the potential of EEG-based deep learning models for early dementia screening, enabling more efficient, scalable, and accessible diagnostic tools.