This research addresses the critical analytical gap facing Small and Medium Enterprises (SMEs). Despite generating substantial operational data, most SMEs lack the technical expertise and resources to extract actionable insights. This study presents a web-based platform designed to bridge this divide by democratizing access to advanced machine learning without requiring programming skills. The proposed framework integrates Exploratory Data Analysis (EDA), clustering algorithms, and classification models within an intuitive interface. The system automates complex technical operations, such as data preprocessing and algorithm selection, enabling non-technical users to perform sophisticated tasks like customer segmentation and predictive analytics. By removing technical barriers, this platform empowers entrepreneurs to transition from intuition-based to data-driven decision-making, optimizing operations and enhancing competitiveness in the modern digital economy.

This paper presents a meticulously conducted and extensive comparative analysis of the absorption patterns of European Structural and Investment Funds (ESIF) throughout the 2014–2020 programming period across a diverse array of European countries, with a particular emphasis on the Greek context. The primary objective of this study is to offer a comprehensive understanding of both shared trends and variations in fund absorption dynamics, thereby providing valuable insights into the intricate factors that shape the utilization of ESIF. Utilizing a robust methodological framework, this research undertakes an exhaustive examination of national and regional data to unravel the multifaceted determinants influencing absorption rates. Key determinants, including macroeconomic conditions, administrative efficiency, institutional capacity, and policy frameworks, are rigorously scrutinized to elucidate their nuanced impacts on ESIF absorption dynamics. Through this methodical approach, the study endeavors to provide nuanced insights into the efficacy of fund allocation mechanisms and the pivotal role played by socio-economic factors in steering regional development trajectories.

This study evaluates the evolution of economic integration within the European Union’s Single Internal Market by developing four de facto indicators that measure the realised intensity of the free movement of goods, services, capital and active population. Using harmonised data for all EU-27 Member States from 1995 to 2020, each indicator is constructed as the normalised average of inward and outward flows relative to GDP or total active population. Data gaps were addressed through controlled interpolation, while cross-source harmonisation ensured conceptual consistency. Integration increased substantially across all freedoms, but unevenly. Goods trade grew by about 50

Background: Ulcerative colitis (UC) is a chronic inflammatory bowel disease characterized by continuous inflammation of the colon and rectum. Accurate disease assessment is essential for effective treatment, with endoscopic evaluation, particularly the Mayo Endoscopic Score (MES), serving as a key diagnostic tool. However, MES measurement can be subjective and inconsistent, leading to variability in treatment decisions. Deep learning approaches have shown promise in providing more objective and standardized assessments of UC severity. Methods: This study utilized publicly available endoscopic images of UC patients to analyze and compare the performance of state-of-the-art deep neural networks for automated MES classification. Several state-of-the-art architectures were tested to determine the most effective model for grading disease severity. The F1 score, accuracy, recall, and precision were calculated for all models, and statistical analysis was conducted to verify statistically significant differences between the networks. Results: VGG19 was found to be the best-performing network, achieving a QWK score of 0.876 and a macro-averaged F1 score of 0.7528 across all classes. However, the performance differences among the top-performing models were very small suggesting that selection should depend on specific deployment requirements. Conclusions: This study demonstrates that multiple state-of-the-art deep neural network architectures could automate UC severity classification. Simpler architectures were found to achieve competitive results with larger models, challenging the assumption that larger networks necessarily provide better clinical outcomes.

Large Language Models (LLMs) have revolutionized natural language processing but present significant technical and legal challenges when confronted with the General Data Protection Regulation (GDPR). This paper examines the complexities involved in reconciling the design and operation of LLMs with GDPR requirements. In particular, we analyze how key GDPR provisions—including the Right to Erasure, Right of Access, Right to Rectification, and restrictions on Automated Decision-Making—are challenged by the opaque and distributed nature of LLMs. We discuss issues such as the transformation of personal data into non-interpretable model parameters, difficulties in ensuring transparency and accountability, and the risks of bias and data over-collection. Moreover, the paper explores potential technical solutions such as machine unlearning, explainable AI (XAI), differential privacy, and federated learning, alongside strategies for embedding privacy-by-design principles and automated compliance tools into LLM development. The analysis is further enriched by considering the implications of emerging regulations like the EU’s Artificial Intelligence Act. In addition, we propose a four-layer governance framework that addresses data governance, technical privacy enhancements, continuous compliance monitoring, and explainability and oversight, thereby offering a practical roadmap for GDPR alignment in LLM systems. Through this comprehensive examination, we aim to bridge the gap between the technical capabilities of LLMs and the stringent data protection standards mandated by GDPR, ultimately contributing to more responsible and ethical AI practices.

Antibiotic resistance is a global health crisis exacerbated by the misuse of antibiotics in healthcare, agriculture, and the environment. In an intensive care unit (ICU), where high antibiotic usage, invasive procedures, and immunocompromised patients converge, resistance risks are amplified, leading to multidrug-resistant organisms (MDROs) and poor patient outcomes. The human microbiome plays a crucial role in the development and dissemination of antibiotic resistance genes (ARGs) through mechanisms like horizontal gene transfer, biofilm formation, and quorum sensing. Disruptions to the microbiome balance, or dysbiosis, further exacerbate resistance, particularly in high-risk ICU environments. This study explores microbiome interactions and antibiotic resistance in the ICU, highlighting machine learning (ML) as a transformative tool. Machine learning algorithms analyze high-dimensional microbiome data, predict resistance patterns, and identify novel therapeutic targets. By integrating genomic, microbiome, and clinical data, these models support personalized treatment strategies and enhance infection control measures. The results demonstrate the potential of machine learning to improve antibiotic stewardship and predict patient outcomes, emphasizing its utility in ICU-specific interventions. In conclusion, addressing antibiotic resistance in the ICU requires a multidisciplinary approach combining advanced computational methods, microbiome research, and clinical expertise. Enhanced surveillance, targeted interventions, and global collaboration are essential to mitigate antibiotic resistance and improve patient care.

Accurate classification of skin lesions into benign or malignant categories is crucial for early diagnosis and treatment of dermatological conditions. In this study, we present a comprehensive evaluation of Vision Transformer (ViT) models for binary classification tasks using a curated subset of the DermNet dataset. By leveraging a pre-trained ViT model fine-tuned on domain-specific data, we achieve a test accuracy of 94

Artificial intelligence (AI) has significantly driven advancement in the healthcare field by enabling the integration of highly advanced algorithms to improve diagnostics, patient surveillance, and treatment planning. Nonetheless, dependence on sensitive health data and automated decision-making exposes such systems to escalating risks of privacy breaches and is under rigorous regulatory oversight. In particular, the EU AI Act classifies AI uses pertaining to healthcare as “high-risk”, thus requiring the application of strict provisions related to transparency, safety, and privacy. This paper presents a comprehensive overview of the diverse privacy attacks that can target machine learning (ML)-based healthcare systems, including data-centric and model-centric attacks. We then propose a novel privacy-preserving architecture that integrates federated learning with secure computation protocols to minimally expose data while ensuring strong model performance. We outline an ongoing monitoring mechanism compliant with EU AI Act specifications and GDPR standards to further improve trust and compliance. We further elaborate on an independent adaptive algorithm that automatically tunes the level of cryptographic protection based on contextual factors like risk severity, computational capacity, and regulatory environment. This research aims to serve as a blueprint for designing trustworthy, high-risk AI systems in healthcare under emerging regulations by providing an in-depth review of ML-specific privacy threats and proposing a holistic technical solution.

Background: Justice-involved adolescents exhibit high rates of mental health disorders with complex comorbidity patterns. Understanding these patterns is crucial for developing targeted interventions in this vulnerable population. Methods: We applied multiple machine-learning techniques to electronic records from 124 justice-involved adolescents (11–21 years; mean = 15.7 ± 1.9). Analyses included association rule mining, K-Means clustering with t-SNE visualization, and topic modeling of clinicians’ recommendation notes. Results: Hyperkinetic disorders (F90.0/F90.1) and family-stress factors (Z63.5) together accounted for approximately 45% of all ICD-10 entries. A four-cluster K-Means solution built on age + F-codes alone showed weak separation (silhouette = 0.044), whereas adding Z-codes markedly improved cohesion (silhouette = 0.468) and isolated a distinct hyperkinetic–family-stress subgroup. Association-rule mining returned one robust rule, F81 → F90.0 (support = 0.048, confidence = 0.46, lift = 1.59), underscoring the frequent co-diagnosis of learning and attention-deficit disorders. Topic modeling of clinicians’ recommendation notes recovered five coherent intervention themes—vocational guidance, parent counseling, psycho-education, family psychotherapy, and psychiatric follow-up—which aligned closely with the data-driven clusters. Conclusions: These findings demonstrate how routine clinical data can reveal actionable comorbidity profiles and guide tailored interventions for complex adolescent mental-health presentations.

The growing epidemic of childhood obesity is a major threat to their overall development and poses a number of challenges for health systems. We propose an integrated framework to comprehensively address childhood obesity. The proposed architecture addresses essential data management and pre-processing functionalities to support scalable, secure, and privacy-preserving data processing in distributed environments. We are also incorporating a health data-driven AI approach for predictive analytics and decision support. There is additionally a User Engagement Layer, which serves as the main point of interaction for users. It connects individuals to system capabilities, facilitating data collection, progress monitoring, and insights. Finally, we present four serious games designed to address protective factors (such as physical activity and healthy eating) and mitigate risk factors (such as excessive screen time and unhealthy food choices). The identified educational objectives were translated into game elements including goal setting, social support, and positive reinforcement. In order to facilitate our approach, we have described the essential data flows and user interactions within our Biobank architecture.

The ASCAPE project aims to improve the health-related quality of life of prostate cancer patients using artificial intelligence-driven solutions. This study tries to unravel the complex relationships between patient data variables and urinary incontinence (UI), and post-radical prostatectomy using the ASCAPE datasets. We employed a Generalized Additive Model to analyze patient-reported outcomes on UI (QLQ PR25 questionnaires over a 12-month period), and objective data derived from wearable devices. Our findings showcase age and comorbidities as the main predictors of incontinence severity, whereas physical activity failed to show any significance in our model. Our study highlights the importance of a personalized approach to incontinence care, where patient characteristics and recovery patterns are considered when developing treatment plans.

Mental fatigue induced by sleep deprivation impairs cognitive performance and disrupts large-scale brain network communication, yet the specific functional connectivity (FC) patterns underlying this disruption remain insufficiently understood. This study proposes an explainable machine learning framework for detecting sleep-related mental fatigue using electroencephalography (EEG)-based FC features during a working memory task. As such, a mental fatigue experiment was performed prior and after 24 hours of sleep deprivation and functional connectivity was estimated using the Phase Lag Index (PLI), producing high-dimensional connectivity matrices in five frequency bands (delta, theta, alpha, beta and gamma). LASSO feature selection was combined with several ML classifiers (SVM, RF, LDA, KNN) to identify the most discriminative features, which were subsequently interpreted according to their physiological role. Nineteen FC features were selected among the different brain frequencies and evaluated across multiple classifiers, with SVM-RBF achieving the highest accuracy (0.97). Topological and spectral analyses showed that fatigue-related alterations were concentrated in frontal and central brain regions, primarily within delta and theta bands, implicating disruptions in executive control, sensorimotor integration, and slow-frequency network dynamics. These findings highlight the sensitivity of low-frequency FC patterns to sleep deprivation and demonstrate that sparse, interpretable models can robustly characterize fatigue-related neural changes.

The growing interest in improved rehabilitation systems and assistive technologies for individuals with motor impairments necessitates the need for new applications of Deep Learning approaches for Brain–Computer Interface (BCI) implementation. This study investigates the application of Deep Learning techniques, specifically the Hierarchical 3D Convolutional Neural Network (H3DCNN) model, for enhancing classification systems utilizing electroencephalography (EEG) data. As such, topographic maps were extracted from EEG signals in a real motion task experiment integrating 4 different motions. The H3DCNN model was then employed in a step-wise fashion to classify and decode the EEG signals, demonstrating its effectiveness in distinguishing between different movement intentions. Moreover, three different optimizers were implemented, including RMSprop, Adam, and Stochastic Gradient Descent (SGD), to further assess and enhance the model performance. The findings indicate that the integration of advanced deep learning techniques can significantly enhance the accuracy and reliability of BCI systems, with RMSprop and SGD showing superior results in terms of accuracy. Moreover, our results illustrate the possibility of decoding neural mechanisms via deep learning paradigms, paving the way for future developments in BCI applications, thus aiming to improve the quality of life for individuals with motor impairments.

Metabolic disorders, including type 2 diabetes mellitus (T2DM), obesity, and metabolic syndrome, are systemic conditions that profoundly impact the skin microbiota, a dynamic community of bacteria, fungi, viruses, and mites essential for cutaneous health. Dysbiosis caused by metabolic dysfunction contributes to skin barrier disruption, immune dysregulation, and increased susceptibility to inflammatory skin diseases, including psoriasis, atopic dermatitis, and acne. For instance, hyperglycemia in T2DM leads to the formation of advanced glycation end products (AGEs), which bind to the receptor for AGEs (RAGE) on keratinocytes and immune cells, promoting oxidative stress and inflammation while facilitating Staphylococcus aureus colonization in atopic dermatitis. Similarly, obesity-induced dysregulation of sebaceous lipid composition increases saturated fatty acids, favoring pathogenic strains of Cutibacterium acnes, which produce inflammatory metabolites that exacerbate acne. Advances in metabolomics and microbiome sequencing have unveiled critical biomarkers, such as short-chain fatty acids and microbial signatures, predictive of therapeutic outcomes. For example, elevated butyrate levels in psoriasis have been associated with reduced Th17-mediated inflammation, while the presence of specific Lactobacillus strains has shown potential to modulate immune tolerance in atopic dermatitis. Furthermore, machine learning models are increasingly used to integrate multi-omics data, enabling personalized interventions. Emerging therapies, such as probiotics and postbiotics, aim to restore microbial diversity, while phage therapy selectively targets pathogenic bacteria like Staphylococcus aureus without disrupting beneficial flora. Clinical trials have demonstrated significant reductions in inflammatory lesions and improved quality-of-life metrics in patients receiving these microbiota-targeted treatments. This review synthesizes current evidence on the bidirectional interplay between metabolic disorders and skin microbiota, highlighting therapeutic implications and future directions. By addressing systemic metabolic dysfunction and microbiota-mediated pathways, precision strategies are paving the way for improved patient outcomes in dermatologic care.

Background/Objectives: Melanoma, an aggressive form of skin cancer, accounts for a significant proportion of skin-cancer-related deaths worldwide. Early and accurate differentiation between melanoma and benign melanocytic nevi is critical for improving survival rates but remains challenging because of diagnostic variability. Convolutional neural networks (CNNs) have shown promise in automating melanoma detection with accuracy comparable to expert dermatologists. This study evaluates and compares the performance of four CNN architectures—DenseNet121, ResNet50V2, NASNetMobile, and MobileNetV2—for the binary classification of dermoscopic images. Methods: A dataset of 8825 dermoscopic images from DermNet was standardized and divided into training (80%), validation (10%), and testing (10%) subsets. Image augmentation techniques were applied to enhance model generalizability. The CNN architectures were pre-trained on ImageNet and customized for binary classification. Models were trained using the Adam optimizer and evaluated based on accuracy, area under the receiver operating characteristic curve (AUC-ROC), inference time, and model size. The statistical significance of the differences was assessed using McNemar’s test. Results: DenseNet121 achieved the highest accuracy (92.30%) and an AUC of 0.951, while ResNet50V2 recorded the highest AUC (0.957). MobileNetV2 combined efficiency with competitive performance, achieving a 92.19% accuracy, the smallest model size (9.89 MB), and the fastest inference time (23.46 ms). NASNetMobile, despite its compact size, had a slower inference time (108.67 ms), and slightly lower accuracy (90.94%). Performance differences among the models were statistically significant (p < 0.0001). Conclusions: DenseNet121 demonstrated a superior diagnostic performance, while MobileNetV2 provided the most efficient solution for deployment in resource-constrained settings. The CNNs show substantial potential for improving melanoma detection in clinical and mobile applications.

Innovation systems consist of different organisations from the quadruple helix, as well as the interactions and linkages between them. Smart technologies and ICT play a key role in the efficiency of systems. At the same time, regional scale is considered crucial for studying innovation in systems. However, the lack of many important data at the regional level compounds the efforts to study them. The paper proposes a novel methodological approach to the regionalisation of national-level indicators in order to address this issue. This is based on the model fit approach, using regressions to “regionalise” national-level indicators based on similar indicators that are available. The approach is tested on the data for Greek NUTS 2 regions and produces regional-level estimates for four innovation indicators, based on four available indicators that are found to be strongly correlated to them. However, the same approach can be used for any EU country or the whole of the EU. The results, their prospects for future research, and potential applications are considered. Overall, the availability of regional-level indicators is considered crucial for the formulation of impactful development policies.

Background: The ASCAPE project aims to improve the health-related quality of life of cancer patients using artificial intelligence (AI)-driven solutions. The current study employs a comprehensive dataset to evaluate sleep and urinary incontinence, thus enabling the development of personalized interventions. Methods: This study focuses on prostate cancer patients eligible for curative treatment with surgery. Forty-two participants were enrolled following their diagnosis and were followed up at baseline and 3, 6, 9, and 12 months after surgical treatment. The data collection process involved a combination of standardized questionnaires and wearable devices, providing a holistic view of patients’ QoL and health outcomes. The dataset is systematically organized and stored in a centralized database, with advanced statistical and AI techniques being employed to reveal correlations, patterns, and predictive markers that can ultimately lead to implementing personalized intervention strategies, ultimately enhancing patient QoL outcomes. Results: The correlation analysis between sleep quality and urinary symptoms post-surgery revealed a moderate positive correlation between baseline insomnia and baseline urinary symptoms (r = 0.407, p = 0.011), a positive correlation between baseline insomnia and urinary symptoms at 3 months (r = 0.321, p = 0.049), and significant correlations between insomnia at 12 months and urinary symptoms at 3 months (r = 0.396, p = 0.014) and at 6 months (r = 0.384, p = 0.017). Furthermore, modeling the relationship between baseline insomnia and baseline urinary symptoms showed that baseline insomnia is significantly associated with baseline urinary symptoms (coef = 0.222, p = 0.036). Conclusions: The investigation of sleep quality and urinary incontinence via data analysis through the ASCAPE project suggests that better sleep quality could improve urinary disorders.

Four indicators corresponding to the four targets of the European Monetary Union were calculated. The study showed that: (a) concerning the deviation of state's general government deficit/surplus from 3% of gross domestic product (GDP), all member states had reached their target, with the exception of Cyprus, which was slightly under the target, (b) concerning the deviation of state's general government debt from 60% of GDP, half of all European Union (EU) member states did not reach their targets, and there was a lot to be done, especially from the EU15 member states, (c) concerning the deviation of state's inflation rate from the mean of the three states with best results of +1.5%, it was observed that the average value of EU28 member states had reached the final target, mainly due to the performances of the EU15 member states, (d) and concerning the deviation of state's interest rate from the mean of the three states with the best results of +2%, it was observed that the average value of EU28 member states had reached the final target.

This research ventures into the critical public health challenge of childhood obesity by exploring the dynamic interplay between psychological well-being and Body Mass Index (BMI) throughout various developmental stages of childhood. It delves into how emotional regulation, attachment dynamics, and social relationships correlate with obesity from early childhood to adolescence. Highlighting key findings, such as the negative correlation between psychological resilience and higher BMI in young children, the impact of social relationships on obesity risk during pre-adolescence, and the link between adaptive emotional strategies and higher BMI in adolescents, this study brings to the fore the nuanced relationship between psychological factors and obesity. Psychological metrics in this study were obtained via referenced questionnaires, leading up to the utilization of the interdisciplinary process of bioinformatics. Utilizing the interdisciplinary process of bioinformatics, this research synergizes psychometric and biomedical data to unearth psychological markers critical for crafting targeted, age-appropriate interventions. This study advocates for a holistic healthcare approach, emphasizing the integration of psychological support within obesity prevention and management strategies, thereby underscoring the indispensable role of psychological factors in the fight against childhood obesity. The application of bioinformatics methods to analyze complex datasets demonstrates how collaboration across medical specialties can enrich our understanding and response to childhood obesity, contributing significantly to the development of comprehensive, bioinformatics-enhanced healthcare solutions.

A crucial aspect in developing machine learning algorithms (or any type of predictive models) is the comparison of different algorithmic candidates based on evaluation criteria that measure their accuracy and practical value in terms of successfully capturing the complexity of the underlying problem and generalizing in a wide range of real-world scenarios. In the domain of diabetes mellitus machine learning techniques are widely employed in the prediction of future values of glucose concentration in the blood in order to assist the patient in avoiding deviations from the normo-glycemic value range and the consequences of hyper- and hypoglycemia. In the relative literature there is an apparent lack of a uniformly adopted evaluation metric which could combine the clarity and direct comparative nature of statistical mathematical formulas dealing with prediction error residuals (e.g., the RMSE) with the clinical insights and the visual-qualitative approach of clinical evaluation tools (e.g., Clarke's EGA). Mean Adjusted Exponent (MADEX) error metric, proposed in this paper, attempts to address this need by providing a validation tool based on an easy to implement mathematical formula, that incorporates adjustable parameters of clinical significance.