Technological advances and societal changes in recent years have contributed to a shift in traditional care models and in the relationship between patients and their doctors/carers, with (in general) an increase in the patient-carer physical distance and corresponding changes in the modes of access to relevant care information by all groups. The objective of this paper is to showcase the research efforts of six projects (that the authors are currently, or have recently been, involved in), CAALYX, eCAALYX, COGKNOW, EasyLine+, I2HOME, and SHARE-it, all funded by the European Commission towards a future where citizens can take an active role into managing their own healthcare. Most importantly, sensitive groups of citizens, such as the elderly, chronically ill and those suffering from various physical and cognitive disabilities, will be able to maintain vital and feature-rich connections with their families, friends and healthcare providers, who can then respond to, and prevent, the development of adverse health conditions in those they care for in a timely manner, wherever the carers and the people cared for happen to be.

(1) Background: Predictive modeling is becoming increasingly relevant in healthcare, aiding in clinical decision making and improving patient outcomes. However, many of the most potent predictive models, such as deep learning algorithms, are inherently opaque, and their decisions are challenging to interpret. This study addresses this challenge by employing Shapley Additive Explanations (SHAP) to facilitate model interpretability while maintaining prediction accuracy. (2) Methods: We utilized Gradient Boosting Machines (GBMs) to predict patient outcomes in an emergency department setting, with a focus on model transparency to ensure actionable insights. (3) Results: Our analysis identifies “Acuity”, “Hours”, and “Age” as critical predictive features. We provide a detailed exploration of their intricate interactions and effects on the model’s predictions. The SHAP summary plots highlight that “Acuity” has the highest impact on predictions, followed by “Hours” and “Age”. Dependence plots further reveal that higher acuity levels and longer hours are associated with poorer patient outcomes, while age shows a non-linear relationship with outcomes. Additionally, SHAP interaction values uncover that the interaction between “Acuity” and “Hours” significantly influences predictions. (4) Conclusions: We employed force plots for individual-level interpretation, aligning with the current shift toward personalized medicine. This research highlights the potential of combining machine learning’s predictive power with interpretability, providing a promising route concerning a data-driven, evidence-based healthcare future.

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.

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.

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.

Ambient-assisted living (AAL) is currently one of the important research and development areas, where accessibility, usability, and learning play a major role and where future interfaces are an important concern for applied engineering. The general goal of AAL solutions is to apply ambient intelligence technology to enable people with specific demands, e.g., handicapped or elderly, to live in their preferred environment longer. The term “Internet of Things” (IoT) is used as an umbrella keyword for covering various aspects related to the extension of the Internet and the Web into the physical realm, by means of the widespread deployment of spatially distributed devices with embedded identification, sensing and/or actuation capabilities. In this context, we propose a new methodology based on self organizing maps (SOMs) and fuzzy C-means (FCM) algorithms for profile generation as regards the activities of the user and their correlation with the available sensors. Moreover, we utilize the provided context to assign the generated profiles to more contextually complex activities. Our methodology is being evaluated into an AAL structure equipped with several sensors. More precisely, we assess the proposed method in a data set generated by accelerometers and its performance over a number of everyday activities.