Search Results (2,299)

Stroke continues to impose an enormous morbidity and mortality burden worldwide. Stroke survivors often incur debilitating consequences that impair motor function, independence in activities of daily living and quality of life. Rehabilitation is a pivotal intervention to minimize disability and promote functional recovery following a stroke. The Internet of Medical Things, a network of connected medical devices, software and health systems that collect, store and analyze health data over the internet, is an emerging resource in neurorehabilitation for stroke survivors. Technologies such as asynchronous transmission to handle intermittent connectivity, edge computing to conserve bandwidth and lengthen device life, functional interoperability across platforms, security mechanisms scalable to resource constraints, and hybrid architectures that combine local processing with cloud synchronization help bridge the digital divide and infrastructure limitations in low-resource environments. This manuscript reviews emerging rehabilitation technologies such as robotic devices, virtual reality, brain–computer interfaces and telerehabilitation in the setting of neurorehabilitation for stroke patients. Full article

Epilepsy remains an active and important area of research due to its complex etiology, significant global burden, and variable response to treatment. Current knowledge has provided valuable insights into the underlying molecular mechanisms of the disease and continues to guide the development of novel therapeutic strategies. This review presents a comprehensive overview of the etiologies of epilepsy, as well as traditional and modern medical and surgical treatment approaches, while highlighting future research directions. Peer-reviewed articles retrieved from PubMed and Google Scholar were analyzed and synthesized to produce this review. The etiological complexity of epilepsy arises from genetic, metabolic, structural, and inflammatory mechanisms, which often coexist rather than act independently. A wide range of anti-seizure drugs (ASDs) is currently available, with many new agents targeting novel mechanisms under development. Surgical approaches, including resection, disconnection, corpus callosotomy, and neuromodulation, are widely used for patients with drug-resistant epilepsy and result in variable seizure outcomes. In addition, minimally invasive techniques such as laser interstitial thermal therapy (LITT), stereoelectroencephalography-guided radiofrequency thermocoagulation, gamma knife radiosurgery, and high-intensity focused ultrasound have gained clinical relevance and continue to be explored. Emerging technologies, including artificial intelligence, machine learning, and precision medicine, offer promising directions for future research. Although several potential biomarkers have been identified, none are yet established for routine clinical use. Continued investigation is essential to improve understanding of epileptogenesis and to develop safer, more effective therapies. Full article

Rheumatoid arthritis-associated interstitial lung disease (RA-ILD) is one of the most common extra-articular manifestations of rheumatoid arthritis (RA) and a leading cause of mortality in RA patients. The diverse and nonspecific clinical presentations of RA-ILD make early diagnosis particularly challenging. In recent years, with a deeper understanding of the pathogenesis of RA-ILD and rapid advancements in medical imaging, artificial intelligence (AI) technologies, and biomarker research, notable progress has been achieved in the diagnostic approaches for RA-ILD. This review summarizes the latest research developments in the diagnosis of RA-ILD, with a focus on the clinical practice guidelines released in 2025. It discusses the application of high-resolution computed tomography (HRCT), the potential of AI in assisting HRCT-based diagnosis, and the discovery and validation of biomarkers. Furthermore, the review addresses current diagnostic challenges and explores future directions, providing clinicians and researchers with a cutting-edge perspective on RA-ILD diagnosis. Full article

Evaluating artificial intelligence (AI) models in clinical medicine requires more than conventional metrics such as accuracy, Area Under the Receiver Operating Characteristic (AUROC), or F1-score, which often overlook key considerations such as fairness, reliability, and real-world utility. We introduce CUES as a multiplicative composite score for clinical prediction models; it is defined as $\mathrm{CUES}=(C\cdot U\cdot E\cdot S{)}^{1/4}$, where C represents calibration, U integrated clinical utility, E equity across patient subpopulations, and S sampling stability. We formally establish boundedness, monotonicity, and differentiability on the domain ${\left(\mathrm{0,1}\right]}^4$, derive first-order sensitivity relations, and provide asymptotic approximations for its sampling distribution via the delta method. To facilitate inference, we propose bootstrap procedures for constructing confidence intervals and for comparative model evaluation. Analytic examples illustrate how CUES can diverge from traditional metrics, capturing dimensions of predictive performance that are essential for clinical reliability but often missed by AUROC or F1-score alone. By integrating multiple facets of clinical utility and robustness, CUES provides a comprehensive tool for model evaluation, comparison, and selection in real-world medical applications. Full article

Background: Artificial intelligence (AI) is transforming medical practice, enhancing diagnostic accuracy, personalisation, and clinical efficiency. However, this transition raises complex ethical challenges related to transparency, accountability, fairness, and human oversight. This study examines how the literature conceptualises and distributes ethical responsibility in AI-assisted healthcare. Methods: This semi-systematic, theory-informed thematic review was conducted in accordance with the PRISMA 2020 guidelines. Publications from 2020 to 2025 were retrieved from PubMed, ScienceDirect, IEEE Xplore databases, and MDPI journals. A semi-quantitative keyword-based scoring model was applied to titles and abstracts to determine their relevance. High-relevance studies (n = 187) were analysed using an eight-category ethical framework: transparency and explainability, regulatory challenges, accountability, justice and equity, patient autonomy, beneficence–non-maleficence, data privacy, and the impact on the medical profession. Results: The analysis revealed a fragmented ethical landscape in which technological innovation frequently outperforms regulatory harmonisation and shared accountability structures. Transparency and explainability were the dominant concerns (34.8%). Significant gaps in organisational responsibility, equitable data practices, patient autonomy, and professional redefinition were reported. A multilevel ethical responsibility model was developed, integrating micro (clinical), meso (institutional), and macro (regulatory) dimensions, articulated through both ex ante and ex post perspectives. Conclusions: AI requires governance frameworks that integrate ethical principles, regulatory alignment, and epistemic justice in medicine. This review proposes a multidimensional model that bridges normative ethics and operational governance. Future research should explore empirical, longitudinal, and interdisciplinary approaches to assess the real impact of AI on clinical practice, equity, and trust. Full article

Over recent years, venous ulcer wound care has experienced significant advancements through the application of machine learning (ML) models. The aim of the present study is a systematic, comprehensive analysis of prior research studies in this field covering the period between 2001 and August 2025. By searching multiple academic databases, including the Web of Science, Scopus, and PubMed, using relevant keywords and different queries, and screening reference lists of previously published manuscripts and review papers with a focus on the application of artificial intelligence in dermatology and medicine, an initial set of potential studies for review was obtained. To ensure the scope and relevance of the review, several inclusion and exclusion criteria were used to derive the final set of relevant research studies upon which a database for research data management was created. As a result, a total of 79 relevant research studies were comprehensively analysed, upon which detailed meta-analysis and analysis of application areas of ML models within venous ulcer wound care were conducted. Afterwards, a summary of benefits for medical systems and patients was given along with a general discussion regarding ML model limitations, trends, and opportunities, as well as research studies’ limitations and possible future research directions. The presented analyses may be valuable for researchers interested in applying ML models not only to venous ulcer wound care but also to other types of chronic wound care. Full article

Artificial intelligence (AI) has revolutionized medical image analysis. Several neural network (NN) architectures were developed and applied across the last decade, becoming essential for automated diagnosis and clinical applications. AI based on NNs has become increasingly integrated into gastroenterology, offering new opportunities for automated lesion detection and workflow optimization. Small-bowel capsule endoscopy (SBCE) has benefited substantially from these advances, addressing long-standing challenges such as time-consuming video review and variability among readers. This systematic review and meta-analysis evaluated neural network-based models for lesion detection in SBCE, assessing pooled diagnostic accuracy and the impact of AI on reading time. A total of 44 primary studies were included: 36 validation studies for accuracy and 9 clinical studies for reading time. All NN architectures demonstrated high diagnostic performance, with a pooled accuracy of 95.3% (95% CI: 94.1–96.5%). More recent architectures, including transformer-based and capsule networks, outperformed classical convolutional neural networks (CNNs). AI assistance significantly reduced SBCE reading time, with a pooled mean reduction of 84% compared to standard review. These findings highlight the strong potential of AI to enhance SBCE efficiency and diagnostic reliability. Full article

Microwave sensing technology is rapidly advancing and increasingly finding its way into biomedical applications, promising significant improvements for medical care. Concurrently, the rise of artificial intelligence (AI) is enabling significant enhancements in the biomedical domain. Close scrutiny of the recent literature reveals intense activity in both fields, with particularly impactful outcomes deriving from the combined use of advanced microwave techniques and AI for biomedical monitoring. In this review, an up-to-date compilation, from the perspective of the authors, of the most significant works published on these topics in recent years is given, focusing on their integration and current challenges. With the objective of analyzing the current landscape, we survey and compare state-of-the-art biosensors and imaging systems at all healthcare levels, from outpatient contexts to specialized medical equipment and laboratory analysis tools. We also delve into the relevant applications of AI in medicine for processing microwave-derived data. As our core focus, we analyze the synergistic integration of AI in the design of microwave devices and the processing of the acquired data, which have shown notable performances, opening new avenues for compact, affordable, and multi-functional medical devices. We conclude by synthesizing the prevailing technical, algorithmic, and translational challenges that must be addressed to realize this potential. Full article

Artificial Intelligence (AI) is a computer science that focuses on developing systems and machines capable of performing tasks that typically require human cognitive abilities. It has widespread applications in medical diagnostics. Its use has led to rapid advancements in diagnostic methodology, enabling the analysis of large datasets. The major applications of AI in medical diagnostics include personalized treatment based on patient genetics, preventive measures, and medical image analysis. AI is employed to analyse genomic data and biomarkers, aiding in the precise tailoring of therapies to individual patient needs. It could also be employed in modern dentistry in the near future, helping to achieve higher efficiency and accuracy in diagnosis and treatment planning. AI may be utilized in screening for oral mucosa lesions and to discriminate between oral potentially malignant disorders and cancers from benign lesions. The potential advantages of AI include high speed and accuracy in the diagnostic process, as well as relatively low costs. The aim of this review was to present the potential applications of AI methods in the diagnosis of selected mucocutaneous diseases. A literature review focuses on oral lichen planus, recurrent aphthous stomatitis, and oral and laryngeal leukoplakia. Full article

Objectives: The purpose of the current research is to examine and analyze the attitudes of healthcare service users towards the integration of remote medical services into ophthalmology in Bulgaria, including teleglaucoma. Methods: A cross-sectional survey study was conducted among 902 healthcare users during the period from May 2023 until December 2024. Descriptive statistics, parametric, and non-parametric tests for hypothesis testing were used. Results: The present study outlined predominantly positive attitudes towards the use of telemedicine services in ophthalmology, with 69.6% of respondents reporting a positive overall opinion in the final assessment. The greatest support was observed during remote consultations with a familiar doctor (77.4%) and during continuous follow-up of eye conditions (55.2%). Willingness to use such services was lower in emergencies or when contacting an unfamiliar specialist. A significant correlation was established between socio-demographic characteristics and attitudes—respondents with greater education levels (p = 0.006), men, and younger participants were more positive towards telemedicine (p < 0.05). The high level of awareness about glaucoma, particularly among those with university-level education, served as a positive prerequisite for the implementation of teleophthalmology services related to its monitoring. Mobile applications and digital solutions were evaluated as beneficial means of facilitating communication and increasing adherence to treatment. Regarding the use of artificial intelligence, certain skepticism and insufficient awareness levels were observed, which required additional efforts to increase trust and digital literacy among users. Conclusions: The implementation of telemedicine services into ophthalmology has potential but outlines the necessity of considering the individual attitudes of applying coherent quality and safety standards and of directed awareness campaigns, especially towards the groups of lower technological and healthcare literacy. Full article

Background: Ward patients who experience clinical deterioration are at high risk of mortality. Conventional rapid response systems (RRS) using track-and-trigger protocols have not consistently demonstrated improved outcomes. This study evaluated the impact of an artificial intelligence (AI)-based cardiac arrest prediction model. Methods: This 1-year, prospective, non-randomized interventional trial assigned hospitalized patients with AI-based software as a medical device (AI-SaMD) high-risk alerts to groups based on their subsequent clinical response; those reassessed or treated within 24 h comprised the AI-SaMD-guided cohort, while the remainder formed the usual care cohort. Alerts prompted an optional but not mandatory treatment review. The primary outcome was ward-based cardiac arrest; the secondary outcome was in-hospital mortality. Multivariable regression analysis was used to adjust for potential confounders. Results: Of 35,627 general ward admissions, 2906 triggered an AI-SaMD alert. Among these, 1409 (48.4%) were allocated to the AI-SaMD-guided cohort. The incidence of cardiac arrest significantly decreased from 2.07% to 1.06% (adjusted risk ratio (RR), 0.54; 95% confidence interval (CI), 0.20–0.88; p < 0.01). In-hospital mortality also significantly declined (adjusted RR, 0.65; 95% CI, 0.32–0.98; p < 0.05). Conclusions: AI-SaMD-guided alerts were associated with reductions in cardiac arrest and in-hospital mortality without requiring additional resources, supporting their integration into current clinical workflows to improve patient safety and optimize RRS performance. Full article

Background/Objectives: Vast amounts of textual data are generated by healthcare organizations every year. Traditional content analysis is time-intensive, error-prone, and potentially biased. This study demonstrates how freely available large language model (LLM) artificial intelligence (AI) tools can efficiently and effectively analyze qualitative healthcare data and uncover insights missed by traditional manual analysis. Interview data from chief nursing officers (CNOs) at top-performing academic medical centers were analyzed to identify factors contributing to their operational and patient quality success. Methods: Semi-structured interviews were conducted with CNOs from top-performing academic medical centers that achieved top-decile quality measures while using resources most efficiently. Interview transcripts were analyzed using a mix of traditional text mining in LSA and Gemini 2.5. The capability of four freely available AI platforms—Gemini 2.5, Scholar AI 5.1, Copilot’s Chat, and Claude’s Sonnet 4.5—was also reviewed. Results: LLM AI analysis identified ten primary factors, comprising twenty-four subtopics, that characterized successful hospital performance. Notably, AI analysis identified a theoretical connection that manual analysis had missed, revealing how the identified framework aligned with Donabedian’s seminal structure, process, outcomes quality model. The AI analysis reduced the required time from weeks to nearly instantaneous. Conclusions: LLM AI tools offer a transformative approach to unlocking insight from the analysis of qualitative textual data in healthcare settings. These tools can provide rapid insight that is accessible to personnel with minimal text-mining expertise and offer a practical solution for healthcare organizations to unlock insight hidden in the vast amounts of textual data they hold. Full article

This review explores the transformative role of three-dimensional (3D) printing in radiation therapy for cancer treatment, emphasizing its potential to deliver patient-specific, cost-effective, and sustainable medical devices. The integration of 3D printing enables rapid fabrication of customized boluses, compensators, immobilization devices, and GRID collimators tailored to individual anatomical and clinical requirements. Comparative analysis reveals that additive manufacturing surpasses conventional machining in design flexibility, lead time reduction, and material efficiency, while offering significant cost savings and recyclability benefits. Case studies demonstrate that 3D-printed GRID collimators achieve comparable dosimetric performance to traditional devices, with peak-to-valley dose ratios optimized for spatially fractionated radiation therapy. Furthermore, emerging applications of artificial intelligence (AI) in conjunction with 3D printing promise automated treatment planning, generative device design, and real-time quality assurance, and are paving the way for adaptive and intelligent radiotherapy solutions. Regulatory considerations, including FDA guidelines for additive manufacturing, are discussed to ensure compliance and patient safety. Despite challenges such as material variability, workflow standardization, and large-scale clinical validation, evidence indicates that 3D printing significantly enhances therapeutic precision, reduces toxicity, and improves patient outcomes. This review underscores the synergy between 3D printing and AI-driven innovations as a cornerstone for next-generation radiation oncology, offering a roadmap for clinical adoption and future research. Full article

Brain tumors, encompassing subtypes with distinct progression and risk profiles, are a serious public health concern. Magnetic resonance imaging (MRI) is the primary imaging modality for non-invasive assessment, providing the contrast and detail necessary for diagnosis, subtype classification, and individualized care planning. In this paper, we evaluate the capability of modern deep learning models to classify gliomas as high-grade (HGG) or low-grade (LGG) using reduced training data from MRI scans. Utilizing the BraTS 2019 best-slice dataset (2185 images in two classes, HGG and LGG) divided in two folders, training and testing, with different images obtained from different patients, we created subsets including 10%, 25%, 50%, 75%, and 100% of the dataset. Six deep learning architectures, DeiT3_base_patch16_224, Inception_v4, Xception41, ConvNextV2_tiny, swin_tiny_patch4_window7_224, and EfficientNet_B0, were evaluated utilizing three-fold cross-validation (k = 3) and increasingly large training datasets. Explainability was assessed using Grad-CAM. With 25% of the training data, DeiT3_base_patch16_224 achieved an accuracy of 99.401% and an F1-Score of 99.403%. Under the same conditions, Inception_v4 achieved an accuracy of 99.212% and a F1-Score of 99.222%. Considering how the models performed across both data subsets and their compute demands, Inception_v4 struck the best balance for MRI-based glioma classification. Both convolutional networks and vision transformers achieved superior discrimination between HGGs and LGGs, even under data-limited conditions. Architectural disparities became increasingly apparent as training data diminished, highlighting unique inductive biases and efficiency characteristics. Even with a relatively limited amount of training data, current deep learning (DL) methods can achieve reliable performance in classifying gliomas from MRI scans. Among the architectures evaluated, Inception_v4 offered the most consistent balance between accuracy, F1-Score, and computational cost, making it a strong candidate for integration into MRI-based clinical workflows. Full article

Understanding the decision-making behavior of machine learning models is essential in domains where individual predictions matter, such as medical diagnosis or sports analytics. While explainable artificial intelligence (XAI) methods such as SHAP provide instance-level feature attributions, they mainly summarize typical decision behavior and offer limited support for systematically exploring atypical yet correctly classified cases. In this work, we introduce the Classification Outlier Variability Score (COVAS), a framework designed to support hypothesis generation through the analysis of explanation variability. COVAS operates in the explanation space and builds directly on SHAP value representations. It quantifies how strongly an individual instance’s SHAP-based explanation deviates from class-specific attribution patterns by aggregating standardized SHAP deviations into a single score. Consequently, the applicability of COVAS inherits the model- and data-agnostic properties of SHAP, provided that explanations can be computed for the underlying model and data. We evaluate COVAS on publicly available datasets from the medical and sports domains. The results show that COVAS reveals explanation-space outliers not captured by feature-space outlier detection or prediction uncertainty measures. Robustness analyses demonstrate stability across parameter choices, class imbalance, model initialization, and model classes. Overall, COVAS complements existing XAI techniques by enabling targeted instance-level inspection and facilitating XAI-guided hypothesis formulation. Full article