Search Results (7,807)

Vibration-based structural health monitoring methods have been widely applied in the field of damage identification. This paper proposes an intelligent diagnostic approach that integrates a multi-scale convolutional neural network with a multi-head attention mechanism (MSCNN-MHA) for the structural health monitoring of jacket-type offshore platforms. Through numerical simulations, acceleration response signals of three-pile and four-pile jacket platforms under random wave excitation are analyzed. Damage localization studies are conducted under simulated crack and pitting corrosion cases. Unlike previous studies that often idealize damage by weakening structural parameters or removing components, this study focuses on small-scale damage forms to better reflect real engineering conditions. To verify the noise resistance of the proposed method, noise is added to the original signals for further testing. Finally, experiments are conducted on the basic structure of the jacket-type offshore platform, simulating small-scale crack and pitting damage under sinusoidal and pulse excitation, to further evaluate the applicability of the method. Compared to previous CNN and MSCNN-based approaches, the results of this study demonstrate that the MSCNN-MHA method achieves higher accuracy in identifying and locating minor damage in jacket-type offshore platforms. Full article

Background: Perinatal research faces significant challenges in understanding placental biology and maternal–fetal interactions due to limited access to human tissues and the lack of reliable models. Emerging technologies, such as liquid biopsy and single-cell analysis, offer novel, non-invasive approaches to investigate these processes. This scoping review explores the current applications of these technologies in placental development and the diagnosis of pregnancy complications, identifying research gaps and providing recommendations for future studies. Methods: This review adhered to PRISMA-ScR guidelines. Studies were selected based on their focus on liquid biopsy or single-cell analysis in perinatal research, particularly related to placental development and pregnancy complications such as preeclampsia, preterm birth, and fetal growth restriction. A systematic search was conducted in PubMed, Scopus, and Web of Science for studies published in the last ten years. Data extraction and thematic synthesis were performed to identify diagnostic applications, monitoring strategies, and biomarker identification. Results: Twelve studies were included, highlighting the transformative potential of liquid biopsy and single-cell analysis in perinatal research. Liquid biopsy technologies, such as cfDNA and cfRNA analysis, provided non-invasive methods for real-time monitoring of placental function and early identification of complications. Extracellular vesicles (EVs) emerged as biomarkers for conditions like preeclampsia. Single-cell RNA sequencing (scRNA-seq) revealed cellular diversity and pathways critical to placental health, offering insights into processes such as vascular remodeling and trophoblast invasion. While promising, challenges such as high costs, technical complexity, and the need for standardization limit their clinical integration. Conclusion: Liquid biopsy and single-cell analysis are revolutionizing perinatal research, offering non-invasive tools to understand and manage complications like preeclampsia. Overcoming challenges in accessibility and standardization will be key to unlocking their potential for personalized care, enabling better outcomes for mothers and children worldwide. Full article

The increasing demand for high-throughput, real-time, and single-molecule protein analysis in precision medicine has propelled the development of novel sensing technologies. Among these, nanopore-based methods have garnered significant attention for their unique capabilities, including label-free detection, ultra-sensitivity, and the potential for miniaturization and portability. Originally designed for nucleic acid sequencing, nanopore technology is now being adapted for peptide and protein analysis, offering promising applications in biomarker discovery and disease diagnostics. This review examines the latest advances in biological, solid-state, and hybrid nanopores for protein sensing, focusing on their ability to detect amino acid sequences, structural variants, post-translational modifications, and dynamic protein–protein or protein–drug interactions. We critically compare these systems to conventional proteomic techniques, such as mass spectrometry and immunoassays, discussing advantages and persistent technical challenges, including translocation control and signal deconvolution. Particular emphasis is placed on recent advances in protein sequencing using biological and solid-state nanopores and the integration of machine learning and signal-processing algorithms that enhance the resolution and accuracy of protein identification. Nanopore protein sensing represents a disruptive innovation in biosensing, with the potential to revolutionize clinical diagnostics, therapeutic monitoring, and personalized healthcare. Full article

Background/Objectives: While the integration of artificial intelligence (AI) into clinical research is rapidly accelerating, a comprehensive analysis of the global AI clinical trial landscape has been limited. This study presents the first systematic characterization of AI-related clinical trials registered in the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP). It aims to map global trends, identify patterns of concentration, and analyze the structure of international collaboration. Methods: A search of the WHO ICTRP was conducted on 20 June 2025. Following a two-stage screening process, the dataset was analyzed for temporal trends, geographic distribution, disease and technology categories, and international collaboration patterns using descriptive statistics and network analysis. Results: We identified 596 AI clinical trials across 62 countries, with registrations growing exponentially since 2020. The landscape is defined by extreme geographic concentration, with China accounting for the largest share of trial participations (35.6%), followed by the USA (8.5%). Research is thematically concentrated in Gastroenterology (22.8%) and Oncology (20.1%), with Diagnostic Support (45.6%) being the most common technology application. Formal international collaboration is critically low, with only 8.7% of trials involving multiple countries, revealing a fragmented collaboration landscape. Conclusions: The global AI clinical trial landscape is characterized by rapid but deeply imbalanced growth. This concentration and minimal international collaboration undermine global health equity and the generalizability of AI technologies. Our findings underscore the urgent need for a fundamental shift toward more inclusive, transparent, and collaborative research models to ensure the benefits of AI are realized equitably for all of humanity. Full article

Background/Objectives: The extension of euthanasia and physician-assisted suicide to individuals with mental disorders presents a profound ethical, clinical, and legal challenge. While increasingly accepted in some jurisdictions, their application in psychiatric contexts—particularly in cases of depression—raises concerns about diagnostic precision, therapeutic adequacy, and the validity of informed consent. This study examines two controversial Belgian cases to explore the complexities of euthanasia for psychological suffering. Methods: A qualitative case analysis was conducted through a qualitative analysis of publicly available media sources. The cases were examined through clinical, psychoanalytic, and medico-legal lenses to assess diagnostic clarity, treatment history, and ethical considerations. No access to official medical records was available. Case Presentation: The first case involved a young woman whose depressive symptoms were reportedly linked to trauma from a terrorist attack. The second concerned a middle-aged woman convicted of infanticide and later diagnosed with Major Depression. Discussion: In both cases, euthanasia was granted on the grounds of “irreversible psychological suffering.” However, the absence of detailed clinical documentation, potential unresolved trauma, and lack of psychodynamic assessment raised doubts about the robustness of the evaluations and the validity of informed consent. Conclusions: These findings highlight the need for a more rigorous, multidisciplinary, and ethically grounded approach to psychiatric euthanasia. This study underscores the importance of precise diagnostic criteria, comprehensive treatment histories, and deeper exploration of unconscious and existential motivations. Safeguarding clinical integrity and ethical standards is essential in end-of-life decisions involving mental illness. Full article

Cancer treatment faces significant challenges due to drug resistance, non-specific toxicity, and limited penetration of therapeutic agents. Here, we discuss the latest advancements in the design and application of tumor-targeted nanoparticles, focusing on polymer-based, biomimetic, and inorganic nanocarriers, as well as innovative surface modification strategies, to enhance diagnostic and therapeutic approaches in cancer treatment, including the co-delivery of chemotherapeutic agents with biologicals or photo/sonosensitizers for synergistic therapeutic effects. This review not only highlights the current importance of nanoparticle design and application for tumor targeting but also provides insights into future directions for more effective cancer therapies. By integrating advanced material science with biology, these strategies hold the potential to transform the landscape of cancer treatment, offering hope for improved patient outcomes and personalized therapeutic approaches. Full article

Objectives: Fluorescein break-up time (FBUT) is commonly used to assess tear film stability. However, the instillation of fluorescein destabilises the tear film, impacting validity and clinical applicability, while the subjective nature and variation in volume and concentration reduces repeatability. Non-invasive break-up time (NIBUT) offers an alternative method with less potential bias. Normal tear break-up time is conventionally accepted as 10 seconds (s); however, FBUT is expected to be lower than NIBUT. This study was designed to compare FBUT and NIBUT values in a pre-operative refractive surgery population, where diagnosis of dry eye disease may alter the risk–benefits ratio and contraindicate surgical procedure(s). Improved understanding of the relationship between these two methods will aid appropriate pre-operative patient counselling and consent. Methods: Data from consecutive participants presenting to a private ophthalmology clinic, for initial refractive surgery pre-operative assessment, were analysed. NIBUT and FBUT were performed. Paired and unpaired comparisons were made using the Wilcoxon signed-rank and Mann–Whitney U tests, respectively, and relationships with demographics were explored using Spearman’s rank correlation coefficient. Results: Median and interquartile range (IQR) for the first NIBUT was 12.5 s (7.0–18.0 s) and 14.2 s (9.4–18.0 s) for the right and left eyes, respectively. Median and IQR for the average NIBUT was 14.0 s (6.9–18.0 s) and 14.6 s (10.1–18.0 s) for the right and left eyes, respectively. Median and IQR for FBUT was 7 s (5–8 s) and 6 s (5–8 s) for the right and left eyes, respectively. There was a statistically significant difference between NIBUT and FBUT (p < 0.001). Conclusions: The findings suggest that the commonly used diagnostic threshold of 10 s cannot be uniformly applied to both FBUT and NIBUT, as FBUT systematically underestimates tear stability. Full article

Depression remains a leading cause of global disability, yet scalable and objective diagnostic tools are still lacking. Speech has emerged as a promising non-invasive modality for automated depression detection, due to its strong correlation with emotional state and ease of acquisition. While convolutional neural networks (CNNs) have achieved state-of-the-art performance in this domain, their high computational demands limit deployment in low-resource or real-time settings. Spiking neural networks (SNNs), by contrast, offer energy-efficient, event-driven computation inspired by biological neurons, but they are difficult to train directly and often exhibit degraded performance on complex tasks. This study investigates whether CNNs trained on audio data from the clinically annotated DAIC-WOZ dataset can be effectively converted into SNNs while preserving diagnostic accuracy. We evaluate multiple conversion thresholds using the SpikingJelly framework and find that the 99.9% mode yields an SNN that matches the original CNN in both accuracy (82.5%) and macro F1 score (0.8254). Lower threshold settings offer increased sensitivity to depressive speech at the cost of overall accuracy, while naïve conversion strategies result in significant performance loss. These findings support the feasibility of CNN-to-SNN conversion for real-world mental health applications and underscore the importance of precise calibration in achieving clinically meaningful results. Full article

Background: Melanoma is one of the most aggressive types of skin cancer. Its diagnosis appears to be challenging due to morphological similarities to benign melanocytic lesions. Even though histopathological evaluation is the diagnostic gold standard, immunohistochemistry (IHC) proves to be useful in challenging cases. Preferentially Expressed Antigen in Melanoma (PRAME) has emerged as a promising diagnostic, prognostic, and therapeutic marker in melanoma. Methods: This review critically examines the role of PRAME across clinical domains. It presents an evaluation of PRAME’s diagnostic utility in differentiating melanomas from benign nevi, its prognostic significance across melanoma subtypes, and therapeutic applications in emerging immunotherapy strategies. An extensive analysis of the current literature was conducted, with a focus on PRAME expression patterns in melanocytic lesions and various malignancies, along with its integration into IHC protocols and investigational therapies. Results: PRAME demonstrates high specificity and sensitivity in distinguishing melanoma from benign melanocytic proliferations, particularly in challenging subtypes such as acral, mucosal, and spitzoid lesions. Its overexpression correlates with poor prognosis in numerous malignancies. Therapeutically, PRAME’s HLA class I presentation enables T-cell-based targeting. Early-phase trials show promising results using PRAME-directed TCR therapies and bispecific ImmTAC agents. However, immune evasion mechanisms (i.e., heterogeneous antigen expression, immune suppression in the tumor microenvironment, and HLA downregulation) pose significant challenges to therapy. Conclusions: PRAME is a valuable biomarker for melanoma diagnosis and a promising target for immunotherapy. Its selective expression in malignancies supports its clinical utility in diagnostic precision, prognostic assessment, and precision oncology. Ongoing research aimed at overcoming immunological barriers will be essential for optimizing PRAME-directed therapies and establishing their place in the personalized management of melanoma. Full article

Metal–organic frameworks (MOFs) have emerged as highly versatile materials for the development of next-generation optical biosensors owing to their tunable porosity, large surface area, and customizable chemical functionality. Recently, MOF-based platforms have shown substantial potential in various optical transduction modalities, including fluorescence, luminescence, and colorimetric sensing, enabling the highly sensitive and selective detection of biological analytes. This review provides a comprehensive overview of recent advancements in MOF-based optical biosensors, focusing on their applications in pathogen detection and environmental monitoring. We highlight key design strategies, including MOF functionalization, hybridization with nanoparticles or dyes, and integration into microfluidic and wearable devices. Emerging methods, such as point-of-care diagnostics, label-free detection, and real-time monitoring, are also discussed. Finally, the current challenges and future directions for the practical deployment of MOF-based optical biosensors in clinical and field environments are discussed. Full article

The COVID-19 pandemic, caused by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), catalyzed unprecedented innovation in molecular diagnostics to address critical gaps in rapid pathogen detection. Over the past five years, CRISPR-based systems, isothermal amplification techniques, and portable biosensors have emerged as transformative tools for nucleic acid detection, offering improvements in speed, sensitivity, and point-of-care applicability compared to conventional PCR. While numerous reviews have cataloged the technical specifications of these platforms, a critical gap remains in understanding the strategic and economic hurdles to their real-world implementation. This review provides a forward-looking analysis of the feasibility, scalability, and economic benefits of integrating these next-generation technologies into future pandemic-response pipelines. We synthesize advances in coronavirus-specific diagnostic platforms and attempt to highlight the need for their implementation as a cost-saving measure during surges in clinical demand. We evaluate the feasibility of translating these technologies—particularly CRISPR-Cas integration with recombinase polymerase amplification (RPA)—into robust first-line diagnostic pipelines for novel viral threats. By analyzing the evolution of diagnostic strategies during the COVID-19 era, we aim to provide strategic insights and new directions for developing and deploying effective detection platforms to better confront future viral pandemics. Full article

Background: Keratinocyte carcinomas (KCs), including basal cell carcinomas (BCCs) and squamous cell carcinomas (SCCs), are the most prevalent malignancies globally, particularly affecting sun-exposed facial areas. Achieving clear surgical margins in these regions is essential to ensure oncologic control while preserving cosmetic outcomes. Reflectance confocal microscopy (RCM) is a noninvasive imaging technique that enables real-time, high-resolution visualization of skin structures and may aid in margin assessment during KC surgery. This systematic review aims to evaluate the role of in vivo RCM in the surgical treatment of KCs. Methods: This review followed PRISMA guidelines. A comprehensive search of PubMed, Scopus, Web of Science, Medline, and EBSCO databases was conducted for studies published between January 1992 and December 2024. Inclusion criteria focused on clinical studies utilizing in vivo RCM for diagnostic or surgical applications in KC management. Results: Eighteen studies involving 1112 patients were included. RCM was used preoperatively in 5 studies and intraoperatively in another 5. Nine studies assessed margin delineation, while eight focused on diagnostic accuracy. RCM improved diagnostic confidence and allowed for more precise margin assessment, potentially reducing the extent of surgical excision in cosmetically sensitive areas. However, its broader clinical adoption is limited by operator dependency, procedural complexity, and lack of standardization. Conclusions: RCM shows promise as a supportive tool in KC surgery, particularly for preoperative planning. While its diagnostic utility is well established, its intraoperative role requires further validation. Larger, standardized, and cost-effective studies are needed to confirm its impact on surgical outcomes and patient quality of life. Full article

The integration of artificial intelligence (AI) in medical imaging has revolutionized diagnostic capabilities, yet the black-box nature of deep learning models poses significant challenges for clinical adoption. Current explainable AI (XAI) approaches, including SHAP, LIME, and Grad-CAM, predominantly focus on post hoc explanations that may inadvertently undermine clinical decision-making by providing misleading confidence in AI outputs. This paper presents a systematic review and meta-analysis of 67 studies (covering 23 radiology, 19 pathology, and 25 ophthalmology applications) evaluating XAI fidelity, stability, and performance trade-offs across medical imaging modalities. Our meta-analysis of 847 initially identified studies reveals that LIME achieves superior fidelity (0.81, 95% CI: 0.78–0.84) compared to SHAP (0.38, 95% CI: 0.35–0.41) and Grad-CAM (0.54, 95% CI: 0.51–0.57) across all modalities. Post hoc explanations demonstrated poor stability under noise perturbation, with SHAP showing 53% degradation in ophthalmology applications (ρ = 0.42 at 10% noise) compared to 11% in radiology (ρ = 0.89). We demonstrate a consistent 5–7% AUC performance penalty for interpretable models but identify modality-specific stability patterns suggesting that tailored XAI approaches are necessary. Based on these empirical findings, we propose a comprehensive three-pillar accountability framework that prioritizes transparency in model development, interpretability in architecture design, and a cautious deployment of post hoc explanations with explicit uncertainty quantification. This approach offers a pathway toward genuinely accountable AI systems that enhance rather than compromise clinical decision-making quality and patient safety. Full article

Recent advances in microbiome research have highlighted the oral cavity as a complex and dynamic ecosystem, home to over 700 microbial species that play critical roles in both oral and systemic health. The oral microbiota not only maintains local tissue homeostasis but also serves as a reservoir for antimicrobial resistance (AMR) genes, contributing to the global spread of resistance. Frequent and sometimes inappropriate antibiotic use in dental practice, along with exposure to antiseptics and biocides, drives the emergence and horizontal transfer of resistance determinants within oral biofilms. This review synthesizes current knowledge on the molecular mechanisms and ecological drivers of AMR in the oral microbiome, emphasizing the clinical implications of dysbiosis and drug-resistant infections. The authors advocate for the development of dental clinical guidelines tailored to the unique characteristics of the oral microbiota, focusing on personalized therapy through molecular diagnostics, standardized AMR risk assessment, and the integration of non-antibiotic strategies such as probiotics and photodynamic therapy. Continuous education in antimicrobial stewardship and the implementation of oral-specific AMR surveillance is also highlighted as an essential component of effective resistance management. To support rational prescribing, a dedicated mobile application has been developed, leveraging microbiota data and resistance profiles to guide evidence-based, targeted therapy and reduce unnecessary antibiotic use. Collectively, these strategies aim to preserve antibiotic efficacy, ensure patient safety, and promote sustainable infection management in the dental field. Full article

Large language models (LLMs) rapidly transform healthcare by automating tasks, streamlining administration, and enhancing clinical decision support. This rapid review assesses current and emerging applications of LLMs in diagnostic-related group (DRG) assignment and clinical decision support systems (CDSS), with emphasis on radiology and nuclear medicine. Evidence shows that LLMs, particularly those tailored for medical domains, improve efficiency and accuracy in DRG coding and radiology report generation, providing clinicians with actionable, context-sensitive insights by integrating diverse data sources. Advances like retrieval-augmented generation and multimodal architecture further increase reliability and minimize incorrect or misleading results that AI models generate, a term that is known as hallucination. Despite these benefits, challenges remain regarding safety, explainability, bias, and regulatory compliance, necessitating ongoing validation and oversight. The review prioritizes recent, peer-reviewed literature on radiology and nuclear medicine to provide a practical synthesis for clinicians, administrators, and researchers. While LLMs show strong promise for enhancing DRG assignment and radiological decision-making, their integration into clinical workflows requires careful management. Ongoing technological advances and emerging evidence may quickly change the landscape, so findings should be interpreted in context. This review offers a timely overview of the evolving role of LLMs while recognizing the need for continuous re-evaluation. Full article