Search Results (1,720)

Class imbalance in clinical electrocardiogram (ECG) datasets limits the diagnostic sensitivity of automated arrhythmia classifiers, particularly for rare but clinically significant beat types. We propose a three-stage hybrid generative pipeline that combines a spectral-guided conditional variational autoencoder (cVAE), a class-conditional latent denoising diffusion probabilistic model (DDPM), and a Quantum Latent Refinement (QLR) module built on parameterized quantum circuits, implemented and evaluated using a classical quantum-circuit simulator, to augment minority arrhythmia classes, and present results based on the MIT-BIH Arrhythmia Database. The QLR module applies a bounded residual correction guided by Maximum Mean Discrepancy minimization to align synthetic latent distributions with real class-specific latent banks. A lightweight 1D MobileNetV2 classifier evaluated over ten independent random seeds and four augmentation ratios serves as the downstream benchmark. Our findings establish latent diffusion augmentation as an effective strategy for imbalanced ECG classification. To our knowledge, the proposed QLR module is the first use of a parameterized quantum circuit as a distributional refiner within a generative augmentation pipeline. While its performance is comparable to that of the classical latent diffusion framework under the present experimental conditions, the proposed approach demonstrates the feasibility of integrating quantum latent operators into generative medical AI pipelines and provides a foundation for future investigations on quantum-enhanced representation learning and data augmentation. Full article

Purpose: Accurate assessment of skeletal maturation is important for timing pediatric dental and orthodontic interventions. This study compared bone age estimates generated by two AI-based programs, CRESCOM MediAI-BA and VUNO Med-BoneAge^®, in Korean children and adolescents undergoing orthodontic diagnostic evaluation. Methods: This retrospective study compared bone-age estimates generated by CRESCOM MediAI-BA and VUNO Med-BoneAge^® using left hand–wrist radiographs from 200 Korean children and adolescents (100 boys and 100 girls; 6–17 years) undergoing orthodontic diagnostic evaluation. Chronological age and program-derived bone age were compared by sex and age group using the Wilcoxon signed-rank test. Standardized differences were summarized using Cohen’s d, and Spearman correlations were calculated among chronological age, program-derived bone age, and the CRESCOM skeletal maturity indicator (SMI). Results: The standardized differences between chronological age and CRESCOM-predicted bone age were very small in boys (d = 0.143) and small in girls (d = 0.258). Corresponding differences for VUNO Med-BoneAge^® were small in boys (d = 0.254) and girls (d = 0.216). CRESCOM SMI correlated strongly with chronological age (ρ = 0.852), VUNO-predicted bone age (ρ = 0.910), and CRESCOM-predicted bone age (ρ = 0.932; all p < 0.001). However, statistically significant age- and sex-specific differences were observed between chronological age and AI-predicted bone age and between the two programs. Conclusions: CRESCOM MediAI-BA may provide clinically useful adjunctive information for skeletal-maturity assessment; nevertheless, AI-derived bone-age estimates from different programs should not be considered interchangeable and should be interpreted in conjunction with clinical findings and established skeletal-maturity indicators. Full article

Background: Maintenance hemodialysis is characterized by repetitive changes in fluid distribution, blood pressure, and organ perfusion. Conventional clinical examination, empirical dry-weight adjustment, and biomarkers do not fully resolve the compartment-specific nature of congestion in this population. This narrative review reframes ultrasound-based volume assessment as a multimodal diagnostic problem involving pulmonary congestion, intravascular filling, systemic venous congestion, cardiac reserve, tissue response, and perfusion vulnerability. Methods: We synthesized clinically relevant evidence indexed in PubMed and Google Scholar for studies published between January 2016 and April 2026, prioritizing dialysis-specific randomized trials, prospective cohorts, systematic reviews, consensus statements, and methodological studies related to diagnostic ultrasound, Doppler-based congestion assessment, contrast-enhanced ultrasound, elastography, artificial intelligence, point-of-care ultrasound, and remote ultrasound monitoring. Results: Lung ultrasound currently has the strongest dialysis-specific evidence for detecting and tracking pulmonary congestion. Inferior vena cava ultrasound provides adjunctive information on intravascular filling and right-sided pressure but is not a surrogate for total body water. Echocardiographic parameters help characterize filling pressure and cardiac tolerance to fluid removal, whereas venous Doppler and the Venous Excess Ultrasound Score provide an emerging approach to systemic venous congestion. Elastography and contrast-enhanced ultrasound remain investigational tools for tissue characterization and perfusion vulnerability, while AI-assisted analysis, handheld point-of-care ultrasound, and tele-ultrasound may improve standardization, automated B-line quantification, and scalability. Conclusions: Different ultrasound modalities answer different diagnostic questions in maintenance hemodialysis. A compartment-specific framework integrating congestion, perfusion, and cardiac-reserve domains may better support individualized ultrafiltration planning, hemodynamic risk assessment, and future outcome-oriented research. Full article

Artificial intelligence (AI) is increasingly being integrated into medical practice, offering promising tools to improve diagnostic accuracy and clinical efficiency. In the field of breast pathology, AI applications, particularly those based on deep learning (DL) and machine learning (ML), are emerging as decision-support tools in both diagnostic and prognostic workflows. This review provides a comprehensive overview of current AI-based approaches, with a focus on their clinical utility in tumor detection, histological classification, biomarker assessment, and prediction of treatment response. In addition to summarizing available AI platforms, the review critically examines their level of clinical validation, regulatory status, and integration into routine practice. Key challenges are also discussed. Overall, AI is expected to play an increasingly important role in supporting pathologists and advancing precision medicine in breast cancer management. Full article

Skin cancer represents a growing global health burden. Current diagnostic pathways combine clinical examination and dermoscopy with histopathological confirmation; however, overlap between benign and malignant lesions can create diagnostic uncertainty and lead to potentially avoidable biopsies. Electrical impedance spectroscopy (EIS) has emerged as a non-invasive technique with potential for portable and cost-efficient implementation that quantifies the dielectric contrast between malignant and healthy tissue, providing objective information that may support clinical decision-making. This review synthesizes the field across four levels. First, we describe the biophysical origins of the impedance contrast in skin cancer, spanning the cellular, tissue architecture, and molecular scales, together with the equivalent circuit and Cole–Cole frameworks used to interpret it. Second, we examine hardware advances, including electrode–skin interface strategies, flexible and wearable architectures, computational electrode design, and the translation from laboratory prototypes to commercial systems such as Nevisense. Third, we critically appraise clinical evidence from large multicenter trials, focusing on the sensitivity–specificity trade-off and the demonstrated reduction in the number needed to excise. Finally, we discuss emerging frontiers, including artificial intelligence-driven analysis and multimodal fusion with dermoscopy, reflectance confocal microscopy, optical coherence tomography, and near-infrared spectroscopy. We conclude that EIS is most valuable as a complementary component within an integrated, AI-supported multimodal diagnostic framework. Full article

Background and Objectives: Artificial intelligence (AI) is increasingly used in clinical dentistry, but its diagnostic accuracy for oral mucosal lesions based on clinical photographs remains insufficiently validated. This prospective observational study compared the Top-1 diagnostic accuracy of ChatGPT-4o and ChatGPT-5 in identifying normal anatomical variants and pathological oral mucosal lesions and evaluated their performance across anatomical sites. Materials and Methods: Seventy adults with either normal anatomical variants (n = 21) or pathological oral mucosal lesions (n = 49) were consecutively recruited at the Department of Dental Medicine, University Hospital of Split, Croatia. One standardized clinical photograph per patient was analyzed by ChatGPT-4o and ChatGPT-5 under image-only and image-plus-text conditions using identical prompts. The reference diagnosis was established by an oral medicine specialist, with histopathological examination (HPE) performed when clinically indicated. Diagnostic performance was assessed using Top-1 accuracy and McNemar’s test. Results: Both models showed low accuracy with image-only input, but performance improved significantly after clinical information was added (p < 0.001). Overall Top-1 accuracy increased from 19.0% to 69.0% for ChatGPT-4o and from 9.0% to 51.0% for ChatGPT-5. For normal anatomical variants, accuracy increased from 14.3% to 81.0% and from 14.3% to 76.2%, respectively. For pathological oral mucosal lesions, accuracy increased from 20.4% to 63.3% and from 6.1% to 40.8%, respectively. ChatGPT-4o showed numerically higher accuracy than ChatGPT-5, particularly for pathological oral mucosal lesions, but no statistically significant difference was found between the models in the corresponding paired comparisons. Conclusions: Diagnostic performance was limited with image-only input but improved substantially when standardized clinical information accompanied the images. The numerical differences between models, particularly for pathological oral mucosal lesions, may be clinically relevant but do not establish superiority or equivalence. Neither model can currently replace conventional clinical diagnosis, and AI should be regarded as a clinical decision-support tool for evaluating oral mucosal lesions. Full article

Osteonecrosis of the femoral head (ONFH) is a progressive and potentially disabling condition caused by compromised blood supply to the femoral head, leading to bone necrosis and collapse. Early diagnosis is essential to enable joint-preserving interventions and improve patient outcomes. Magnetic resonance imaging (MRI) is currently considered the most sensitive modality for early detection, whereas computed tomography (CT) provides superior assessment of subchondral bone integrity and structural collapse. In recent years, artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL) techniques, has emerged as a promising tool to enhance diagnostic accuracy, automate lesion segmentation, and predict disease progression. This review aims to provide an overview of current AI applications in MRI for ONFH, focusing on early diagnostic, disease staging and classification, volumetric assessment, differential diagnosis and prognostic prediction. A total of 61 articles were initially identified, of which 13 studies (2021–2025) met the inclusion criteria. Results indicate that DL models, particularly convolutional neural networks (CNNs), achieve excellent diagnostic performance, with reported accuracies up to 98.4% and area under the curve (AUC) values reaching 0.98 for early-stage detection. Several models demonstrated performance comparable to or exceeding that of experienced clinicians, particularly in differentiating ONFH from other hip pathologies and in early disease recognition. AI algorithms also showed high accuracy in staging and classification (AUC up to 99.7% in internal validation), as well as in automated segmentation and volumetric assessment (Dice coefficients up to 0.89), enabling objective quantification of necrotic lesions. Furthermore, prognostic models integrating radiomics and ML techniques demonstrated promising results in predicting femoral head collapse (AUC up to 0.85). From a clinical perspective, AI appears to function primarily as a supportive tool, improving diagnostic consistency, efficiency, and reproducibility, and acting as a “second reader” capable of reducing variability among less experienced clinicians. However, significant limitations remain, including dataset heterogeneity, predominance of retrospective and monocentric studies, and limited integration of clinical data. In conclusion, AI-based MRI analysis shows strong potential to enhance the diagnosis, staging, and management of ONFH. Future research should focus on multicenter prospective validation, integration of multimodal clinical data, and development of explainable and generalizable models to facilitate widespread clinical adoption. Full article

Background/Objectives: We aimed to evaluate the methodological quality, translational limitations, and clinical applicability of current artificial intelligence (AI) applications in periodontology and to propose a framework for validated prediction and decision support. Methods: A structured narrative review based on a targeted, non-systematic literature search was conducted using PubMed and cross-disciplinary sources (January 2015–April 2026). Evidence from primary studies, systematic reviews, and methodological guidance for AI prediction models and clinical decision-support systems was synthesized with a focus on clinical applicability. Results: Current periodontal AI research is dominated by retrospective studies focusing on radiographic phenotyping, where deep learning models demonstrate promising diagnostic performance for detecting and quantifying periodontal bone loss. However, substantial limitations persist, including heterogeneous endpoints, inconsistent reporting, limited external validation, and insufficient calibration assessment. Importantly, there is little evidence that AI-based tools improve clinical decision-making or patient-relevant outcomes. Emerging work on prognostic modeling and multimodal data integration highlights the potential for individualized periodontal risk prediction but remains undervalidated, with limited evidence for clinical implementation. Conclusions: Although AI-based models show promising diagnostic performance, translational progress in periodontology is currently limited by insufficient validation and the lack of evidence for clinical utility. Future research should prioritize clinically actionable prediction models, robust external validation, and prospective evaluation of AI-supported decision-making within real-world periodontal care pathways. Full article

Commercial artificial intelligence (AI)-assisted electrocardiography systems require local assessment of the acquisition-to-review pathway before patient-facing use. We conducted a single-site, simulation-based functional assessment of one commercial 12-lead ECG configuration. A physiological signal simulator, ECG acquisition unit, AI-assisted review platform, multiparameter monitor, and central monitoring system were used to examine 13 predefined rhythm and conduction categories, alarm behavior, local export pathways, a representative five-lead signal-display subset, and formative expert feedback. Expected category-level functional agreement was observed in 12 of 13 predefined categories. In the retained record, the discordant asystole scenario was categorized as lead-off/electrode disconnection, identifying a signal-integrity boundary that requires verification of electrodes, leads, cables, waveforms, and context before clinician adjudication. Signal-to-noise ratio and root-mean-square error values from 90 derived windows in five leads were reported descriptively. These findings cannot be generalized to the seven unmeasured diagnostic leads or to complete 12-lead signal fidelity. Local PDF, HL7 v2.x, DICOM, and platform synchronization were demonstrated without independent conformance or semantic-integrity testing. Formative input from six purposively selected experts comprised 30 ordinal ratings (median 5; observed range 4–5) and qualitative comments. These data do not constitute usability or educational-effectiveness validation. The study provides bounded functional evidence for the tested configuration and identifies requirements for prospective technical, human-factors, multivendor, and patient-level validation before clinical deployment. Full article

Accurate morphological analysis of blood smears is vital for hematological diagnosis, yet manual examination is labor-intensive and subjective. While deep learning offers automation, its black-box nature and computational demands often hinder clinical trust and deployment. We propose XHIC-Net, an Explainable Hybrid Involution–Convolution Network designed for efficient and transparent cell classification. By integrating spatially adaptive involution operations with convolutional layers within a residual framework, XHIC-Net captures both contextual and fine-grained features efficiently. To enhance interpretability, a Grad-CAM-based explainable AI (XAI) module visualizes the cellular regions driving model predictions. The proposed framework was evaluated on a dataset comprising 12,879 microscopic blood smear images belonging to 12 morphological cell categories. Experimental results demonstrate that XHIC-Net achieves an overall accuracy of 98.88%, precision of 98.89%, recall of 98.87%, F1-score of 0.9887, and Cohen’s Kappa score of 0.9887. It outperformed established models, including DL models such as EfficientNetV2S, MobileNet family, DenseNet family, and VGG16, while using fewer parameters and requiring shorter training times. Furthermore, the XAI maps consistently highlighted biologically relevant structures, validating the model’s decision-making process. XHIC-Net is a strong, effective, and clear research model for automated hematology. With future clinical validation, it has the potential to be modified for point-of-care diagnostics in healthcare settings with limited resources. Full article

AI-generated synthetic media are evolving from tools of digital manipulation into a practical antidote for persistent challenges in digital health implementation. However, the computational capabilities that characterise these technologies, their applications and downstream health-related benefits remain fragmented and insufficiently synthesised. This systematic review identified the computational capabilities that characterise AI-generated synthetic media in health, examined their applications and benefits, and developed an integrative framework linking these domains. Twenty-four studies published between 2021 and 31 May 2026 were included. Methodological quality was assessed using the Mixed Methods Appraisal Tool (MMAT) and findings were synthesised through thematic analysis. The synthesis revealed an integrated set of capabilities spanning photorealistic medical-image generation, modality-specific synthesis of clinical images and physiological signals, synthetic non-image health-data creation, preservation of statistical distributions, temporal patterns and clinical relationships, generation of diverse, novel and non-memorised samples and controlled transformation of medical and audiovisual content. Privacy-oriented synthesis and deepfake detection emerged as distinct components supporting privacy-conscious data use, clinical verification and healthcare safety. These demonstrated capabilities were linked to empirically evaluated and indicated applications, including data augmentation, AI model training, diagnostic model development, privacy-oriented health-data sharing, medical education, patient-facing communication, therapeutic support, clinical safety, health-system analytics and planning. The resulting Computational Capability–Application–Benefit (CAB) Framework conceptualises synthetic media as an evidence-graded pathway distinguishing demonstrated computational capabilities, evaluated health-related applications and reported, indicated or potential downstream benefits requiring further validation. AI-generated synthetic media, therefore, represent an emerging computational infrastructure with potential to support safer, privacy-conscious, adaptive and data-intensive healthcare. Full article

Micromachined ultrasonic transducers (MUTs) represent a notable advance in miniaturized sensing, enabling compact, low-power, and complementary metal-oxide-semiconductor (CMOS)-integrated platforms that extend ultrasonic capabilities into wearable, implantable, and edge-computing domains. The integration of artificial intelligence (AI) has introduced new approaches for signal interpretation, adaptive control, and data-driven optimization, enhancing performance in specific areas such as compressed sensing, neural beamforming, and learned image enhancement that complement conventional signal processing. Meanwhile, sensor fusion strategies that combine ultrasonic data with complementary modalities have improved robustness, contextual awareness, and diagnostic accuracy across applications ranging from industrial monitoring to clinical diagnostics. This review provides a comprehensive analysis of this active research area, systematically covering transducer hardware platforms, design methodologies, and intelligent signal processing frameworks. While traditional bulk piezoelectric transducers remain the benchmark for high-power applications, capacitive and piezoelectric micromachined variants offer superior acoustic impedance matching and monolithic CMOS compatibility essential for portable systems. We examine the evolution from deterministic analytical and numerical modeling toward AI-powered inverse design, which enables the discovery of non-intuitive, high-performance geometries beyond human intuition. Furthermore, the integration of machine learning (ML) for signal recovery, image enhancement, and multi-modal sensor fusion is discussed as a pathway to compensate for hardware constraints such as limited aperture, sparse sampling, and low signal-to-noise ratio (SNR), while pointing out that AI technology cannot overcome fundamental physical limits including acoustic attenuation, thermal noise floors, and transduction efficiency boundaries. By synthesizing recent advancements, this review demonstrates how the convergence of classical acoustic physics and data-driven intelligence is guiding the development of of intelligent ultrasonic systems. Full article

Background/Objectives: Rosacea is a chronic inflammatory skin disease that requires long-term management and continuous patient education regarding triggers, skincare practices, and treatment adherence. In recent years, patients have increasingly turned to online platforms and artificial intelligence (AI)-based chatbots for health-related information. Although ChatGPT has been evaluated in the context of rosacea, evidence regarding the performance of other AI chatbots remains limited. This study aimed to evaluate the accuracy, reliability, quality, readability, and diagnostic relevance of AI-generated responses to common rosacea-related patient questions and to assess their potential role as sources of health-related information. Methods: Between 21 December 2025 and 22 February 2026, rosacea-related questions were collected from the publicly accessible Quora platform using a systematic screening process. Twenty clinically relevant and representative questions covering diagnosis, triggers, treatment options, skincare practices, and disease manifestations were selected. Each question was independently submitted to three AI chatbots (Claude Sonnet 4.5, ChatGPT-5.2, and DeepSeek-V3.2). Responses were evaluated by domain experts using the modified DISCERN (mDISCERN) for reliability, the Global Quality Scale (GQS) for overall quality, the Flesch Reading Ease Score (FRES) for readability, and a 5-point Likert scale for accuracy. Reference hallucinations were assessed through manual verification of cited sources. Statistical comparisons were performed using the Friedman test with Bonferroni-adjusted post hoc analyses, and effect sizes were calculated using Kendall’s coefficient of concordance (Kendall’s W). Results: Significant differences were observed among the AI chatbots across all evaluation domains (p < 0.05), with moderate to large effect sizes (Kendall’s W = 0.272–0.683). ChatGPT-5.2 and DeepSeek-V3.2 demonstrated significantly higher reliability and accuracy scores than Claude Sonnet 4.5. DeepSeek-V3.2 achieved the highest overall quality scores, whereas ChatGPT-5.2 produced the most readable responses. Reference analysis revealed variable hallucination rates among the evaluated models. Conclusions: Generative AI chatbots demonstrate considerable potential as sources of health-related information for rosacea-related queries. However, variability in performance and reference hallucination rates highlights the need for careful validation before their widespread use as complementary sources of patient health information. Full article

Artificial intelligence (AI) is transforming cardiovascular medicine through applications in disease detection, diagnosis, risk prediction, and clinical decision support. However, the clinical implementation of these technologies remains poorly characterised. This systematic review evaluated the current landscape of AI applications in cardiovascular medicine, focusing on implementation maturity and clinical translation. The review followed PRISMA guidelines and a prospectively registered PROSPERO protocol. Data extraction included study characteristics, cardiovascular domain, AI methodology, clinical application, validation strategy and implementation maturity, assessed using a predefined five-level framework. AI methodologies were classified as conventional machine learning, deep learning, hybrid ML/deep learning, multimodal AI, or large language models/generative AI. Seventy-four studies met the eligibility criteria. Conventional machine learning was the most frequently used methodology (47.3%), followed by deep learning (41.9%), whereas multimodal AI (5.4%), hybrid ML/deep learning (2.7%), and large language models/generative AI (2.7%) were uncommon. Applications focused mainly on screening and early detection (31.1%), risk stratification and prognosis (25.7%), treatment planning (16.2%), diagnosis (13.5%), and monitoring (12.2%). Most studies reached implementation maturity Level 3 (clinical validation, 47.3%) or Level 2 (technical validation, 35.1%), while only 13.5% achieved routine clinical implementation (Level 5) and 4.1% reached clinical deployment (Level 4). Although AI demonstrated promising diagnostic and prognostic performance across multiple cardiovascular conditions, most applications remain at the validation stage. Future research should prioritise implementation science, pragmatic evaluation, and real-world evidence to facilitate routine adoption and maximise patient benefit and healthcare value. Full article

Globally, respiratory diseases such as asthma, chronic obstructive pulmonary disease (COPD) and pneumonia affect populations significantly, requiring early and accurate diagnosis for effective clinical management. Manual auscultation and expert interpretation are the common shortcomings in conventional diagnostic approaches, as they lead to time-consuming and inconsistent analysis. To address these limitations, an artificial intelligence-driven framework for respiratory disease classification using multi-spectral feature extraction and deep learning architectures is proposed to classify four different respiratory conditions: Asthma, COPD, Pneumonia and Healthy. The dataset is collected from Kaggle’s respiratory sound database and the COUGHVID V3 database, which together contain 322 Asthma signals, 746 COPD signals, 323 Pneumonia signals and 174 Healthy signals. Subsequently, the features are extracted using four different feature extraction techniques—Constant Q Transform (CQT), a Gammatone spectrogram, Mel-Frequency Cepstral Coefficients (MFCC) and Perceptual Linear Prediction (PLP)—and these extracted spectral representations are provided as inputs to various deep learning models such as a Deep Convolutional Neural Network (Deep CNN), a Temporal Attention Network (TAN) and an Autoencoder for automated feature learning and disease classification. The proposed framework is evaluated using several performance metrics, and the experimental results clearly indicate that the performance of the proposed classification framework strongly depends on the selection of spectral feature extraction techniques and deep learning models. Among all the evaluated combinations, it is evident that the Autoencoder model integrated with CQT features exhibited the best classification performance, with an accuracy of 98.72%, precision of 98.74%, recall of 98.72%, Matthews correlation coefficient (MCC) of 98.11%, Cohen’s kappa value of 98.10% and the least log loss of 0.025. The proposed artificial intelligence (AI)-enabled respiratory disease classification framework has demonstrated the ability to produce a reliable computer-aided diagnostic system which is suitable for smart healthcare applications and automated pulmonary disease screening. Full article