Machine learning (ML) models are increasingly proposed to support clinical decision-making, yet their evidentiary basis remains weaker than their publication volume suggests. This editorial argues that the problem is not only translational, regulatory, or infrastructural, but methodological. Many medical ML pipelines rely on uncertain ground truths, optimize performance around clinically irrelevant thresholds, report unstable or prevalence-dependent metrics, neglect calibration and uncertainty, and lack rigorous external and temporal validation. These weaknesses produce optimistic estimates that do not reliably anticipate performance in heterogeneous clinical settings. We call for an evidence-based medical AI grounded in more reliable annotation practices, explicit modeling of uncertainty, clinically meaningful threshold selection, calibration and decision-utility analyses, robustness testing, external validation on independent datasets, and post-deployment monitoring. The editorial also invites authors, reviewers, users, and vendors to adopt stricter standards so that predictive models can become credible, accountable, and clinically useful tools in everyday practice, rather than merely publishable artifacts.

Diagnosing ischemic stroke from computed tomography (CT) images is a highly challenging and detailed process that requires precise and careful analysis by a medical professional. Deep learning techniques offer an effective solution to this issue because of their remarkable performance. Nevertheless, most of those methods still lack the uncertainty quantification (UQ) and eXplainable artificial intelligence (XAI) features, which are essential for clinical practice and acceptance. We present TrustNet, a small but powerful convolutional neural network that uses Monte Carlo dropout and quantitative Grad-CAM. This technique helps visualize the issues related to two independent factors: uncertainty in the model’s classification and inconsistency in recognizing the relevant visual features. The model was validated on a set of 2023 brain CT scans and compared with networks that are generally used for classification purposes. TrustNet was able to achieve an accuracy of 94.67%, with 100% specificity, 91.6% sensitivity, and 100% precision, competing against various conventional architectures. The introduction of the UQ and XAI methods led to a consistent performance enhancement over the baseline models by limiting the number incorrect predictions, which is crucial for stroke diagnosis. With this performance, our approach can also provide an explanation for the reasoning and estimate confidence, which is essential for model deployment. This method is an indispensable tool for eliminating diagnostic bias and thus controlling the safety of AI in the clinical workflow.

Cancer is a chaotic disease known as the plague of our age and there are many subtypes of the cancer. Cancer is commonly seen disorder and its mortality rate is very high. Therefore, many researchers have worked/studied on the cancer detection and treatment. To contribute cancer studies according to machine learning, we have presented a new generation convolutional neural network (CNN) termed ShortNeXt in this research. The presented ShortNeXt has inspired by ResNet, ConvNeXt and MobileNet architectures to use the advantages these CNNs together. This model, which aims to extract robust feature map using convolution-based residual blocks, is named ShortNeXt because it incorporates more than one shortcut. The ShortNeXt architecture has four main stages and these stages are: (i) an input/stem, (ii) ShortNeXt, (iii) downsampling, and (iv) output. In this CNN architecture, convolution, batch normalization and the Gaussian Error Linear Unit (GELU) activation functions have been utilized. In this aspect, the implementation of the recommended ShortNeXt is simple. The stem stage uses a 4 x 4 sized convolution with stride 4 like ConvNeXt and Swin Transformer and this operation is named patchify operation. Additionally, a 2 x 2 patchify block has been used in the downsampling block. In the ShortNeXt block, an inverted bottleneck has been used, and both 1 x 1 and 3 x 3 convolution blocks are employed in the expansion phase. The output layer has increased the number of filters from 768 to 1280 by using pixel-wise convolution, drawing inspiration from MobileNetV2 and a final feature map with a length of 1280 has been obtained by deploying global average pooling (GAP). In the classification phase, fully connected and softmax operators have been used. To get comparative results about to the recommended ShortNeXt, a publicly available histopathological image dataset has been used and this dataset contains nine classes, and the proposed ShortNeXt has achieved 97.82% and 97.86% validation and test accuracy, respectively. The obtained results and findings openly showcases that ShortNeXt is an effective deep learning method for histopathological image classification for cancer detection/classification.

The presence of artifacts in whole slide images (WSIs), such as tissue folds, air bubbles, and out-of-focus regions, can significantly impact WSI digitization, pathologists' evaluation, and the accuracy of downstream analyses. We present SlideInspect, a novel AI-based framework for comprehensive artifact detection and quality control in digital pathology. Our system leverages deep learning techniques to segment multiple artifact types across diverse tissue types and staining methods. SlideInspect provides a hierarchical output: a color-coded slide quality indicator (green, yellow, red) with recommended actions (no action, re-scan, re-mount, re-cut) based on artifact type and extent, and pixel-level segmentation masks for detailed analysis. The system operates at multiple magnifications (1.25x for tissue segmentation, 5x for artifact detection) and also incorporates stain quality assessment for histological stain evaluation. We validated SlideInspect on a large, multi-centric, multi-scanner dataset of over 3000 WSIs, demonstrating robust performance across different tissue types, staining methods, and scanning platforms. The system achieves high segmentation accuracy for various artifacts while maintaining computational efficiency (average processing time: 72.7 s per WSI). Pathologist evaluations confirmed the clinical relevance and accuracy of SlideInspect's quality assessments. By providing actionable insights at multiple levels of granularity, SlideInspect significantly improves the efficiency and standardization of digital pathology workflows. Its vendor-agnostic design and multi-stain capability make it suitable for integration into diverse clinical and research settings.

Recent trends in computational pathology favour increasingly complex deep learning architectures, raising the question of whether such complexity is necessary for routine diagnostic tasks. This study challenges this assumption through a comprehensive analysis of the relationship between model complexity, data pre-processing, and performance across four fundamental digital pathology tasks: nuclei counting, steatosis quantification, glomeruli detection, and Ki67 proliferation index (PI) assessment. We evaluated five deep learning models of varying complexity (lightweight: MobileNetV2, U-Net, and more complex: ConvNeXt, K-Net, and Swin Transformer) combined with different image pre-processing techniques. To evaluate model performance without extensive ground truth (GT) annotations, we introduced a validation strategy utilizing the relative absolute deviation (RAD) between network predictions and correlation of performance metrics. Our findings demonstrate that pre-processing strategies, particularly stain normalization (NORM), can be more impactful than model complexity, reducing error rates by up to 50% compared to processing original (ORIG) images. With appropriate pre-processing, lightweight models achieved comparable or superior results to complex models while reducing processing times by up to 40%. Only specific tasks involving complex morphological features, such as glomeruli detection, significantly benefited from more sophisticated architectures. This study provides an evidence-based framework for selecting optimal model-pre-processing combinations in clinical settings, suggesting that investing in pre-processing pipelines rather than model complexity may be more beneficial for routine computational pathology applications.

Accurate segmentation of the prostate gland in magnetic resonance imaging (MRI) is important for the early detection and treatment of prostate cancer. Manual segmentation is time-consuming and prone to interobserver variability. This study evaluates and compares two deep learning models, U-Net and E-Net, for automated prostate segmentation using axial T2-weighted MRI scans. Both models were trained and validated using a five-fold cross-validation approach, and tested on a hold-out dataset. Performance was assessed using metrics such as Dice Similarity Coefficient (DSC), precision, recall, and surface distance measures. E-Net achieved superior results with a mean DSC of 81.61

ABSTRACT Gastrointestinal (GI) diseases remain among the leading causes of global mortality, with early detection directly linked to survival outcomes. While previous reviews have focused on single imaging modalities, this systematic review uniquely examines artificial intelligence applications across endoscopic, radiological, and histological approaches, reflecting actual clinical diagnostic pathways. This systematic review analyzes 76 high‐quality studies (2016–2024) and provides the first comprehensive assessment of how AI performs across different imaging techniques for GI abnormality detection. This multi‐modal perspective is particularly timely as healthcare systems move toward integrated diagnostic workflows. Our analysis reveals endoscopy as the most widely used modality ( n = 44), particularly for Helicobacter pylori , colorectal polyps, and ulcerative colitis detection. Histological analysis emerges as the second most common approach ( n = 25), especially for celiac disease and ulcerative colitis, while CT imaging ( n = 10) primarily supports colorectal polyp detection. Deep learning methods significantly outnumber traditional machine learning techniques (68 vs. 8 studies), consistently achieving 90%–99% diagnostic accuracy across multiple disease categories. However, these systems face significant implementation barriers to clinical adoption. Most validation is still conducted in controlled, single‐center settings using curated datasets that poorly reflect clinical complexity. Future studies must prioritize multicenter validation, standardized imaging protocols and preprocessing pipelines, and the integration of interpretable AI models capable of providing transparent diagnostic rationale. This review maps the current technical landscape while highlighting critical translational challenges that must be addressed to enable real‐world impact. This article is categorized under: Technologies > Data Preprocessing Technologies > Artificial Intelligence

Background Fuzzy logic techniques have gained significant prominence in healthcare, primarily due to their ability to address and manage the inherent imprecision and uncertainty in healthcare data analysis. We conducted a comprehensive review investigating how fuzzy techniques have developed and been applied in healthcare between 2017 and 2025. Methods We conducted a systematic literature review following PRISMA guidelines, analyzing 91 papers from major medical and engineering databases. Our analysis focused on three distinct methodological streams: classical fuzzy systems, combined fuzzy-machine learning approaches, and emerging fuzzy-enhanced deep learning frameworks. We evaluated each paper’s methodology, implementation details, and clinical relevance. Results The distribution of research approaches showed a balanced landscape across methodologies, with traditional fuzzy systems comprising 30.1%, hybrid approaches 34.4%, and fuzzy-deep learning implementations 33.3% of studies. Medical imaging dominated the application domains, led by MRI studies (36.3%) and CT applications (12.1%). Biosignal analysis also showed strong representation, particularly in EEG (22%) and ECG (7.7%) applications. Performance analysis revealed that both deep learning and conventional feature engineering methods achieved comparable accuracy rates of approximately 96.5%, with some variations in consistency across different applications. Conclusions This research area has undergone significant evolution, particularly since 2023, with an increased emphasis on incorporating fuzzy techniques into deep learning frameworks. This transition shows that fuzzy approaches, originally designed as standalone solutions, are now becoming critical components of modern healthcare AI systems, providing unique benefits in dealing with medical data uncertainty.

Background and Purpose Validation of deformable image registration (DIR) remains predominantly contourbased; this study evaluated inverse consistency error (ICE) as an automated voxelwise metric for DIR accuracy. Materials and Methods Synthetic ground-truth DVFs were generated using geometric and head-and-neck (HN) digital phantoms undergoing controlled global and local deformations. DIR was performed with the ANACONDA algorithm in RayStation. ICE maps derived from clinical DVFs were compared with ground-truth registration error (GTRE), target registration error (TRE) from 20 anatomical landmarks, and mean distance to agreement (MDA) for 22 propagated ROIs. Results Ground-truth DVFs showed negligible ICE values, confirming mathematical invertibility. In HN phantoms, median ICE and GTRE were 0.8 ± 0.2 mm and 1.6 ± 0.4 mm, respectively. ICE correlated strongly with GTRE (R = 0.85, p < 0.001) and moderately with TRE (R = 0.68, p < 0.001). No significant correlation was found with contourbased MDA (2.47 ± 0.18 mm). Voxel-wise analysis showed that ICE captured spatial patterns of uncertainty consistent with regions of higher GTRE, while underestimating error for global homogeneous deformations >15 mm due to DIR regularisation. Across all datasets, ICE correctly identified high-uncertainty subregions that were not detected by contour-based metrics. Conclusions ICE enables automated voxel-wise quantification of DIR uncertainty directly from clinical DVFs. It complements traditional contour-based metrics and may support patient-specific QA and more reliable dose mapping in adaptive and re-irradiation radiotherapy workflows.

Myocardial infarction (MI) detection and localization through echocardiography are crucial for effective patient management. However, current diagnostic approaches rely heavily on visual assessment, which can be subjective. In this work we developed a cascade framework for automated MI diagnosis and localization in echocardiograms. Our method combines deep learning for left ventricle wall segmentation with machine learning classification using clinically relevant features. Specifically, we employ a U-Net architecture for segmentation, followed by a two-stage Random Forest classifier for MI detection and localization. We trained and evaluated our approach on two public datasets - CAMUS and HMC-QU. The proposed method achieved 100 % sensitivity and 89.8 % specificity for segment identification, outperforming single-stage classification methods. To the best of our knowledge, this is the first study to apply a multi-step artificial intelligence system combining segmentation and classification for MI diagnosis from echocardiography. This interpretable cascade framework exhibits high performance for early detection and localization of myocardial infarction, demonstrating potential as a clinical decision support tool.

The integration of wavelet transformation and artificial intelligence techniques has demonstrated significant potential in healthcare applications. Wavelet analysis enables multi-scale signal decomposition and feature extraction that, when combined with machine and deep learning approaches, enhance the accuracy and efficiency of medical data analysis. This systematic review synthesizes 112 relevant studies from 2013 to 2023 exploring wavelet-based artificial intelligence in healthcare. Our analysis reveals that the discrete wavelet transform dominates (43% of studies), primarily used for feature extraction from biosignals (82%) and medical images. Major applications include cardiac abnormality detection (29%), neurological disorder diagnosis (27%), and mental health assessment (16%), with classification accuracies frequently exceeding 95%. Key findings indicate a shift from traditional machine learning to deep learning approaches after 2020, with emerging trends in hybrid architectures. The review identifies critical challenges in computational efficiency, optimal wavelet selection, and clinical validation. Future developments should focus on real-time processing optimization, interpretable deep learning models, multi-modal data fusion, and validation on larger clinical datasets, advancing the translation of these systems into practical clinical tools.

One of the main causes of death and permanent disability is ischemic stroke, for which prompt and precise diagnosis is essential to successful treatment. This study introduces a novel dual-stream deep learning framework for ischemic stroke classification using Computed Tomography (CT) images, specifically addressing challenges in accuracy, computational efficiency, and clinical interpretability. Three significant innovations are included in the suggested architecture: (1) a hybrid Dual Attention Mechanism that combines Dynamic Routing and Cross-Attention for improved region-specific feature discrimination; (2) a Multi-Scale Feature Extraction Module with parallel convolutional pathways that captures both contextual and fine-grained features; and (3) an Adaptive Random Vector Functional Link layer that significantly reduces training time while maintaining high classification performance. When tested on a single-center dataset, the model achieves state-of-the-art classification accuracy of 98.83% across normal, acute and chronic stroke categories. We demonstrate the strong generalization capabilities of the proposed framework by achieving 92.42% accuracy on a diverse, multi-center dataset of 7,842 CT images. The integration of explainable Artificial Intelligence tools improve clinical trustworthiness by offering clear insight into the model’s decision-making process. These outcomes demonstrate the model’s potential to use in actual clinical settings for quick and accurate stroke diagnosis, along with its interpretability and computational efficiency.

Infrared thermography (IRT) is a non-invasive, radiation-free imaging technique that uses an infrared (IR) camera to record and produce an image using IR radiation emitted from the body. IRT imaging has shown promise as a screening method for breast cancer, diabetic foot ulcers, and dry eye disease, among other medical disorders. The aim of this systematic review is to present a complete overview of the applications of artificial intelligence (AI) techniques with IRT imaging for medical decision support systems over the course of the last ten years (2013–2024). Several scientific databases, including PubMed, IEEE, and Google Scholar, were searched using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. After meeting the requirements for inclusion, 131 papers were selected. The reviewed studies demonstrated how various AI techniques, including deep learning and classical machine learning, can be used to develop automated diagnosis systems using IRT images. The efficacy of these AI systems differed depending on the medical field; for example, they could identify dry eye disease with 90–100% accuracy, classify diabetic foot ulcers with 85–95% accuracy, and detect breast cancer with 80–100% accuracy. This review highlights the value of IRT imaging in early disease detection, especially when combined with AI techniques. This work discusses challenges in using deep learning (DL) models in healthcare, including data scarcity and ethical considerations. It also, proposes three main recommendations: dataset standardization for ethical data management, clear governance models for ethical practices, and the use of Multimodal Large Language Models (MLLMs) to address explainability issues.

Suicide is a major global public health concern, and the application of artificial intelligence (AI) methods, such as natural language processing (NLP), machine learning (ML), and deep learning (DL), has shown promise in advancing suicide prediction and prevention efforts. Recent advancements in AI - particularly NLP and DL have opened up new avenues of research in suicide prediction and prevention. While several papers have reviewed specific detection techniques like NLP or DL, there has been no recent study that acts as a one-stop-shop, providing a comprehensive overview of all AI-based studies in this field. In this work, we conduct a systematic literature review to identify relevant studies published between 2019 and 2023, resulting in the inclusion of 156 studies. We provide a comprehensive overview of the current state of research conducted on AI-driven suicide prevention and prediction, focusing on different data types and AI techniques employed. We discuss the benefits and challenges of these approaches and propose future research directions to improve the practical application of AI in suicide research. AI is highly capable of improving the accuracy and efficiency of risk assessment, enabling personalized interventions, and enhancing our understanding of risk and protective factors. Multidisciplinary approaches combining diverse data sources and AI methods can help identify individuals at risk by analyzing social media content, patient histories, and data from mobile devices, enabling timely intervention. However, challenges related to data privacy, algorithmic bias, model interpretability, and real-world implementation must be addressed to realize the full potential of these technologies. Future research should focus on integrating prediction and prevention strategies, harnessing multimodal data, and expanding the scope to include diverse populations. Collaboration across disciplines and stakeholders is essential to ensure that AI- driven suicide prevention and prediction efforts are ethical, culturally sensitive, and person-centered.

Fibromyalgia syndrome (FMS) is a long-lasting medical condition that poses significant challenges for diagnosis and management because of its complex and poorly understood nature. It affects millions of people around the globe, predominantly women, causing widespread pain, fatigue, cognitive impairments, and mood disturbances. The lack of objective measures to address FMS complicates its assessment, often leading to delayed or misdiagnosed cases. By hindering daily activities and productivity, FMS negatively impacts the quality of the patient’s life. Innovative approaches that use medical data, such as bio-signals and bioimaging, combined with machine learning techniques, hold the promise of deepening our knowledge of FMS, which might in turn lead to systems that offer efficient, precise, and personalized physician support. Furthermore, artificial intelligence-driven identification of biomarkers and patient subgroups could improve FMS management. In this systematic review, we explore the role of artificial intelligence in understanding FMS pathophysiology, discuss the present limitations, and shed light on future research avenues, aiming to translate findings into improved clinical outcomes.

The main aim of this study is to propose a novel convolutional neural network, named BrainNeXt, for the automated brain disorders detection using magnetic resonance images (MRI) images. Furthermore, we aim to investigate the performance of our proposed network on various medical applications. To achieve high/robust image classification performance, we gathered a new MRI dataset belonging to four classes: (1) Alzheimer's disease, (2) chronic ischemia, (3) multiple sclerosis, and (4) control. Inspired by ConvNeXt, we designed BrainNeXt as a lightweight classification model by incorporating the structural elements of the Swin Transformers Tiny model. By training our model on the collected dataset, a pretrained BrainNeXt model was obtained. Additionally, we have suggested a feature engineering (FE) approach based on the pretrained BrainNeXt, which extracted features from fixed-sized patches. To select the most discriminative/informative features, we employed the neighborhood component analysis selector in the feature selection phase. As the classifier for our patch-based FE approach, we utilized the support vector machine classifier. Our recommended BrainNeXt approach achieved an accuracy of 100% and 91.35% for training and validation. The recommended model obtained the test classification accuracy of 94.21%. To further improve the classification performance, we suggested a patch-based DFE approach, which achieved a test accuracy of 99.73%. The obtained results, surpassing 90% accuracy on the test dataset, demonstrate the effectiveness and high classification performance of the proposed models.

OBJECTIVE:Parkinson's disease (PD) presents challenges in early diagnosis and progression prediction. Recent advancements in machine learning, particularly convolutional-neural-networks (CNNs), show promise in enhancing diagnostic accuracy and prognostic capabilities using neuroimaging data. The aims of this study were: (i) develop a 3D-CNN based on MRI to distinguish controls and PD patients and (ii) employ CNN to predict the progression of PD. METHODS:Three cohorts were selected: 86 mild, 62 moderate-to-severe PD patients, and 60 controls; 14 mild-PD patients and 14 controls from Parkinson's Progression Markers Initiative database, and 38 de novo mild-PD patients and 38 controls. All participants underwent MRI scans and clinical evaluation at baseline and over 2-years. PD subjects were classified in two clusters of different progression using k-means clustering based on baseline and follow-up UDPRS-III scores. A 3D-CNN was built and tested on PD patients and controls, with binary classifications: controls vs moderate-to-severe PD, controls vs mild-PD, and two clusters of PD progression. The effect of transfer learning was also tested. RESULTS:CNN effectively differentiated moderate-to-severe PD from controls (74% accuracy) using MRI data alone. Transfer learning significantly improved performance in distinguishing mild-PD from controls (64% accuracy). For predicting disease progression, the model achieved over 70% accuracy by combining MRI and clinical data. Brain regions most influential in the CNN's decisions were visualized. CONCLUSIONS:CNN, integrating multimodal data and transfer learning, provides encouraging results toward early-stage classification and progression monitoring in PD. Its explainability through activation maps offers potential for clinical application in early diagnosis and personalized monitoring.

Left Ventricular Hypertrophy (LVH) is a significant cardiovascular risk marker that manifests in several clinical conditions, including Hypertension (HTN), Chronic Kidney Disease (CKD), and Hypertrophic Cardiomyopathy (HCM). This systematic review examines Artificial Intelligence (AI) approaches for the automated identification of these conditions using cardiac ultrasound (US) imaging. Following the PRISMA guidelines, 37 relevant articles (7 reviews, 30 research papers) published between 2010 and 2025 were analysed. The analysis revealed three primary methodological approaches: feature learning pipelines, end-to-end Deep Learning (DL), and hybrid methods that combine both techniques. For CKD detection, only one study using cardiac US was identified, which achieved 99.09% classification accuracy using Support Vector Machine (SVM) with steerable Gaussian filters and entropy features. HTN classification studies have demonstrated high performance across different approaches: traditional Machine Learning (ML) classifiers (decision trees with transform features: 99.11%, weighted k-nearest neighbors: 98%) and DL methods (AUC: 0.92-0.94). HCM studies ranged from binary classification (42.3% of studies) to multi-class problems of increasing complexity (3-class: 38.4%, 4-class: 11.5%, 5-class: 7.6%), with SVM achieving 95.2% average sensitivity and DL models reaching an average AUC of 0.94. Current limitations include a predominant focus on binary classification problems, limited research on cardiac-based CKD detection, and a lack of standardized datasets. Future research directions include developing hybrid methodologies that combine traditional and DL approaches, creating standardized multimodal databases, implementing explainable AI techniques, and integrating IoT technologies for continuous monitoring.