Jul 2026· International Conference on Digital Health· pp. 52-60· 0 citations· 33 references
Abstract
Peripheral Artery Disease (PAD) is a serious and potentially limb and life-threatening condition that demands a timely and accurate diagnosis to prevent severe complications. Traditional diagnostic methods are often slow and error prone. Although PAD localization remains relatively less explored using machine learning, the development of efficient and deployable models for real-time arterial blockage identification demands more investigation. In this paper, we propose PAD-Net, a novel end-to-end object detection framework specifically optimized for the identification and localization of arterial blockage. By integrating Light-Weight Feature Adapter (LFA) and Transformer block into You Only Look Once (YOLO) architecture, our model enhances the possibility of capturing multi-scale vascular structures often obscured in complex medical backgrounds/conditions. The proposed model was evaluated on a novel dataset of lower-extremity CT angiograms collected from the National Institute of Cardiovascular Disease, Bangladesh. Experimental results demonstrate that the model achieves a Mean Average Precision of 0.87 (mAP@0.5) and 0.423 (mAP@0.5:0.95), achieving competitive performance compared with state-of-theart YOLO and RT-DETR (Real Time Detection Transformer) models. At the same time, it maintains real-time speed to make predictions. The proposed PAD-Net gives a strong and scalable solution for AI-based PAD diagnosis. Furthermore, to validate real-world applicability, the model is deployed on a web-based Streamlit Community Cloud. The deployment of the proposed model highlights the effectiveness of reliable PAD detection and its integration into clinical workflows.
Abstract Aims Early identification of obstructive coronary artery disease (ObCAD) is crucial because it is strongly associated with acute myocardial infarction. We developed a weighted average ensemble model integrating deep learning (DL) and machine learning (ML) to leverage imaging and clinical data for enhancing the detection of ObCAD. Methods and results A retrospective cohort of 1054 patients was used to develop an ensemble model combining a 3D Vision Transformer with eXtreme Gradient Boosting and CatBoost for binary classification of ObCAD (>50% stenosis). Unstructured data comprised 3D cardiac non-contrast computed tomography (CT) scans, while structured data included 11 demographic and clinical features. Obstructive coronary artery disease labels were derived from corresponding coronary CT angiography. Model performance was evaluated using 10-fold cross-validation with fold-wise Wilcoxon signed-rank testing. The ensemble model achieved a mean receiver operating characteristic area under the curve (ROC AUC) of 0.81 ± 0.04 and an accuracy of 0.76 ± 0.04. It demonstrated a statistically significantly higher ROC AUC than individual component models. Feature importance analysis identified age, chest pain, and sex as the most influential predictors of ObCAD. Gradient-weighted class activation mapping visualization indicated that the 3D Vision Transformer primarily focused on cardiac regions containing coronary artery calcium deposits. Conclusion Integrating DL-based imaging analysis with ML-based clinical modelling enhances the discriminative performance for ObCAD detection with complementary interpretability. This ensemble framework demonstrates potential to support clinical decision-making by identifying high-risk patients using routine cardiac CT combined with patient-level clinical data. Future studies using external validation and coronary artery calcium scores may further improve risk prediction.
Doyoung Park, Lin-Xuan Yan, Arman Ahmad Khan et al.· European Heart Journal - Dig...· 0 citations
Heart disease is significant health burden and a major cause of death worldwide, hence the need to develop automated diagnostic systems with correctness and computational efficiency in order to rescue lives through early clinical intervention. This paper introduces a fine-grained deep transfer learning architecture to classify cardiac disease images into multiple classes using two large convolutional backbones ResNet50 and DenseNet121, which are trained systematically with AdamW, stochastic gradient descent (SGD), and Lion under a single experimental environment. The suggested pipeline combines standardized image preprocessing, stratified division of data, transfer learning hierarchical feature extraction based on the feature, hyper parameter convergence analysis, and 50-epoch supervised fine-tuning, and then thorough performance and efficiency assessment. The experimental findings indicate the stable optimization dynamic in all six backbone optimizer setups, with the most consistent validation convergence in ResNet50 with Lion optimizer. The most successful configuration obtained nearly 93.9% accuracy of validation, precision, recall, and F1-score and a robust ROC-AUC of 0.995, which verified a high inter-class separability and stable threshold-independent result. The loss and validation accuracy curves also show that the convergence is quick and the ability to generalize is high. In terms of deployment, DenseNet121 was found to have much lower architectural complexity (8.0M parameters) and reduced inference latency (8.4 ms/image) than ResNet50 (25.6M parameters, 12.8 ms/image), and yet have a competitive level of classification. The graph driven analysis also shows that optimizer selection mostly influences the smoothness of convergence, predictive calibration, but backbone architecture influences a tradeoff between representational richness and computational efficiency. On the whole, the suggested framework is a clinically applicable and deployment-focused solution to smart computer aided diagnosis of cardiac diseases.
One of the primary causes of death worldwide is still heart disease. Although echocardiography is a commonly used method for identifying cardiovascular diseases, precise interpretation of echocardiogram pictures necessitates specialist medical knowledge. In order to overcome this difficulty, this paper presents a deep learning-based method for automatically classifying heart conditions from echocardiography data using the EfficientNetB0 architecture. For medical picture analysis, EfficientNetB0 offers a lightweight yet effective solution thanks to its compound scaling technique, which balances network depth, width, and resolution. In order to lessen the need for human interpretation, the model is trained to automatically extract intricate and distinctive features from echocardiographic images. EfficientNetB0 is especially well-suited for real-time clinical use since it guarantees great accuracy at a cheap computing cost by utilizing its efficiency and good generalization potential. This strategy seeks to assist healthcare providers in enhancing diagnostic accessibility, consistency, and efficiency. The suggested approach has the potential to improve cardiovascular disease prognosis and early detection, thereby increasing the scalability of sophisticated diagnostic capabilities in a variety of healthcare settings.
Taha Tahseen, Afshan Fatima· International Journal of Eng...· 0 citations
Early detection of individuals at high risk of cardiovascular disease remains a major challenge in preventive healthcare, particularly in regions with limited access to advanced diagnostic facilities. Traditional risk assessment methods rely on laboratory tests and invasive procedures, which restrict large-scale screening.This paper proposes a hybrid deep learning framework for heart disease risk prediction using retinal fundus images. The proposed model integrates deep visual features extracted using convolutional neural networks with handcrafted retinal vascular features such as vessel density, tortuosity, and fractal dimension. While deep learning captures complex spatial patterns, handcrafted features provide clinically interpretable insights. A feature-level fusion strategy is employed to combine both representations and classify individuals into low-, medium-, and high-risk categories. Experimental results demonstrate that the proposed model achieves superior performance compared to standard deep learning models, particularly in identifying highrisk cases. The proposed framework offers a scalable, noninvasive, and interpretable solution for early cardiovascular risk screening.
Dr. M A Srinuvasu, Kumar ProfessorMentor, Dept. of Aids et al.· International Conference on...· 0 citations
Worldwide, cardiovascular diseases (CVDs) are the leading cause of mortality, making early detection a critical public health objective. The analysis of electrocardiograms (ECGs) is a fundamental tool for CVD diagnosis, and recent developments in deep learning have greatly enhanced its diagnostic accuracy. A 12-lead ECG dataset provides valuable information for enhancing diagnostic performance, but extracting and revealing these patterns requires complex and hidden models with numerous parameters and high computational demands. In portable and wearable healthcare devices, GPUs with low power are not suitable for deploying these models. The purpose of this study is to address these limitations by presenting a knowledge distillation (KD) framework that transfers knowledge from a high-capacity teacher model (T-CST) to a lightweight student model (S-CS). A forward lead selection strategy is introduced to identify the most informative single lead or optimal small-lead combinations for efficient diagnosis with minimal data input. The teacher model integrates CNN, SENet, and Transformer modules to capture both local and long-range temporal dependencies, enabling it to extract a diverse set of hidden and comprehensive patterns from 12-lead ECG signals, including subtle temporal relationships and intricate local features. The compact student model, consisting of a CNN and SENet, leverages this transferred knowledge to achieve comparable performance with substantially lower computational cost. Experiments on the PTB-XL dataset demonstrate that S-CS attains near-teacher accuracy using only three leads while reducing computation by 63 times, highlighting its promise for efficient, accurate, and accessible CVD detection in portable and wearable healthcare systems.
Nasim Beigzadeh, A. Fathi· Journal of Medical Engineeri...· 0 citations
Cardiovascular disease (CVD) remains the leading cause of global mortality, necessitating non-invasive, accurate, and interpretable screening tools. Retinal fundus imaging offers an accessible, low-cost means of assessing systemic vascular condition, since microvascular changes visible in the retina tend to parallel those occurring in the coronary vessels. This paper presents a deep learning framework that integrates Contrast-Limited Adaptive Histogram Equalization (CLAHE) preprocessing with a fine-tuned EfficientNet-B0 architecture for binary CVD risk classification. Gradient-weighted Class Activation Mapping (Grad-CAM) is incorporated to generate spatially localized saliency overlays, enabling clinically interpretable predictions by highlighting retinal regions that most influence model decisions. A quadrant-level activation analysis further breaks down attention patterns relative to underlying coronary risk. Evaluation on a publicly available retinal fundus dataset shows that the proposed model attains 94.7% classification accuracy, 93.2% sensitivity, and 95.8% specificity, surpassing several existing deep learning baselines. CLAHE improves vessel contrast and cuts down illumination artifacts, and this turns out to matter a lot for how well the model performs. The Grad-CAM outputs back this up too: the attention patterns line up with what we’d expect clinically, and in high-risk cases we see diffuse activation in the superior-temporal region that matches arteriovenous nicking and vessel tortuosity. Collectively, these findings suggest a practical direction forward: an interpretable retinal imaging pipeline for cardiovascular disease (CVD) screening that could feasibly be integrated into ophthalmology clinics and resource-constrained primary care settings.
Shruthi T, Dr. Supriya Shrivastav· International Research Journ...· 0 citations
