Cardiovascular diseases (CVDs) continue to present a leading cause of mortality worldwide, emphasizing the importance of early and accurate prediction. Electrocardiogram (ECG) signals, central to cardiac monitoring, have increasingly been integrated with Deep Learning (DL) for real-time prediction of CVDs. However, DL models are prone to performance degradation due to concept drift and to catastrophic forgetting. To address this issue, we propose a real-time CVDs prediction approach, referred to as ADWIN-GFR that combines Convolutional Neural Network (CNN) layers, for spatial feature extraction, with Gated Recurrent Units (GRU), for temporal modeling, alongside adaptive drift detection and mitigation mechanisms. The proposed approach integrates Adaptive Windowing (ADWIN) for real-time concept drift detection, a fine-tuning strategy based on Generative Features Replay (GFR) to preserve previously acquired knowledge, and a dynamic replay buffer ensuring variance, diversity, and data distribution coverage. Extensive experiments conducted on the MIT-BIH arrhythmia dataset demonstrate that ADWIN-GFR outperforms standard fine-tuning techniques, achieving an average post-drift accuracy of 95.4%, a macro F1-score of 93.9%, and a remarkably low forgetting score of 0.9%. It also exhibits an average drift detection delay of 12 steps and achieves an adaptation gain of 17.2%. These findings underscore the potential of ADWIN-GFR for deployment in real-world cardiac monitoring systems, including wearable ECG devices and hospital-based patient monitoring platforms.

Skin cancer is among the most common malignancies worldwide, but its mortality burden is largely driven by aggressive subtypes such as melanoma, with outcomes varying across regions and healthcare settings. These variations emphasize the importance of reliable diagnostic technologies that support clinicians in detecting skin malignancies with higher accuracy. Traditional diagnostic methods often rely on subjective visual assessments, which can lead to misdiagnosis. This study addresses these challenges by developing HybridFusionNet, a novel model that integrates Convolutional Neural Networks (CNN) with 1D feature extraction techniques to enhance diagnostic accuracy. Utilizing two extensive datasets, BCN20000 and HAM10000, the methodology includes data preprocessing, application of Synthetic Minority Oversampling Technique combined with Edited Nearest Neighbors (SMOTEENN) for data balancing, and optimization of feature selection using the Tree-based Pipeline Optimization Tool (TPOT). The results demonstrate significant performance improvements over traditional CNN models, achieving an accuracy of 0.9693 on the BCN20000 dataset and 0.9909 on the HAM10000 dataset. The HybridFusionNet model not only outperforms conventional methods but also effectively addresses class imbalance. To enhance transparency, it integrates post-hoc explanation techniques such as LIME, which highlight the features influencing predictions. These findings highlight the potential of HybridFusionNet to support real-world applications, including physician-assist systems, teledermatology, and large-scale skin cancer screening programs. By improving diagnostic efficiency and enabling access to expert-level analysis, the model may enhance patient outcomes and foster greater trust in artificial intelligence (AI)-assisted clinical decision-making.