Wavelet-Enhanced Neural Framework for Accurate Detection of Cardiac Arrhythmia

Abstract

Cardiac arrhythmia is a life-threatening cardiovascular disorder characterized by irregular heart rhythms, requiring accurate and early detection for effective diagnosis and treatment. Traditional electrocardiogram (ECG) analysis methods often struggle with noise interference, feature extraction limitations, and reduced classification accuracy in real-world clinical environments. To address these challenges, this study proposes a Wavelet-Enhanced Neural Framework for accurate detection of cardiac arrhythmia by integrating discrete wavelet transform (DWT) for signal denoising and feature enhancement with deep neural network-based classification. The wavelet transform decomposes ECG signals into multi-resolution frequency components, enabling effective extraction of clinically significant features, while the neural network performs automated pattern recognition and arrhythmia classification. The proposed framework improves robustness against noise, enhances feature representation, and increases detection accuracy compared to conventional machine learning and signal processing approaches. Experimental evaluation demonstrates that the hybrid wavelet-neural model achieves superior performance in terms of accuracy, sensitivity, specificity, and F1-score. The results confirm that the proposed method is highly effective for real-time, reliable, and automated cardiac arrhythmia detection in healthcare monitoring systems.