Abstract: In this study, we propose an optimized approach for electrocardiogram (ECG) classification, leveraging wavelet decomposition and convolutional neural networks (CNNs). Through wavelet decomposition, we extract informative features from ECG signals, which are then fed into a CNN model for accurate classification. Our results demonstrate improved performance in ECG classification, showcasing the efficacy of our optimized methodology. We utilized a comprehensive dataset to benchmark our approach against traditional methods, achieving superior accuracy, sensitivity, and specificity. This paper also discusses the potential clinical implications of our method, emphasizing its robustness in handling noisy and complex ECG signals. The proposed method holds promise for real-time medical diagnostics and automated healthcare solutions.

Keywords: ECG classification, wavelet decomposition, convolutional neural network, signal processing, deep learning.

| DOI: 10.17148/IARJSET.2024.11637