A Hybrid Deep Learning Framework using CNN and Cascaded Discrete Wavelet Transform for ECG-Based Cardiovascular Disease Detection

Lakshmi Behra1, Dr. Saidabhi2 Sharon Elizabeth3, A.Sravani4, Saamiya Afreen5, K.Kaushik6
Student1,3-5, BTech (CSE) From Sphoorthy Engineering College, Hyderabad.
Assistant Professor2, Dep of CSE, Sphoorthy Engineering College, Hyderabad.

Abstract:Cardiovascular diseases (CVDs) remain one of the leading causes of mortality worldwide, making early and accurate diagnosis essential for effective treatment. Electrocardiogram (ECG) signals play a vital role in detecting cardiac abnormalities, but their complex patterns often require advanced automated analysis methods. To address this need, we propose a hybrid deep learning framework that combines Convolutional Neural Networks (CNNs) with a CascadedDiscrete Wavelet Transform (CadDWT) approach for enhanced ECG-based CVD classification. The CNN is used to extract spatial features, while the wavelet transform captures critical time–frequency information that traditional CNNs may overlook. Principal Component Analysis (PCA) is further applied to reduce dimensionality and improve model efficiency. Experimental results demonstrate that the proposed CNN–DWT hybrid model significantly outperformstraditional machine learning methods such as K-Nearest Neighbors (KNN) and Support Vector Machines (SVM), achieving superior accuracy, precision, recall, and F1-score. These findings highlight the effectiveness of integrating spatial and frequency-domain features for robust ECG analysis and underscore the potential of t e proposed framework as a reliable tool for automated cardiovascular disease detection. Keywords: ECG Signal Classification, Cardiovascular Disease Detection, Convolutional Neural Networks (CNN) AndDiscrete Wavelet Transform (DWT)