ECG Signal Classification Using 1D ResNet-18 with Integrated CBAM and Auxiliary Classifier

Electrocardiogram (ECG) classification is essential for the early diagnosis of heart diseases. In this study, we propose an enhanced method using a one-dimensional ResNet-18 model combined with a Convolutional Block Attention Module (CBAM) and an auxiliary classifier, both commonly applied in computer vision tasks, to process 1D ECG data. To address class imbalance, we applied ADASYN for data balancing and implemented data augmentation techniques on the MIT-BIH Arrhythmia Database. Experimental results demonstrate that the proposed method achieves significant improvements in classification accuracy compared to other approaches. The combination of CBAM and auxiliary classifier, along with balanced and augmented data, provides a robust solution for accurate ECG classification, potentially contributing to better diagnostic tools in healthcare.