ECG Signal Classification Algorithm Based on CCA-DSC ResGNet

With the widespread application of deep learning in biomedical signal processing, electrocardiogram (ECG) signal classification methods have continued to evolve. However, existing approaches still face challenges such as high computational complexity and insufficient capability to capture long-range dependencies. This paper proposes an ECG classification algorithm based on criss-cross attention and depthwise separable convolution residual gated network (CCA-DSC ResGNet). The proposed method leverages criss-cross attention (CCA) to enhance global information modeling, employs depthwise separable convolution (DSC) to reduce computational complexity, and integrates residual network (ResNet) to mitigate gradient vanishing issues. Additionally, gated recurrent units (GRU) are incorporated to model temporal dependencies, thereby improving the classification performance of abnormal heartbeats. Experimental results on the MIT-BIH Arrhythmia Database demonstrate that the proposed method outperforms existing state-of-the-art approaches in terms of classification accuracy and F1-score, validating its effectiveness and feasibility.