The Self-discipline Learning Model with Imported Backpropagation Algorithm

The electrocardiogram (ECG) has been proved to be the most common and effective method of studying cardiovascular disease because it is simple, noninvasive, and inexpensive. However, the differences between ECG signals are difficult to distinguish. In this paper, a model combining convolutional neural networks (CNN) with self-discipline learning (SDL) is proposed to realize the classification and identification of cardiac arrhythmia data. Comparison with a variety of deep learning frameworks based on the MIT-BIH arrhythmia dataset shows that, this model achieves a higher level of accuracy with less structure.