Improved Convolutional Neural Network and Its Training Method for Identifying Heart Rhythm Types

Abstract

This application relates to an improved convolutional neural network and its training method for identifying heart rhythm types. ECG data known as atrial fibrillation or non-atrial fibrillation is used as a training database to train the neural network. The convolutional neural network consists of several convolutional neural networks. Layer and several pooling layers, a fully connected layer and a classifier layer, the loss function used is: using this loss function can make the convolutional neural network increase the penalty for false negatives during training, so as to achieve guaranteed accuracy Under the premise of reducing the missed detection rate and improving accuracy.

Description

technical field [0001] The present application belongs to the technical field of electrocardiogram processing, and in particular relates to an improved convolutional neural network for identifying heart rhythm types and a training method thereof. Background technique [0002] An electrocardiogram (ECG) is a graph formed by recording the changes in the electrical activity of the heart every cardiac cycle from the body surface. A variety of heart diseases in humans can be characterized by an electrocardiogram. Among them, atrial fibrillation (abbreviated as atrial fibrillation) is the most common sustained arrhythmia. There is no unified classification of atrial fibrillation at present, and the identification of atrial fibrillation is also complicated. [0003] Convolutional Neural Networks (CNN) is a type of Feedforward Neural Networks (Feedforward Neural Networks) that includes convolution calculations and has a deep structure. In recent years, the use of neural networks ...

AI Technical Summary

Problems solved by technology

Han Xiaocen's master's thesis "Atrial Fibrillation Detection Based on Atrial Activity Characteristics and Convolutional Neural Networks" also researched the method of using convolutional neural networks for atrial fibrillation identification. The neural network is directly applied to the identification of atrial fibrillation without any optimization for the identification of atrial fibrillation, which leads to the inability to further improve the accuracy of the identification of atrial fibrillation in the electrocardiogram

Images