Unbalanced heart rate data set processing method and system based on generative adversarial network

• Main Authors: WANG BAOGUO, CONG HAIFANG, LAN TIANYU, ZHANG WEIHUA, ZHU YAODONG, YANG YANG, LI MINGQIU, HUANG XUPENG, LI FENGTIAN, JIANG SHUHUA, CAO XINYAN
• Format: Patent
• Language: Chinese; English
• Published: 03.05.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DIAGNOSIS; ELECTRIC DIGITAL DATA PROCESSING; HUMAN NECESSITIES; HYGIENE; IDENTIFICATION; MEDICAL OR VETERINARY SCIENCE; PHYSICS; SURGERY

The invention belongs to the technical field of medical data processing, and discloses an unbalanced heart rate data set processing method and system based on a generative adversarial network. According to the method, a multilayer wavelet transform decomposition method is adopted, baseline drift and noise of heart rate signals are filtered out, and heart rate data are converted from one-dimensional data to low-latitude representation data; performing reversible mapping on high-dimensional data and low-dimensional representation data of the electrocardiosignals by using the constructed CECG-GAN neural network model, and training the CECG-GAN neural network model to obtain a synthetic data set for expanding scarce samples; and judging whether the obtained synthetic data set is an effective sample or not by combining the accuracy rate, the recall rate and the F1-score value of the classification judgment model, and storing the effective sample. The high synthesis quality and diversity of the CECG-GAN neural netw