ECG signal noise pollution degree and category intelligent assessment method

Abstract

The invention provides an ECG signal noise pollution degree and category intelligent assessment method. The method comprises the following steps: S1, carrying out ECG signal preprocessing; S2, performing mathematical statistical calculation on ECG signal component characteristics; S3, carrying out ECG signal noise pollution degree judgment; S4, judging the low-frequency baseline drift noise pollution degree; S5, judging the high-frequency component noise pollution type and pollution degree; and S6, comprehensively evaluating the overall noise pollution degree and category of the ECG signal. According to the evaluation method, M-CEEMD signal decomposition and an SVC network cluster are combined, ECG signal noise pollution degree assessment and pollution type automatic classification are accurately completed, screening of the available ECG signals is achieved, a basis is provided for selecting a targeted filtering mode, original ECG data information is reserved to the maximum extent while the filtering effect is improved, the ECG intelligent diagnosis false positive rate is reduced, meanwhile, the diagnosis accuracy is improved, and the ECG intelligent diagnosis method is high in clinical generalization and classification accuracy.

Description

technical field [0001] The present invention relates to the field of artificial intelligence technology and smart medical technology, in particular to an intelligent evaluation method for ECG signal noise pollution degree and category Background technique [0002] The monitoring and analysis of ECG signals is the main means to reduce the mortality of cardiovascular diseases. According to the World Health Organization report in 2017, cardiovascular diseases (CVD) are the first cause of death in the world. According to statistics, the number of cardiovascular patients in my country is as high as 290 million, and its mortality rate accounts for more than 40% of the causes of death among residents, which is much higher than that of tumors and other diseases. With the aging process in our country, the morbidity and mortality of CVD are also increasing year by year. Since CVD diseases have the characteristics of rapid deterioration of chronic diseases in a short period of time, l...

AI Technical Summary

Problems solved by technology

However, most of the existing filtering technologies can only obtain better filtering effects for specific noises. However, if the current signal noise pollution degree and pollution type are unknown, direct filtering processing will not only fail to obtain good filtering results, but may even cause ECG The waveform of the signal itself is deformed, which further deteriorates the intelligent diagnosis results of the ECG signal

Images