Method for quantitatively detecting microvolt T-wave alternans

Abstract

The invention discloses a method for quantitatively detecting microvolt T-wave alternans. A new effective quantitative TWA (T-wave alternans) detection index is provided by analyzing external shape of a T-wave alternans scatter diagram according to the relation of the external shape and the T-wave alternans. By means of improved T-wave window analysis method, the method includes: subjecting 128 continuous heartbeat T-wave sampling point sequences to first difference and drawing a scatter diagram, analyzing the relation of the external shape of the scatter diagram and the T-wave alternans, researching an effective 'cross searching method' to subject the scatter diagram to edge extraction, then defining three quantitative parameters of short axis, long axis and the ratio of the short axis to the long axis at the 'edge' of the scatter diagram, utilizing the ratio of the short axis to the long axis as quantitative detection index Axial_ratio, and finding a proper threshold to judge whether the T-wave alternans exist or not.

Description

technical field [0001] The invention relates to a detection method, in particular to a quantitative detection method for microvolt-level T wave alternation. The invention belongs to the technical field of biomedical signal processing. Background technique [0002] T wave alternans (TWave Alternans, TWA) refers to the beat-by-beat alternation of T wave amplitude, shape and even polarity on the body surface ECG when the heart rhythm is regular. Since Traube first discovered the phenomenon of electrical alternation in 1872, many studies have shown that TWA is related to myocardial ischemia, coronary artery disease, electrolyte disturbance and long QT syndrome. Research in the past 30 years has shown that TWA is a sign of myocardial electrical activity instability and is Independent, statistically significant indicator for predicting ventricular arrhythmia and sudden cardiac death. Compared with other non-invasive ECG detection methods, TWA has higher accuracy in predicting ar...

AI Technical Summary

Problems solved by technology

[0006] Chinese patent application 201210064226 discloses a method for alternately detecting T waves in an electrocardiogram. Correlation analysis is used to detect TWA signals from a time-domain perspective. Although it can dynamically track unsteady TWA phenomena, it has high requirements for signal input quality. Adjacent frequency noise is very sensitive, which can easily cause false detection and missed detection

The complexity of these methods is very high. Although the main application of WVD is high time-frequency concentration, the cross-interference is serious, the complexity is high, and the clarity and complexity are trade-offs. The large amount of calculation leads to poor real-time performance. In terms of the processing power of the hardware, there are great defects in practical applications.

[0008] Chinese patent application 200410003212.1 discloses a method and system for improving the measurement of T-wave alternans, a method for improving TWA detection by removing certain beats from the measurement of T-wave alternans, and only involves ECG signal preprocessing for TWA detection , did not propose a specific TWA detection method

Images