QRS wave starting point end point positioning method based on regularized least square regression learning

Abstract

A QRS wave starting point end point positioning method based on regularized least square regression learning comprises the following steps: building a series of detection methods, detection criterions and threshold parameters according to function quadrature decomposition idea in the Hilbert space and the regularized least square regression learning algorithm, thus finally detecting the QRS wave and positioning the starting point and end point; carrying out inner product operation on an electrocardiosignal and a gauss function first order derived function so as to obtain an inner product sequence; selecting the reproducing kernel Hilbert space formed by the gauss kernel function as the approximation space of the regularized least square regression learning algorithm; using a real symmetric matrix square root decomposition method solving algorithm to obtain the novel inner product sequence.

Description

technical field [0001] The invention relates to the location measurement and analysis of the starting point and the ending point of the QRS wave in the dynamic electrocardiogram and the conventional electrocardiogram, specifically a method for locating the starting point and the ending point of the QRS wave based on regularized least squares regression learning, belonging to the technical field of electrocardiogram (ECG) automatic diagnosis and analysis. Background technique [0002] Electrocardiogram is one of the most widely used examination methods in clinical practice. It can help diagnose arrhythmia, myocardial ischemia, myocardial infarction, determine the location of infarction, determine the impact of drugs or electrolytes on the heart, and help implant artificial cardiac pacemakers, etc. . In addition, the tracing electrocardiogram is easy to operate and low in cost, and is especially suitable for use in areas with underdeveloped medical conditions. [0003] At pre...

AI Technical Summary

Problems solved by technology

At present, most of the mainstream methods for locating the start and end points of QRS waves on the market are only applicable to a certain type or several types of QRS waves. Figure 1.2 ), severe baseline drift ( Figure 1.3 ) and other waveforms, the analysis and positioning accuracy of QRS waves is low, and it is more difficult to make accurate analysis and judgment for waveforms with poor signal quality

Images