Heart sound analysis and labeling method and system

Abstract

The invention provides a heart sound analysis and labeling method and system. The method comprises the steps of obtaining a heart sound map through conversion of an imported heart sound signal sample,wherein the heart sound map comprises a phonocardiogram and a heart sound frequency response image; and playing the heart sound map, obtaining input label information, and labeling a heart sound analysis result in a corresponding interval range. According to the method, the imported heart sound sample is converted into the phonocardiogram and the heart sound frequency response image; the loudnessof a heart sound is judged by observing the time domain change trends and conditions of the heart sound frequency response image and the phonocardiogram; the division of the loudness color of the heart sound is observed, the normal heart sound and the abnormal heart sound are distinguished, and an imported audio file is heard in cooperation; through speed adjustment and time axis amplification, the occurrence interval of the abnormal heart sound and the change of the loudness are judged more accurately; and the interval of the abnormal heart sound is labeled and exported to form a sample database, so that later heart sound calculation and research analysis of a patient disease database are facilitated.

Description

technical field [0001] The invention relates to the technical field of medical testing, in particular to a method and system for analyzing and labeling heart sounds. Background technique [0002] Heart sounds are produced by vibrations caused by myocardial contraction, heart valve closure, and blood hitting the ventricular wall and aorta wall. It is one of the most important physiological signals of the human body. When some diseases occur in the human body, heart sounds will produce some important pathological Information, and these pathological information are characteristically reflected in many diseases, so observing the changes of heart sound signals can provide a certain basis for the judgment of symptoms. [0003] In the existing technology, most of the classification and recognition functions of heart sounds are only to distinguish the signals of normal and abnormal heart sounds, and in the segmentation process of heart sounds, the segmentation is usually performed b...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, most of the classification and recognition functions of heart sounds are only to distinguish the signals of normal and abnormal heart sounds, and in the segmentation process of heart sounds, the segmentation is usually performed by calculating a single threshold. This calculation method is easy to implement. However, the accuracy of the calculated results is relatively inaccurate, and the anti-interference ability to external noise is not strong, which will lead to serious errors in the segmentation of heart sounds, which will affect the patient's symptom judgment results.

Images