Electronic heart-lung sound data annotation management method and device

Abstract

The invention discloses an electronic heart-lung sound data annotation management method and device. The method comprises the following steps: conducting screening processing in a heart-lung sound annotation system on the basis of input screening information to obtain to-be-annotated heart-lung audio data; playing the to-be-annotated heart-lung audio data in the heart-lung sound annotation system,and acquiring an audiogram and a frequency spectrogram during playing; selecting to-be-annotated heart-lung audio data slots which need to be annotated in the audiogram and the frequency spectrogram,and conducting annotation and evaluation to obtain annotated and evaluated heart-lung audio data slots; and downloading the annotated and evaluated heart-lung audio data slots to a local client on the basis of the HTTP protocol. In the electronic heart-lung sound data annotation management method and device, a user can conveniently conduct visual heart-lung sound pre-processing, so that the accuracy of auscultation and judgement of respiratory cycles, cardiac cycles and abnormal heart-lung sound by a doctor is greatly improved; and by means of unified standardized audio annotation, the user can conveniently conduct unified terminology processing on mass data, so that the problem of disordered audio data labels is solved.

Description

technical field [0001] The invention relates to the technical field of medical data management, in particular to a method and device for labeling and managing electronic heart-lung sound data. Background technique [0002] The existing labeling management of heart and lung sounds is quite chaotic. In the case of mass audio management, it is difficult for users to quickly find out which part of the audio is abnormal. Auscultation discrimination is limited by subjective factors and auscultation techniques, and misjudgment is prone to occur. When analyzing big data audio, standard machine learning samples cannot be provided. Contents of the invention [0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides a method and device for labeling and managing electronic heart-lung sound data, which can facilitate users to perform visualized heart-lung sound preprocessing, and greatly improve doctors' understanding...

AI Technical Summary

Problems solved by technology

[0002] The existing labeling management of heart and lung sounds is quite chaotic. In the case of mass audio management, it is difficult for users to quickly find out which part of the audio is abnormal. Auscultation discrimination is limited by subjective factors and auscultation techniques, and misjudgment is prone to labeling

In big data audio analysis, standard machine learning samples cannot be provided

Images