METHOD AND SYSTEM FOR RECOGNIZING CRACKLES

The present disclosure provides a method for recognizing crackles, including: processing collected lung sound signal to extract moist rale component for a respiratory cycle; calculating a power spectrum of the moist rale component, and according to the power spectrum, calculating at least one of a ratio of power of each preset frequency band in a plurality of preset frequency bands to total power of all preset frequency bands and the total power of all the preset frequency bands and selecting at least one of them as a frequency domain parameter, and/or calculating at least one of a ratio of the number of occurrence of moist rale in the late inspiratory phase to the total number of occurrence of moist rale in the entire inspiratory phase and a maximum amplitude of moist rale in the entire inspiratory phase and selecting at least one of them as a time domain parameter; inputting the frequency domain parameter and/or the time domain parameter serving as a parameter feature into a classification model for recognition. The present disclosure further provides a system for recognizing crackles. The method and the system can improve the accuracy for recognizing crackles.