Traditional Chinese medicine auscultation and diagnosis syndrome differentiation analysis method, system, electronic device and medium

Abstract

The application discloses a traditional Chinese medicine auscultation and diagnosis method and system, electronic equipment and medium, and relates to the technical field of traditional Chinese medicine auscultation and diagnosis, and comprises the following steps: S100, utilizing a trained audio boundary regression model to preprocess acquired voice and cough audio data, so as to locate and extract effective audio signal segments; S200, inputting the effective audio signal segments into a two-way mixed expert network model to respectively output voice and cough symptom element classification results; S300, based on a preset qualitative and positioning weight mapping table between symptom elements and traditional Chinese medicine syndrome types, generating and outputting the final traditional Chinese medicine auscultation and diagnosis result through weighted calculation and rule matching of the obtained voice symptom element classification result and cough symptom element classification result. The defects of the prior art, such as difficulty in balancing the processing accuracy and efficiency of the auscultation audio, inability to explicitly model the common symptom characteristics between different sound modalities, and non-transparent syndrome differentiation decision process, are effectively overcome.

Images