Diagnosis of sleep disorders in traditional Chinese medicine based on adaptive neuro-fuzzy inference system

• DOI: 10.1016/j.bspc.2021.102942
• Publication Date: 2021-07-15T10:34:53Z
• Authors: Jianfang Wu; Ruo Hu; Ming Li; Shanshan Liu; Xizheng Zhang; Jun He; Jiaxu Chen; Xiangjun Li
• Keywords: Objectification; Palpation

Traditional Chinese medicine (TCM) relies on a combination of the four diagnostic methods inspection, listening and smelling, inquiry, palpation to diagnose sleep disorders. This method doctor's practice experience his mastery TCM theory is subjective in nature, it necessary make diagnosis objective. We propose using an Adaptive neuro-fuzzy inference system (ANFIS) disorders TCM, with adaptive reasoning capabilities objectivity. Sleep disorder symptoms are first compressed by genetic algorithm based mutual information. According process model has inputs: inspection(A), smelling(O), inquiry(F), palpation(P). quantified linguistic variables A, O, F, P fed them into model, which was trained generate membership each input variable, used derivation rules form output complete diagnosis. After testing, accuracy proposed 97.6%. In test comparison six doctors more than 10 years experience, our tested correctly all 100 medical cases. The diagnoses were consistent correct for 92 cases, however, there 8 cases where their diverged, indicating that ability objective reasoning. Our work provides reference objectification