Research on the construction of a medical cases recommendation model for Qin Bowei's diagnosis and treatment of liver disease based on semantic similarity
Objective To use AI(artificial intelligence)technology to empower the inheritance of the veteran TCM phycians Qin Bowei's clinical experience in diagnosis and treatment of liver disease,and to assist Chinese medicine clinical decision-making.Methods Based on semantic similarity,a recommendation model of Qin BoWei's liver disease diagnosis and treatment.Positive samples were established based on the data of Qin Bowei's liver disease-related medical cases,and negative samples were composed of unrelated liver disease cases.Long Short-Term Memory(LSTM)network was used to model the correlations between symptom clusters.The AdamW optimizer function was used to optimize the LSTM model through the back-propagation algorithm and the gradient of the model parameters.The symptom clusters were represented vectorically,the cosine similarity of the two syndromes was calculated,and the recommended results were output for clinicians'reference.Results The samples used for model training in this study included a total of 2,132 positive and negative samples,including 1,816 symptom descriptions.244 medical cases were tested,and the application scenarios of the model were displayed through random validation tests,and similar medical case recommendations were achieved.Conclusion The model constructed in this study has a good recommendation effect,which is helpful for the inheritance,promotion and application of the diagnosis and treatment experience of veteran TCM phycians.At the same time,it can serve as a valuable reference for clinical decision-making,and improve the level of TCM diagnosis and treatment.
Veteran TCM phyciansSemantic similarityLSTMMedical case recommendation
万方数据
维普
