基于证素辨证原理的"症状-证素"诊断权值归一化的研究与思考OA

Objective To resolve the inconsistency between varying summed"symptom-pattern element"contribution scores and a fixed diagnostic threshold in syndrome element differentiation,this study explored the effect of normalizing"symptom-pattern element"diagnostic weights on pattern element identification before and after normalization,using coronary heart disease as the investigative model.Methods Clinical cases of coronary heart disease treated with traditional Chinese medicine were collected and processed via the"Ancient and Modern Medical Case Cloud Platform".Syndrome element scores were normalized through proportional scaling,and the accuracy of syndrome element identification was compared before and after normalization.Results Coronary heart disease involved 207 symptoms,including chest tightness,wiry pulse,shortness of breath,and others.A total of 38 syndrome elements were identified,among which 13 accounted for more than 10％of cases.The primary disease-location syndrome elements were the heart,heart spirit,liver,and spleen,while the main disease-nature syndrome elements included qi deficiency,blood stasis,yin deficiency,and qi stagnation.After normalizing the diagnostic weights of"symptom-syndrome element"relationships using the proportional scaling method based on 300 renowned medical case records of coronary heart disease,the diagnostic consistency of the syndrome elements phlegm,heart,blood deficiency,and heart spirit with the original medical diagnoses improved compared to the original syndrome element integral pattern differentiation results.Conclusion The normalization of diagnostic weights for syndrome elements enhance the output performance of imbalanced syndrome elements and improve the accuracy of pattern differentiation for syndrome elements with lower total integral values.