A study aimed at addressing the rising prevalence of insomnia and gastralgia has developed a machine learning-based diagnostic model for Liver-Spleen Disharmony Syndrome (LSDS), which is significant in Traditional Chinese Medicine (TCM). The research, involving 575 patients, focused on accurate syndrome differentiation essential for effective treatment.
Read More
Participants were evaluated using structured scales that included demographic information and clinical symptoms related to LSDS. The study employed six machine learning algorithms: Chi-square Automatic Interaction Detector (CHAID), C5.0 decision tree, Back-Propagation Neural Network (BPNN), Radial Basis Function (RBF) network, Bayesian Network (BN), and Binomial Logistic Regression Analysis (BLRA) to create diagnostic models. The models' performance was assessed using various metrics, including receiver operating characteristic (ROC) curves.
Results revealed that the CHAID model outperformed the others, achieving an area under the ROC curve (AUC) of 0.889, an accuracy of 96%, and a sensitivity rate of 97%, marking it as a promising tool for assisting clinicians with LSDS diagnosis. The study identified depression or irritability as the most significant symptom variable in predicting LSDS, followed by epigastric fullness and various gastrointestinal symptoms.
The study underscores the potential of machine learning in TCM, illustrating its capability to provide more standardized and objective diagnoses. However, the authors noted that independent external validation is required to further assess the model's clinical applicability before widespread use.