A Multi-dimensional Diagnostic Research Path for Syndromes Based on the Combination of AI"Macro micro"Perspectives:A Case Study of Non-alcoholic Fatty Liver Disease
- VernacularTitle:基于AI"宏-微"观结合的证候多维诊断研究路径——以非酒精性脂肪性肝病为例
- Author:
Caiying HE
1
;
Baixue LI
;
Ju CHEN
;
Hang ZHOU
;
Dong WANG
Author Information
- Publication Type:Journal Article
- Keywords: "Macro-micro"integration; Syndrom; Non-alcoholic fatty liver disease; Machine learning; Interpretable model
- From: World Science and Technology-Modernization of Traditional Chinese Medicine 2025;27(11):3157-3171
- CountryChina
- Language:Chinese
- Abstract: With the advancement of modern research methods,machine learning(ML)algorithms have been widely applied in traditional Chinese medicine(TCM)diagnosis,transforming subjective syndrome differentiation into a more objective process,thereby providing a feasible pathway for the objectification and quantification of TCM diagnostics.However,challenges persist,including the dual"black-box"nature of disease-syndrome models(lacking interpretability),missing spatiotemporal dynamic data,and the"disconnect"between clinical phenotyping and molecular biomarker research.Focusing on non-alcoholic fatty liver disease(NAFLD),this study proposes a novel framework guided by the biological"multiple-hit"theory and TCM's"disease-syndrome-symptom-stage"approach,which involves:Constructing a mathematical model of NAFLD progression(simple steatosis→steatohepatitis→fibrosis→cirrhosis)via multi-level(phenotypic-cellular-molecular)network modules;And developing an interpretable multidimensional model integrating syndrome-imaging phenomics(macro)and metabolomics-derived biomarkers(micro)to enable personalized NAFLD diagnosis.
