The combination of artificial intelligence and TCM and the bulk replication and promotion of the famous doctor treatment model will help solve the problem of scarce quality medical resources in TCM. The elements of syndromes are derived from the decomposition of syndromes, which is a representation of the etiology and pathogenesis of syndromes. On the one hand, it helps to reduce the dimensionality of the TCM data consistent with the symptoms, facilitating the standardisation of the data, and on the other hand, it supports flexible combinations by order, helping to guide the use of prescriptions in a targeted manner. The elements of syndromes have the characteristics of being superimposable and combinable in multiple dimensions such as etiology, disease nature and disease situation, and are suitable for mathematical calculations, which can transform the complex multi-dimensional treatment process of TCM into an artificial intelligence calculation process. This study provided an in-depth analysis of the reasons limiting the deep integration of TCM and artificial intelligence, and designs a TCM artificial intelligence-assisted diagnosis and treatment system for coronary artery disease based on the respective characteristics of the elements of syndromes and TCM to bridge the two fields of TCM and artificial intelligence, with a view to providing a reference for a breakthrough in the deep integration of TCM and artificial intelligence.

OBJECTIVE To systematically evaluate the efficacy and safety of Shenfu injection combined with chemical medicine in the treatment of coronary heart disease combined with heart failure. METHODS Retrieved from CNKI， CBM， VIP， Wanfang， PubMed， Embase and the Cochrane Library， randomized controlled trials （RCTs） about Shenfu injection combined with chemical medicine （trial group） versus chemical medicine （control group） in the treatment of heart failure with coronary heart disease were collected during the inception to August 2022. After literature screening and data extraction， the qualities of included literature were evaluated and rated by using Cochrane manual and GRADE system. Meta-analysis and Egger’s were performed with RevMan 5.3 software， and TSA 0.9.5.10 Beta software was used for trial sequential analysis. RESULTS Seventeen studies were included， with a total sample of 1 355 patients. The quality grade evidence of GRADE was all low. Meta-analysis showed that cardiac function efficacy [RR＝1.23， 95%CI （1.16，1.30）， P＜0.000 01]， the decrease of brain natriuretic peptide [MD＝－96.06， 95%CI （－116.47， －75.64）， P＜0.000 01] and the increase of left ventricular ejection fraction [MD＝5.32， 95%CI （4.03，6.60）， P＜0.000 01] in trial group were significantly better than control group； there was no statistical significance in the incidence of ADR between 2 groups [RR＝0.52，95%CI（0.22，1.22），P＝0.13]. The results of sequential analysis showed that the sample size included in this study met the requirements of meta-analysis； the results of Egger’s test showed that the results were robust and publication bias had no significant effect on the results. CONCLUSIONS Shenfu injection combined with chemical medicine in the treatment of coronary heart disease combined with heart failure can further improve the clinical symptoms and related indicators， and no serious adverse reaction is observed.

ObjectiveTo observe the effect of Xiaoke drink on insulin resistance in ob/ob mice and explore the mechanism. MethodEighteen ob/ob mice were randomly assigned into model， Xiaoke drink （17.68 g·kg^-1）， and atorvastatin （0.01 g·kg^-1） groups （n=6）， and six C57BL/6 mice were selected as the normal group. Mice in the normal and model groups were administrated with the same amount of distilled water. Fasting body weight， weekly food intake， and weekly water intake were measured at a fixed time. Fasting plasma glucose （FPG） and 2-hour post-load plasma glucose （2 hPG） were measured before and after 8-week intervention. After intervention， total cholesterol （TC）， triglyceride （TG）， fasting insulin （FINS）， Homeostasis Model Assessment-Insulin Resistance （HOMA-IR）， blood routine， and alkaline phosphatase （ALP） were measured. Western blot was employed to determine the expression levels of ubiquitin-specific protease 20 （USP20） and 3-hydroxy-3-methylglutaryl coenzyme A reductase （HMGCR） in the liver. The pancreas was stained with hematoxylin-eosin for observation. ResultCompared with the model group， the Xiaoke drink group showed decreased body weight of ob/ob mice （P<0.05， P<0.01）， declined growth trend of body weight （P<0.05， P<0.01）， reduced weekly average water intake， lowered levels of FPG， 2 hPG， TC， and HOMA-IR （P<0.05， P<0.01）， and down-regulated expression level of USP20 in the liver （P<0.05）. HMGCR content was positively correlated with USP20 expression. In addition， Xiaoke drink promoted the recovery of islet tissue morphology and function in ob/ob mice. ConclusionXiaoke drink can ameliorate insulin resistance in ob/ob mice by inhibiting USP20/HMGCR expression， reversing cholesterol biosynthesis process， and reducing cholesterol level.