Intelligent prescription recommendation has become an important research direction in traditional Chinese medicine (TCM), offering new opportunities to support clinical decision-making and promote the modernization of TCM practice. With the rapid development of artificial intelligence (AI), a variety of computational approaches have been proposed to learn prescription patterns from clinical data and generate personalized treatment recommendations. However, despite increasing research activity, systematic and comprehensive reviews of AI-driven methods for TCM prescription recommendation remain limited. In this study, we present a comprehensive review of computational approaches for herbal prescription recommendation (HPR) in TCM. Existing methods are systematically categorized into several major paradigms, including traditional machine learning methods, topic model methods, sequential generative methods, deep learning and graph-based methods, and large language model–based frameworks. In addition to summarizing methodological developments, we also review commonly used public datasets and evaluation metrics in this field. Furthermore, representative models with publicly available implementations are experimentally evaluated on multiple benchmark datasets to provide a comparative analysis of their performance on the HPR task. Finally, we discuss the key challenges that hinder the practical deployment of intelligent prescription recommendation systems, including data heterogeneity, limited interpretability, and insufficient integration of TCM domain knowledge. Future research directions are outlined to facilitate the development of more reliable, interpretable, and clinically applicable AI-assisted HPR systems for TCM.

OBJECTIVE To reveal the pharmacological mechanism of Sofren Injection in the treatment of Qi deficiency and blood stasis syndrome of angina pectoris in coronary heart disease,and to preliminarily verify the reliability of the prediction results by cell experiments.METHODS Firstly,we screened the main chemical components of Sofren injection and their targets from biomedical databases and literature.Then,using the DIAMOnD algorithm,we constructed the angina disease module by screening Qi deficiency and blood stasis syndrome-related genes from the GeneCards and MalaCards databases.Next,we conducted gene functional enrichment analysis of the core targets in the"angina pectoris-Qi deficiency and blood stasis-Sofren Injection"network to identify key pathways.Finally,we performed cell experiments to verify the effect of Sofren Injection on the expression of key pathway proteins in hypoxic H9C2 cardiomyocytes.RESULTS We identified 7 main chemical components of Sofren Injection,targeting a total of 362 genes.We screened 232 known angina pectoris-related genes and added 100 predicted genes by constructing the angina pectoris dis-ease module.A total of 2 960 genes related to Qi deficiency and blood stasis syndrome were obtained.Network topological analysis re-vealed 30 core targets for Sofren Injection in treating coronary heart disease angina pectoris with Qi deficiency and blood stasis syn-drome,including STAT3,EGFR,TNF,and IL-6.Gene functional enrichment analysis identified 82 pathways.Literature analysis combined with the results indicated that STAT3 and the JAK2/STAT3 pathway might be key pathways for Sofren Injection in treating coronary heart disease angina pectoris with Qi deficiency and blood stasis syndrome.Cell experimental results showed significant de-creases in mRNA and protein expression of JAK2 and STAT3 in the SI group and the nicorandil group compared to the model group(P<0.05).CONCLUSION Disease module analysis and cell experiments confirm that STAT3 is a key gene in the pathological mecha-nism of coronary heart disease angina pectoris with Qi deficiency and blood stasis syndrome,and the JAK2/STAT3 pathway is a core pathway for Sofren Injection in treating this condition.This study demonstrates the effectiveness and novelty of combining disease mod-ule for mining the treatment of TCM formulas with specific disease and syndrome.

Objective To develop and evaluate a fine-tuned large language model(LLM)for traditional Chinese medicine(TCM)prescription recommendation named TCMLLM-PR.Methods First,we constructed an instruction-tuning dataset containing 68 654 samples(ap-proximately 10 million tokens)by integrating data from eight sources,including four TCM textbooks,Pharmacopoeia of the People's Republic of China 2020(CHP),Chinese Medicine Clinical Cases(CMCC),and hospital clinical records covering lung disease,liver disease,stroke,diabetes,and splenic-stomach disease.Then,we trained TCMLLM-PR using Chat-GLM-6B with P-Tuning v2 technology.The evaluation consisted of three aspects:(i)compari-son with traditional prescription recommendation models(PTM,TCMPR,and PresRecST);(ii)comparison with TCM-specific LLMs(ShenNong,Huatuo,and HuatuoGPT)and general-domain ChatGPT;(iii)assessment of model migration capability across different disease datasets.We employed precision,recall,and F1 score as evaluation metrics.Results The experiments showed that TCMLLM-PR significantly outperformed baseline models on TCM textbooks and CHP datasets,with F1@10 improvements of 31.80%and 59.48%,respectively.In cross-dataset validation,the model performed best when migrating from TCM textbooks to liver disease dataset,achieving an F1@10 of 0.155 1.Analysis of real-world cases demonstrated that TCMLLM-PR's prescription recommendations most closely matched actual doctors'prescriptions.Conclusion This study integrated LLMs into TCM prescription recommendations,leverag-ing a tailored instruction-tuning dataset and developing TCMLLM-PR.This study will pub-licly release the best model parameters of TCMLLM-PR to promote the development of the decision-making process in TCM practices(https://github.com/2020MEAI/TCMLLM).

In 2009,the World Health Organization included snakebite on the list of neglected tropical diseases,acknowledging it as a common occupational hazard for farmers,plantation workers,and others,causing tens of thousands of deaths and chronic physical disabilities every year.This guideline aims to provide practical information to help clinical professionals evaluate and treat snakebite victims.These recommendations are based on clinical experience and clinical research evidence.This guideline focuses on the following topics:snake venom,clinical manifestations,auxiliary examination,diagnosis,treatments,and prevention.

OBJECTIVE To reveal the pharmacological mechanism of Sofren Injection in the treatment of Qi deficiency and blood stasis syndrome of angina pectoris in coronary heart disease,and to preliminarily verify the reliability of the prediction results by cell experiments.METHODS Firstly,we screened the main chemical components of Sofren injection and their targets from biomedical databases and literature.Then,using the DIAMOnD algorithm,we constructed the angina disease module by screening Qi deficiency and blood stasis syndrome-related genes from the GeneCards and MalaCards databases.Next,we conducted gene functional enrichment analysis of the core targets in the"angina pectoris-Qi deficiency and blood stasis-Sofren Injection"network to identify key pathways.Finally,we performed cell experiments to verify the effect of Sofren Injection on the expression of key pathway proteins in hypoxic H9C2 cardiomyocytes.RESULTS We identified 7 main chemical components of Sofren Injection,targeting a total of 362 genes.We screened 232 known angina pectoris-related genes and added 100 predicted genes by constructing the angina pectoris dis-ease module.A total of 2 960 genes related to Qi deficiency and blood stasis syndrome were obtained.Network topological analysis re-vealed 30 core targets for Sofren Injection in treating coronary heart disease angina pectoris with Qi deficiency and blood stasis syn-drome,including STAT3,EGFR,TNF,and IL-6.Gene functional enrichment analysis identified 82 pathways.Literature analysis combined with the results indicated that STAT3 and the JAK2/STAT3 pathway might be key pathways for Sofren Injection in treating coronary heart disease angina pectoris with Qi deficiency and blood stasis syndrome.Cell experimental results showed significant de-creases in mRNA and protein expression of JAK2 and STAT3 in the SI group and the nicorandil group compared to the model group(P<0.05).CONCLUSION Disease module analysis and cell experiments confirm that STAT3 is a key gene in the pathological mecha-nism of coronary heart disease angina pectoris with Qi deficiency and blood stasis syndrome,and the JAK2/STAT3 pathway is a core pathway for Sofren Injection in treating this condition.This study demonstrates the effectiveness and novelty of combining disease mod-ule for mining the treatment of TCM formulas with specific disease and syndrome.

Objective To develop and evaluate a fine-tuned large language model(LLM)for traditional Chinese medicine(TCM)prescription recommendation named TCMLLM-PR.Methods First,we constructed an instruction-tuning dataset containing 68 654 samples(ap-proximately 10 million tokens)by integrating data from eight sources,including four TCM textbooks,Pharmacopoeia of the People's Republic of China 2020(CHP),Chinese Medicine Clinical Cases(CMCC),and hospital clinical records covering lung disease,liver disease,stroke,diabetes,and splenic-stomach disease.Then,we trained TCMLLM-PR using Chat-GLM-6B with P-Tuning v2 technology.The evaluation consisted of three aspects:(i)compari-son with traditional prescription recommendation models(PTM,TCMPR,and PresRecST);(ii)comparison with TCM-specific LLMs(ShenNong,Huatuo,and HuatuoGPT)and general-domain ChatGPT;(iii)assessment of model migration capability across different disease datasets.We employed precision,recall,and F1 score as evaluation metrics.Results The experiments showed that TCMLLM-PR significantly outperformed baseline models on TCM textbooks and CHP datasets,with F1@10 improvements of 31.80%and 59.48%,respectively.In cross-dataset validation,the model performed best when migrating from TCM textbooks to liver disease dataset,achieving an F1@10 of 0.155 1.Analysis of real-world cases demonstrated that TCMLLM-PR's prescription recommendations most closely matched actual doctors'prescriptions.Conclusion This study integrated LLMs into TCM prescription recommendations,leverag-ing a tailored instruction-tuning dataset and developing TCMLLM-PR.This study will pub-licly release the best model parameters of TCMLLM-PR to promote the development of the decision-making process in TCM practices(https://github.com/2020MEAI/TCMLLM).

Objective:To explore the therapeutic characteristics of population with gout achieving treat-to-target (T2T) indicators through real-world research and evaluate their safety.Methods:A total of 3 287 patients diagnosed with gout by rheumatologists in 21 first-class tertiary hospitals in 10 provinces, municipalities, and autonomous regions in China from January 2015 to December 2021 were included in this polycentric cross-sectional study. The database included patients′ general information, disease characteristics, and clinical application of traditional Chinese and Western medicine treatment measures. SPSS and Excel software were used for data analysis. Frequency analysis, cluster analysis, and factor analysis were used to summarize the characteristics and rules of treatment measures for patients with gout who achieved the target after treatment. The occurrence of adverse events (AE) was recorded during treatment.Results:After treatment, 691 visits (7%) achieved the serum urate (SUA) target, and the most frequent use of urate-lowering therapy (ULT) was febuxostat, followed by benzbromarone. The most common treatment options were following: GroupⅠ: traditional Chinese medicine (TCM) decoction-TCM external treatment-physical exercise-proprietary Chinese medicine; GroupⅡ: ferulic acid-nonsteroidal anti-inflammatory drugs (NSAIDs); Group Ⅲ: allopurinol-sodium bicarbonate-benzbromarone; Group Ⅳ: glucocorticoid-colchicine; Group Ⅴ: febuxostat. A total of 5 898 visits (60%) chieved manifestations of joint pain VAS scores target, and the most frequently used drug to control joint symptoms was NSAIDs. The frequency of use of drugs to control joint symptoms were 2 118 times (usage rate reached 35.9%), while the frequency of ULT were 2 504 times (usage rate reached 42.5%), which was higher than the joint symptom control drug. The most common treatment options were following: Group Ⅰ: proprietary Chinese medicine-TCM decoction-TCM external treatment-physical exercise; Group Ⅱ: NSAIDs-colchicine hormones; Group Ⅲ: allopurinol, Group Ⅳ: benzbromarone; Group Ⅴ: febuxostat. A total of 59 adverse events occurred during treatment.Conclusion:The proportions of gout patients who reach target serum urate level & good control of joint symptoms are both very low, and ULT and anti-inflammatory prescription patterns are very different from international guidelines, so it is necessary to strengthen the standardized management of gout patients. At the same time, life intervention measures account for a certain proportion of the treatment plans for the T2T population, and further exploration is needed.

Objective:To investigate the risk factors for different types of single subcortical infarction (SSI) in middle cerebral artery territory and the risk factors for early neurologic deterioration (END).Methods:Patients with SSI in middle cerebral artery territory admitted to the Department of Neurology, People's Hospital Affiliated to Jiangsu University from January 2020 to April 2021 were enrolled retrospectively. According to the distribution of infarction, the patients were divided into proximal SSI (pSSI) and distal SSI (dSSI). The demographics, vascular risk factors and baseline clinical data were collected. END was defined as new signs and/or symptoms of neurological deficit or aggravation of any neurological deficit within 2 weeks after onset. Multivariate logistic regression analysis was used to determine the independent risk factors for pSSI and END. Results:Seventy-six patients with acute SSI in the middle cerebral artery territory were included, 41 patients (53.9%) in the pSSI group, 35 (46.1%) in the dSSI group; 13 (17.1%) in the END group, and 63 (82.9%) in the non-END group. There were no significant differences in age, gender, vascular risk factors and baseline National Institutes of Health Stroke Scale score between the pSSI group and the dSSI group. The total cholesterol, fasting blood glucose levels and the ratio of pSSI in the END group were significantly higher than those in the non-END group ( P<0.05), while the high-density lipoprotein cholesterol level was significantly lower than that of the non-END group ( P<0.05). Multivariate logistic regression analysis showed that pSSI was an independent risk factor for the occurrence of END in patients with SSI (odds ratio 6.75, 95% confidence interval 1.26-36.23; P=0.026). Conclusion:There was no significant difference in risk factors between pSSI and dSSI, but patients with pSSI were more prone to END.

Objective:To describe the incidence, diagnosis rate, treatment rate and treatment pattern of hyperkalemia, and serum potassium retesting rate among hyperkalemia patients in the emergency department.Methods:Data were derived from Military Data Center for Rational Use of Drugs. Patients who accessed emergency medical services (≥18 years old) with record(s) of serum potassium between 2015 and 2017 were included. The data of laboratory test, diagnosis, and treatment were analyzed. The main outcomes included the incidence of hyperkalemia, the diagnosis rate, the treatment rate, treatment pattern and the 7-day retesting rate.Results:A total of 1 039 245 patients who met the above criteria were included, of whom, 36 615 (3.52%) had at least one hyperkalemia event. Among the emergency patients with chronic kidney disease, heart failure, diabetes mellitus and hypertension, the proportions of patients who experienced hyperkalemia were 47.69%, 29.13%, 21.69% and 10.16%, respectively. The diagnosis rate of emergency hyperkalemia patients was 9.23%. The overall hyperkalemia treatment rate was 42.1%. Insulin + glucose injection was the most commonly used therapy for emergency hyperkalemia patients. The overall serum potassium retesting rate within 7 days was 28.8%.Conclusions:Hyperkaliemia is more common and more severe in patients with chronic kidney disease, heart failure, diabetes and hypertension. The diagnosis rate and retesting rate of hyperkalemia are low, suggesting that the identification and management of hyperkaliemia in emergency patients should be strengthened.

The clinical pathological data of a patient large cell undifferentiated bladder carcinoma was retrospectively analyzed and understand.The clinical and imaging findings of large cell undifferentiated bladder carcinoma was nonspecific.Diagnosis depended on the pathological and immuno-histochemical staining.The tumor is aggressive with high risk of recurrence.It is mainly treated with radical resection.