Traditional Chinese medicine (TCM) represents a paradigmatic approach to personalized medicine, developed through the systematic accumulation and refinement of clinical empirical data over more than 2000 years, and now encompasses large-scale electronic medical records (EMR) and experimental molecular data. Artificial intelligence (AI) has demonstrated its utility in medicine through the development of various expert systems (e.g., MYCIN) since the 1970s. With the emergence of deep learning and large language models (LLMs), AI's potential in medicine shows considerable promise. Consequently, the integration of AI and TCM from both clinical and scientific perspectives presents a fundamental and promising research direction. This survey provides an insightful overview of TCM AI research, summarizing related research tasks from three perspectives: systems-level biological mechanism elucidation, real-world clinical evidence inference, and personalized clinical decision support. The review highlights representative AI methodologies alongside their applications in both TCM scientific inquiry and clinical practice. To critically assess the current state of the field, this work identifies major challenges and opportunities that constrain the development of robust research capabilities-particularly in the mechanistic understanding of TCM syndromes and herbal formulations, novel drug discovery, and the delivery of high-quality, patient-centered clinical care. The findings underscore that future advancements in AI-driven TCM research will rely on the development of high-quality, large-scale data repositories; the construction of comprehensive and domain-specific knowledge graphs (KGs); deeper insights into the biological mechanisms underpinning clinical efficacy; rigorous causal inference frameworks; and intelligent, personalized decision support systems.
Medicine, Chinese Traditional/methods* ; Artificial Intelligence ; Humans ; Precision Medicine ; Decision Support Systems, Clinical

Advancements in artificial intelligence (AI) and emerging technologies are rapidly expanding the exploration of chemical space, facilitating innovative drug discovery. However, the transformation of novel compounds into safe and effective drugs remains a lengthy, high-risk, and costly process. Comprehensive early-stage evaluation is essential for reducing costs and improving the success rate of drug development. Despite this need, no comprehensive tool currently supports systematic evaluation and efficient screening. Here, we present druglikeFilter, a deep learning-based framework designed to assess drug-likeness across four critical dimensions: 1) physicochemical rule evaluated by systematic determination, 2) toxicity alert investigated from multiple perspectives, 3) binding affinity measured by dual-path analysis, and 4) compound synthesizability assessed by retro-route prediction. By enabling automated, multidimensional filtering of compound libraries, druglikeFilter not only streamlines the drug development process but also plays a crucial role in advancing research efforts towards viable drug candidates, which can be freely accessed at https://idrblab.org/drugfilter/.

Drug development encompasses multiple processes, wherein protein subcellular localization is essential. It promotes target identification, treatment development, and the design of drug delivery systems. In this research, a deep learning framework called LocPro is presented for predicting protein subcellular localization. Specifically, LocPro is unique in (a) combining protein representations from the pre-trained large language model (LLM) ESM2 and the expert-driven tool PROFEAT, (b) implementing a hybrid deep neural network architecture that integrates convolutional neural network (CNN), fully connected (FC) layer, and bidirectional long short-term memory (BiLSTM) blocks, and (c) developing a multi-label framework for predicting protein subcellular localization at multiple granularity levels. Additionally, a dataset was curated and divided using a homology-based strategy for training and validation. Comparative analyses show that LocPro outperforms existing methods in sequence-based multi-label protein subcellular localization prediction. The practical utility of this framework is further demonstrated through case studies on drug target subcellular localization. All in all, LocPro serves as a valuable complement to existing protein localization prediction tools. The web server is freely accessible at https://idrblab.org/LocPro/.

Objective To evaluate the association between self-reported neighborhood walkability environments and mortality in China.Methods The Prospective Urban Rural Epidemiology study in China(PURE-China)recruited 47 931 participants aged 35-70 from 12 provinces in China between 2005 and 2009.Neighborhood environmental indicators were collected using the Neighborhood Environment Walkability Scale(NEWS)questionnaire,with higher scores indicating better walkable environments.The primary outcomes were all-cause mortality and cardiovascular mortality,using Cox fragile model to evaluate the association between community walkability and outcomes,as well as exploring mediating pathways.Results Of 35 490 participants included in this study,60%were female,with a mean(SD)age of 51.5(9.6)years.The median follow-up was 11.7 years.This study found an association between higher community walkability score and reduced risk of all-cause mortality,with the total score(HR=0.85;95%CI,0.80-0.89),land-use mix(HR=0.84;95%CI,0.79-0.88),and crime safety(HR=0.84;95%CI,0.80-0.89)showing the most significant associations.NEWS can affect long-term adverse outcomes through lifestyle.Conclusions In the Chinese population,favorable community walkability is associated with lower all-cause mortality risk,which may support policymakers to take actions to mitigate the adverse effects of poor community en-vironments on health.

Objective To discuss the prediction value of the early efficacy of hepatic arterial infusion chemotherapy(HAIC)in treating stage Ⅱ-Ⅲ hepatocellular carcinoma(HCC).Methods The clinical data of 81 patients with stage Ⅱ-Ⅲ HCC,who received at least 3 times of HAIC at the Nanchang Municipal Central Hospital of China from November 2021 to March 2024,were retrospectively analyzed.CT or MRI was used to compare patient's local tumor response after each treatment cycle.Based on modified Response Evaluation Criteria in Solid Tumors(mRECIST),the curative effects of patients after receiving the first,the second,and the last HAIC treatment were compared between each other.The prediction value of the early efficacy of HAIC in treating patients with stage Ⅱ-Ⅲ HCC was analyzed.Results In the 67 patients,the efficacy of the last time HAIC was equal or similar to that of the first time HAIC,and in the remaining 14 patients the efficacy of the last time HAIC was different from that of the first time HAIC,with an efficacy prediction rate of 82.72％.The efficacy of the last time HAIC was equal or similar to that of the second time HAIC in 71 patients,and in the remaining 10 patients the efficacy of the last time HAIC was different from that of the second time HAIC,with an efficacy prediction rate of 87.65％.Conclusion In treating stage Ⅱ-Ⅲ HCC with HAIC,the early efficacy can be used to predict the final efficacy after completion of the total treatment course.

Drug development encompasses multiple processes,wherein protein subcellular localization is essential.It promotes target identification,treatment development,and the design of drug delivery systems.In this research,a deep learning framework called LocPro is presented for predicting protein subcellular localization.Specifically,LocPro is unique in(a)combining protein representations from the pre-trained large language model(LLM)ESM2 and the expert-driven tool PROFEAT,(b)implementing a hybrid deep neural network architecture that integrates convolutional neural network(CNN),fully connected(FC)layer,and bidirectional long short-term memory(BiLSTM)blocks,and(c)developing a multi-label framework for predicting protein subcellular localization at multiple granularity levels.Additionally,a dataset was curated and divided using a homology-based strategy for training and validation.Compar-ative analyses show that LocPro outperforms existing methods in sequence-based multi-label protein subcellular localization prediction.The practical utility of this framework is further demonstrated through case studies on drug target subcellular localization.All in all,LocPro serves as a valuable complement to existing protein localization prediction tools.The web server is freely accessible at https://idrblab.org/LocPro/.

Advancements in artificial intelligence(AI)and emerging technologies are rapidly expanding the exploration of chemical space,facilitating innovative drug discovery.However,the transformation of novel compounds into safe and effective drugs remains a lengthy,high-risk,and costly process.Comprehensive early-stage evaluation is essential for reducing costs and improving the success rate of drug development.Despite this need,no comprehensive tool currently supports systematic evaluation and efficient screening.Here,we present druglikeFilter,a deep learning-based framework designed to assess drug-likeness across four critical dimensions:1)physicochemical rule evaluated by systematic determination,2)toxicity alert investigated from multiple perspectives,3)binding affinity measured by dual-path analysis,and 4)compound synthesizability assessed by retro-route prediction.By enabling automated,multidimensional filtering of compound libraries,druglikeFilter not only streamlines the drug development process but also plays a crucial role in advancing research efforts towards viable drug candidates,which can be freely accessed at https://idrblab.org/drugfilter/.

Objective To determine the time point when porcine pancreatic elastase(PPE)induced abdominal aortic aneurysm(AAA)reaches the regression phase in mice and observe the histological characteristics of AAA in regression phase.Methods AAAs were induced by transient intraluminal infusion of PPE in C57BL/6J mice.The diameters of the mouse abdominal aortas were measured before PPE infusion and sacrifice time,day 14 for AAA progression phase or day 56 for regression phase after PPE infusion,respectively.The histological characteristics of the aneurysm lesion site on day 14 and day 56 after surgery were compared and analyzed.Results The diameters of the abdominal aortas were significantly increased in both day 14 and day 56 after PPE infusion groups(diameter growth rate 147％and 155％,respectively)as compared to the baseline diameters.In the day 14 group,the infused aortas showed typical AAA characteristics,such as elastin break/degradation,medial smooth muscle cells depletion,and inflammatory cell diffused infiltration.In the day 56 group after PPE infusion,although the artery diameter did not change significantly as compared to the day 14 group,histology showed that elastin was partially repaired,new smooth muscle cells were added to the damaged aorta media,the infiltrated inflammatory cells were significantly subsided,and the adventitia neovascularization was reduced,showing a significant feature of the disease regression phase.Conclusion In the PPE-induced mouse AAA model,day 56 after surgery is an appropriate time point for observing aneurysm regression,and the histological characteristics of the regression are obvious.

BACKGROUND:Overactive osteoclasts disrupt bone homeostasis and play a bad role in the pathological mechanisms of related skeletal diseases,such as osteoporosis,fragility fractures,and osteoarthritis.Studies have confirmed that ellagic acid and ellagtannin have the potential to inhibit osteoclast differentiation.As their natural metabolites,urolithin A has antioxidant,anti-inflammatory,anti-proliferative and anti-cancer effects,but its effect on osteoclast differentiation and its underlying molecular mechanisms remain unclear. OBJECTIVE:To explore the effect of urolithin A on osteoclast differentiation induced by receptor activator for nuclear factor-κB ligand and its mechanism. METHODS:Mouse mononuclear macrophage leukemia cells(RAW264.7)that grew stably were cultured in vitro.Toxicity of urolithin A(0,0.1,0.5,1.5,2.5 μmol/L)to RAW264.7 cells were detected by cytotoxic MTS assay to screen out the safe concentration.Different concentrations of urolithin A were used again to intervene with receptor activator for nuclear factor-κB ligand-induced differentiation of RAW264.7 cells in vitro.Then,tartrate-resistant acid phosphatase staining and F-actin ring and nucleus staining were performed to observe its effect on the formation and function of osteoclasts.Finally,the expressions of urolithin A on upstream and downstream genes and proteins in the MAPK signaling pathway were observed by western blot and RT-qPCR assays. RESULTS AND CONCLUSION:Urolithin A inhibited osteoclast differentiation and F-actin ring formation in a concentration-dependent manner and 2.5 μmol/L had the strongest inhibitory effect.Urolithin A inhibited the mRNA expression of Nfatc1,Ctsk,Mmp9 and Atp6v0d2 and the protein synthesis of Nfatc1 and Ctsk,related to osteoclast formation and bone resorption.Urolithin A inhibited the activity of osteoclasts by downregulating the phosphorylation of p38 protein to inhibit the mitogen-activated protein kinase signaling pathway.