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

Perceptual decision making refers to the process by which individuals make choices and judgments based on sensory information, serving as a fundamental ability for human adaptation to complex environments. While traditional research has focused on perceptual decision making in isolated contexts, growing evidence highlights the profound influence of social contexts prevalent in real-world scenarios. As a crucial factor supporting individual survival and development, social context not only provides rich information sources but also shapes perceptual decision making through top-down processing mechanisms, prompting researchers to recognize the inherently social nature of human decisions. Empirical studies have demonstrated that social information, such as others’ choices or group norms, can systematically bias individuals’ perceptual decisions, often manifesting as conformity behaviors. Social influence can also facilitate performance under certain conditions, particularly when individuals can accurately identify and adopt high-quality social information. The impact of social context on perceptual decisions is modulated by a variety of external and internal factors, including group characteristics(e.g., group size, response consistency), attributes of peers (e.g., familiarity, social status, distinctions between human and artificial agents), as well as individual differences such as confidence, personality traits, and developmental stage. The motivations driving social influence encompass three primary mechanisms: improving decision accuracy through informational influence, gaining social acceptance through normative influence, and maintaining positive self-concept. Recent computational approaches have employed diverse theoretical frameworks to provide valuable insights into the cognitive mechanisms underlying social influence in perceptual decision making. Reinforcement learning models demonstrate how social feedback shapes future choices through reward-based updating. Bayesian inference frameworks describe how individuals integrate personal beliefs with social information based on their respective reliabilities, dynamically updating beliefs to optimize decisions under uncertainty. Drift diffusion models offer powerful tools to decompose social influence into distinct cognitive components, allowing researchers to differentiate between changes in perceptual processing and shifts in decision criteria. Collectively, these models establish a comprehensive methodological foundation for disentangling the multiple pathways by which social context shapes perceptual decisions. Neuroimaging and electrophysiological studies provide converging evidence that social context influences perceptual decision making through multi-level neural mechanisms. At early perceptual processing stages, social influence modulates sensory evidence accumulation in parietal cortex and directly alters primary visual cortex activity, while guiding selective attention to stimulus features consistent with social norms through attentional alignment mechanisms. At higher cognitive levels, the reward system (ventral striatum, ventromedial prefrontal cortex) is activated during group-consistent decisions; emotion-processing networks (anterior cingulate cortex, insula, amygdala) regulate experiences of social acceptance and rejection; and mentalizing-related brain regions (dorsomedial prefrontal cortex, temporoparietal junction) support inference of others’ mental states and social information integration. These neural circuits work synergistically to achieve top-down multi-level modulation of perceptual decision making. Understanding the mechanisms by which social context shapes perceptual decision making has broad theoretical and practical implications. These insights inform the optimization of collective decision-making, the design of socially adaptive human-computer interaction systems, and interventions for cognitive disorders such as autism spectrum disorder and anorexia nervosa. Future studies should combine computational modeling and neuroimaging approaches to systematically investigate the multi-level and dynamic nature of social influences on perceptual decision making.

Objective To propose a brain electrical phase prediction method based on long short-term memory network(LSTM)to improve the accuracy and robustness of phase synchronization prediction in transcranial magnetic stimulation(TMS).Methods First,an LSTM consisting of an input layer,an LSTM layer,an ReLU activation layer,a fully connected layer and a regression layer was constructed to capture the EEG signal features through the synergistic action of input gates,forgetting gates and output gates.Second,eye-open resting-state EEG data from 30 healthy subjects were trained using the LSTM to obtain a predictive model for EEG signal and EEG phase prediction.Finally,the LSTM method and the traditional autoregressive(AR)method were compared in terms of the phase prediction errors at the overall and individual levels and the prediction performance for peaks and troughs.A regression model was used to explore the relationships between instantaneous EEG amplitude,signal-to-noise ratio and phase prediction error with the LSTM method.Results The LSTM method achieved a total phase prediction error of 0.04°±5.69°,which was lower than that of the traditional AR method(-3.36°±51.13°).For each subject,the LSTM method demonstrated superior phase prediction accuracy compared to the traditional AR method(P＜0.001).The accuracy for predicting peaks(troughs)by the LSTM method(about 89％)was higher than that by the traditional AR method(about 10％).Unlike the traditional AR method,the LSTM method didnot result in linear relationships between instantaneous EEG amplitude,signal-to-noise ratio and phase prediction error,with Pvalues being 0.58 and 0.18,respectively.Conclusion The LSTM-based brain electrical phase prediction method shows high accuracy and robustness when used for EEG phase-synchronized TMS.[Chinese Medical Equipment Journal,2025,46(3):1-8]

Objective:To investigate the impact of correcting rotational subluxation through circumferential fusion and transforaminal lumbar interbody fusion (TLIF) on postoperative coronal plane imbalance in degenerative scoliosis.Methods:A retrospective analysis was conducted on the data of 108 patients with type A degenerative scoliosis in the Nanjing classification who underwent primary multi-segment posterior column osteotomy (PCO) with deformity correction and internal fixation at Nanjing Gulou Hospital from June 2017 to June 2021. Patients were divided into two groups based on the presence of preoperative rotational subluxation: the rotational subluxation group and the non-rotational subluxation group. The rotational subluxation group consisted of 60 patients, with 8 males and 52 females, aged 63.7±5.5 years (range, 56-75 years). The non-rotational subluxation group included 48 patients, with 5 males and 43 females, aged 64.4±5.2 years (range, 53-72 years). Within the rotational subluxation group, depending on whether TLIF was performed on the rotational subluxation segment, they were further categorized into the TLIF group and the PCO group. The TLIF group comprised 28 patients, while the PCO group had 32 patients. Full-spine anteroposterior and lateral X-rays were taken preoperatively, postoperatively, and at the last follow-up to measure coronal balance types and radiographic parameters. The differences in the lumbar Cobb angle, coronal balance distance (CBD), and the Cobb angle of the lumbosacral curve (Cobb-Fra angle) were compared between the rotational subluxation group and the non-rotational subluxation group, as well as between the TLIF group and the PCO group.Results:The average surgery duration ranged from 200 to 310 min, with a mean of 235±47 min. The intraoperative blood loss ranged from 700 to 2,400 ml, with an average of 950±355 ml. The number of fused segments in the rotational subluxation group was 7.6±2.1, ranging from 5 to 11 segments, while in the non-rotational subluxation group, it was 7.4±2.0, ranging from 5 to 10 segments. Postoperatively, 13%(8/60) of patients in the rotational subluxation group developed type C coronal imbalance, significantly higher than the 2%(1/48) in the non-rotational subluxation group. The immediate postoperative and final follow-up lumbar Cobb angles, CBD, and Cobb-Fra angles in the rotational subluxation group were 20.60°±10.73° and 20.33°±10.92°, 22.53±16.45 mm and 18.53±17.31 mm, 13.14°±4.40° and 11.23°±4.92°, respectively, which were higher than those in the non-rotational subluxation group (13.92°±7.02° and 12.92°±6.64°, 18.62±17.44 mm and 8.83±8.95 mm, 11.91°±3.03° and 9.52°±3.30°), with statistical significance ( P<0.05).. Among patients in the rotational subluxation group, the probability of new-onset coronal imbalance postoperatively was 4%(1/28) in the TLIF group, which was lower than the 22%(7/32) in the PCO group, with a statistically significant difference (χ 2=4.330, P=0.037). The immediate postoperative and final follow-up lumbar Cobb angles, CBD, and Cobb-Fra angles in the PCO group were 25.63°±11.00° and 25.13°±11.04°, 27.37±18.95 mm and 25.25±18.67 mm, 15.50°±3.62° and 14.08°±4.77°, respectively, which were significantly higher than those in the TLIF group (14.86°±6.96° and 14.86°±5.37°, 17.08±10.94 mm and 10.86±7.86 mm, 10.14°±3.37° and 8.46°±2.66°), with statistical significance ( P<0.05). Conclusion:For patients with Type A degenerative scoliosis combined with rotational subluxation according to the Nanjing classification, performing a 360-degree circumferential release and interbody fusion at the segment with rotatory subluxation can reduce the risk of developing new postoperative coronal imbalances.

In this paper, near-infrared spectroscopy(NIRS) was employed to analyze 129 batches of commercial products of Reduning Injection. The batch reporting rate was estimated according to the report of Reduning Injection in the direct adverse drug reaction(ADR) reporting system of the drug marketing authorization holder of the Center for Drug Reevaluation of the National Medical Products Administration(National Center for ADR Monitoring) from August 2021 to August 2022. According to the batch reporting rate, the samples of Reduning Injection were classified into those with potential risks and those being safe. No processing, random oversampling(ROS), random undersampling(RUS), and synthetic minority over-sampling technique(SMOTE) were then employed to balance the unbalanced data. After the samples were classified according to appropriate sampling methods, competitive adaptive reweighted sampling(CARS), successive projections algorithm(SPA), uninformative variables elimination(UVE), and genetic algorithm(GA) were respectively adopted to screen the features of spectral data. Then, support vector machine(SVM), logistic regression(LR), k-nearest neighbors(KNN), naive bayes(NB), random forest(RF), and artificial neural network(ANN) were adopted to establish the risk prediction models. The effects of the four feature extraction methods on the accuracy of the models were compared. The optimal method was selected, and bayesian optimization was performned to optimize the model parameters to improve the accuracy and robustness of model prediction. To explore the correlations between potential risks of clinical use and quality test data, TreeNet was employed to identify potential quality parameters affecting the clinical safety of Reduning Injection. The results showed that the models established with the SVM, LR, KNN, NB, RF, and ANN algorithms had the F1 scores of 0.85, 0.85, 0.86, 0.80, 0.88, and 0.85 and the accuracy of 88%, 88%, 88%, 85%, 91%, and 88%, respectively, and the prediction time was less than 5 s. The results indicated that the established models were accurate and efficient. Therefore, near infrared spectroscopy combined with machine learning algorithms can quickly predict the potential risks of clinical use of Reduning Injection in batches. Three key quality parameters that may affect clinical safety were identified by TreeNet, which provided a scientific basis for improving the safety standards of Reduning Injection.
Spectroscopy, Near-Infrared/methods* ; Drugs, Chinese Herbal/administration & dosage* ; Machine Learning ; Algorithms ; Humans ; Quality Control

This study investigated the mechanism of Chaishao Kaiyu Decoction in improving hippocampal neuroinflammation in depressed rats based on complement component 3(C3)/C3 receptor(C3aR). A total of 60 SD rats were randomly divided into a blank group, a model group, high, medium, and low dose groups of Chaishao Kaiyu Decoction, and a positive drug group, with 10 rats in each group. Except for the blank group, chronic unpredictable mild stress(CUMS) was used to construct depression models in other groups. Sucrose preference, open-field experiment, forced swimming, and water maze were used to detect the changes in depression-like behavior in each group. Enzyme-linked immunosorbent assay(ELISA) was used to detect the serum inflammatory factor level in rats, and hematoxylin-eosin(HE) staining and Nissl staining were employed to observe the pathological damage of hippocampal neurons. Golgi-Cox staining was used to observe the dendritic spine damage of hippocampal neurons, and immunofluorescence and Western blot were utilized to detect the expression of microglial marker Iba-1 and C3/C3aR protein in the hippocampus of rats. The behavioral results showed that compared with the model group, Chaishao Kaiyu Decoction could significantly strengthen the sugar water preference, increase the distance and number of voluntary activities, shorten the immobility time in forced swimming and the successful incubation period of positioning navigation, and prolong the stay time of space exploration in the target quadrant. ELISA results showed that the content of inflammatory factors in the hippocampus of depressed rats was significantly higher than that of the blank group, and the content of inflammatory factors decreased significantly after the intervention of Chaishao Kaiyu Decoction. In addition, Chaishao Kaiyu Decoction could relieve pathological damage such as cell swelling and loose arrangement of hippocampus tissue. In the Western blot experiment, the expression levels of C3 and C3aR proteins in the model group were higher than those in the blank group, while the expression of C3 and C3aR in Chaishao Kaiyu Decoction could be down-regulated. Immunofluorescence results showed that compared with the model group, the fluorescence intensity of microglia marker Iba-1 decreased significantly after the intervention of Chaishao Kaiyu Decoction and positive drugs. The antidepressant effect of Chaishao Kaiyu Decoction may be related to the down-regulation of C3/C3aR signaling pathway-related proteins, thus alleviating hippocampal inflammation.
Animals ; Hippocampus/metabolism* ; Rats, Sprague-Dawley ; Drugs, Chinese Herbal/administration & dosage* ; Rats ; Male ; Depression/metabolism* ; Complement C3/metabolism* ; Receptors, Complement/metabolism* ; Humans ; Neuroinflammatory Diseases/genetics*

This study aims to explore the effects of Buzhong Yiqi Decoction(BYD) on the cyclic guanosine monophosphate-adenosine monophosphate synthase(cGAS)-stimulator of interferon genes(STING) signaling pathway in the mouse model of autoimmune thyroiditis(AIT) and the mechanism of BYD in alleviating the immune injury. Forty-eight NOD.H-2~(h4) mice were assigned into normal, model, low-, medium-, and high-dose BYD, and selenium yeast tablets groups(n=8). Mice of 8 weeks old were treated with 0.05% sodium iodide solution for 8 weeks for the modeling of AIT and then administrated with corresponding drugs by gavage for 8 weeks before sampling. High performance liquid chromatography was employed to measure the astragaloside Ⅳ content in BYD. Hematoxylin-eosin staining was employed to observe the pathological changes in the mouse thyroid tissue. Enzyme-linked immunosorbent assay was employed to measure the serum levels of thyroid peroxidase antibody(TPO-Ab), thyroglobulin antibody(TgAb), and interferon-γ(IFN-γ). Flow cytometry was employed to detect the distribution of T cell subsets in the spleen. The immunohistochemical method was used to detect the expression of cGAS, STING, TANK-binding kinase 1(TBK1), and interferon regulatory factor 3(IRF3). Real-time PCR and Western blot were employed to determine the mRNA and protein levels, respectively, of markers related to the cGAS-STING signaling pathway in the thyroid tissue. The results showed that the content of astragaloside Ⅳ in BYD was(7.06±0.08) mg·mL~(-1). Compared with the normal group, the model group showed disrupted structures of thyroid follicular epithelial cells, massive infiltration of lymphocytes, and elevated levels of TgAb and TPO-Ab. Compared with the model group, the four treatment groups showed intact epithelial cells, reduced lymphocyte infiltration, and lowered levels of TgAb and TPO-Ab. Compared with the normal group, the model group showed increases in the proportions of Th1 and Th17 cells, a decrease in the proportion of Th2 cells, and an increase in the IFN-γ level. Compared with the model group, the four treatment groups presented decreased proportions of Th1 and Th17 cells and lowered levels of IFN-γ, and the medium-dose BYD group showed an increase in the proportion of Th2 cells. Compared with the normal group, the modeling up-regulated the mRNA levels of cGAS, STING, TBK1, and IRF3 and the protein levels of cGAS, p-STING, p-TBK1, and p-IRF3. Compared with the model group, the four treatment groups showed reduced levels of cGAS, STING, TBK1, and IRF3-positive products, down-regulated mRNA levels of cGAS, STING, and TBK1, and down-regulated protein levels of cGAS and p-STING. The high-dose BYD group showed down-regulations in the mRNA level of IRF3 and the protein levels of p-TBK1 and p-IRF3. The above results indicate that BYD can repair the imbalance of T cell subsets, alleviate immune injury, and reduce thyroid lymphocyte infiltration in AIT mice by inhibiting the cGAS-STING signaling pathway.
Animals ; Drugs, Chinese Herbal/administration & dosage* ; Signal Transduction/drug effects* ; Thyroiditis, Autoimmune/metabolism* ; Mice ; Membrane Proteins/metabolism* ; Mice, Inbred NOD ; Humans ; Female ; Nucleotidyltransferases/metabolism* ; Male ; Disease Models, Animal

Continuous manufacturing, as an innovative pharmaceutical production model, offers advantages such as high production efficiency and ease of control compared to traditional batch production, aligning with the future trend of drug production moving toward greater efficiency and intelligence. However, the development of continuous manufacturing technology in wet granulation has been slow. On one hand, this is closely related to its high technical complexity, substantial equipment investment costs, and stringent process control requirements. On the other hand, the long-term use of the traditional batch production model has created strong path dependence, and the lack of mature standardized processes further increases the difficulty of technological transformation. To promote the deep integration of wet granulation technology with continuous manufacturing, this review systematically outlines the current application of wet granulation in continuous manufacturing. It focuses on the development of key technologies such as online detection, process modeling, and process scale-up, with the aim of providing a reference for process innovation and application in wet granulation.
Drug Compounding/instrumentation* ; Technology, Pharmaceutical/methods* ; Drugs, Chinese Herbal/chemistry* ; Models, Theoretical

This paper aimed to use non-targeted urine metabolomics to reveal the potential biomarkers of toxicity in rats with hepatic injury induced by aqueous extracts of Dictamni Cortex(ADC). Forty-eight SD rats were randomly assigned to a blank group and high-dose, medium-dose, and low-dose ADC groups, with 12 rats in each group(half male and half female), and they were administered orally for four weeks. The hepatic injury in SD rats was assessed by body weight, liver weight/index, biochemical index, L-glutathione(GSH), malondialdehyde(MDA), and pathological alterations. The qPCR was utilized to determine the expression of metabolic enzymes in the liver and inflammatory factors. Differential metabolites were screened using principal component analysis(PCA) and partial least squares-discriminant analysis(PLS-DA), followed by a metabolic pathway analysis. The Mantel test was performed to assess differential metabolites and abnormally expressed biochemical indexes, obtaining potential biomarkers. The high-dose ADC group showed a decrease in body weight and an increase in liver weight and index, resulting in hepatic inflammatory cell infiltration and hepatic steatosis. In addition, this group showed elevated levels of MDA, cytochrome P450(CYP) 3A1, interleukin-1β(IL-1β), and tumor necrosis factor-α(TNF-α), as well as lower levels of alanine transaminase(ALT) and GSH. A total of 76 differential metabolites were screened from the blank and high-dose ADC groups, which were mainly involved in the pentose phosphate pathway, tryptophan metabolism, purine metabolism, pentose and glucuronic acid interconversion, galactose metabolism, glutathione metabolism, and other pathways. The Mantel test identified biomarkers of hepatotoxicity induced by ADC in SD rats, including glycineamideribotide, dIDP, and galactosylglycerol. In summary, ADC induced hepatotoxicity by disrupting glucose metabolism, ferroptosis, purine metabolism, and other pathways in rats, and glycineamideribotide, dIDP, and galactosylglycerol could be employed as the biomarkers of its toxicity.
Animals ; Male ; Rats, Sprague-Dawley ; Rats ; Metabolomics ; Biomarkers/metabolism* ; Liver/metabolism* ; Drugs, Chinese Herbal/adverse effects* ; Female ; Chemical and Drug Induced Liver Injury/metabolism* ; Glutathione/metabolism* ; Humans