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.

Sleep is an instinctive behavior alternating awakening state, sleep entails many active processes occurring at the cellular, circuit and organismal levels. The function of sleep is to restore cellular energy, enhance immunity, promote growth and development, consolidate learning and memory to ensure normal life activities. However, with the increasing of social pressure involved in work and life, the incidence of sleep disorders (SD) is increasing year by year. In the short term, sleep disorders lead to impaired memory and attention; in the longer term, it produces neurological dysfunction or even death. There are many ways to directly or indirectly contribute to sleep disorder and keep the hormones, including pharmacological alternative treatments, light therapy and stimulus control therapy. Exercise is also an effective and healthy therapeutic strategy for improving sleep. The intensities, time periods, and different types of exercise have different health benefits for sleep, which can be found through indicators such as sleep quality, sleep efficiency and total sleep time. So it is more and more important to analyze the mechanism and find effective regulation targets during sleep disorder through exercise. Dopamine (DA) is an important neurotransmitter in the nervous system, which not only participates in action initiation, movement regulation and emotion regulation, but also plays a key role in the steady-state remodeling of sleep-awakening state transition. Appreciable evidence shows that sleep disorder on humans and rodents evokes anomalies in the dopaminergic signaling, which are also implicated in the development of psychiatric illnesses such as schizophrenia or substance abuse. Experiments have shown that DA in different neural pathways plays different regulatory roles in sleep behavior, we found that increasing evidence from rodent studies revealed a role for ventral tegmental area DA neurons in regulating sleep-wake patterns. DA signal transduction and neurotransmitter release patterns have complex interactions with behavioral regulation. In addition, experiments have shown that exercise causes changes in DA homeostasis in the brain, which may regulate sleep through different mechanisms, including cAMP response element binding protein signal transduction, changes in the circadian rhythm of biological clock genes, and interactions with endogenous substances such as adenosine, which affect neuronal structure and play a neuroprotective role. This review aims to introduce the regulatory effects of exercise on sleep disorder, especially the regulatory mechanism of DA in this process. The analysis of intracerebral DA signals also requires support from neurophysiological and chemical techniques. Our laboratory has established and developed an in vivo brain neurochemical analysis platform, which provides support for future research on the regulation of sleep-wake cycles by movement. We hope it can provide theoretical reference for the formulation of exercise prescription for clinical sleep disorder and give some advice to the combined intervention of drugs and exercise.

Objective To explore the relationship between axillary artery and axillary vein by combining computed tomography angiography computed tomography angiography(CTA)and computed tomography venography(CTV),and to provide imaging data for establishing an effective blood circulation pathway for vascular injury in clinical transaxillary surgery.Methods The image data of 30 patients who underwent left upper limb and axillary CTA and CTV at the same time were collected.After three-dimensional reconstruction of the images by GE AW4.6 workstation,the course,branch type and variation of axillary artery,the course of axillary vein,the reflux of subordinate branches and the relationship between axillary artery and axillary vein were observed.The length of the whole segment,the length of segment and the inner diameter of the starting point of each segment were measured and statistically analyzed.Results According to the number of branches from the main trunk,the axillary artery was divided into 7 types.According to the variation of the number of blood vessels,the axillary vein was divided into 5 types;The unknown vein branches converge into 32 branches,of which the first segment accounted for 37.5％,the second segment accounted for 46.9％,and the third segment accounted for 15.6％.According to the absence of arterial branches,the relationship between axillary artery and axillary vein was divided into 2 types.Conclusion There are many types of branches of axillary arteries and branches of axillary veins,and the variation types are complex.The changes of branches of axillary arteries affect the distribution of branches of axillary veins.Combined with CTA and CTV images,the relationship between axillary vessels can be reflected clearly and intuitively,which can provide imaging reference for the establishment of new ways of blood supply and reflux in clinical axillary treatment.

Objective To investigate the value of multimodal model integrating digital mammography(MG)and MRI radiomics features for preoperative prediction of lymphovascular invasion(LVI)status in breast cancer.Methods The clinical and imaging data from 336 patients with pathologically confirmed invasive breast cancer were retrospectively analyzed and randomly divided into a training group(235 cases)and a test group(101 cases)according to the ratio of 7∶3.Feature dimensionality reduction was carried out by Pearson correlation analysis followed by least absolute shrinkage and selection operator(LASSO)regression.Radiomics models were constructed based on MG craniocaudal(CC),dynamic contrast enhancement(DCE),T2 WI,and integrated MRI sequences;a multimodal model was further developed by incorporating clinical high-risk factors.The predictive efficiency of each model was evaluated by plotting receiver operating characteristic(ROC)curve.Results The ROC curve analysis showed that the multimodal model performed the best predictive efficiency,with area under the curve(AUC)of 0.989 and 0.861,accuracy of 0.949 and 0.782,sensitivity of 0.923 and 0.828,and specificity of 0.962 and 0.764 in the training group and test group respectively.Conclusion The multimodal model,integrating MG and MRI radiomics features,show optimal performance and can be served as a preoperative prediction of LVI status in breast cancer.

Neuroscience,a cutting-edge field in interdisciplinary research,consistently draws considerable research interest,of which quantitatively probing the neurochemical dynamics is essential for brain science research.In vivoelectrochemical analysis,featuring with high sensitivity,high spatiotemporal resolution,free from transfection,and designable electrode/solution interfaces,provides important tools for in vivo neurochemicals sensing.Fast scan cyclic voltammetry combined with microelectrodes can not only enable precise detection of dopamine but also is compatible with existing neurosurgical equipment.This offers new opportunities for the clinical application of in vivo electrochemical analysis and paves new avenues for the diagnosis and treatment of neurological diseases.This review summarized recent progress of in vivo electrochemical techniques for brain neurochemistry and addressed key clinical challenges and their potential solutions.

Objective:To explore the screening value of platelet parameters from blood cell analysis for MYH9-related disorders(MYH9-RD).Methods:A cross-sectional study was conducted with 38 patients diagnosed with MYH9-RD at Shenzhen Children's Hospital from May 1, 2016, to August 31, 2024, including 24 males and 14 females; the median age was 11.5 (3.8, 35) years; categorized by gene mutation location into "head region" ( n=8 ) and "tail region" ( n=30); and by clinical manifestations into " isolated hematological manifestations" ( n=16) and "hematological manifestations with extra-hematological involvement"( n=22). The control groups included 39 cases of immune thrombocytopenia (ITP), 38 cases of acute lymphoblastic leukemia (ALL), and 40 healthy individuals. Platelet-related parameters were detected by hematology analyzer, and platelet counts and sizes were confirmed by manually counting and microscopic observation. Kruskal-Wallis test was used to compare platelet parameters between MYH9-RD and control groups. Receiver operating characteristic (ROC) curves were used to assess the diagnostic efficacy of platelet parameters for MYH9-RD. Results:In MYH9-RD patients the median value of mean platelet volume (MPV) was 13.4 (11.2, 14.7) fl, immature platelet fraction (IPF) was 52.7% (43.5%, 58.0%), platelet large cell ratio(PLCR) was 57.6 %(45.0%, 62.9%), and microscopic large platelet ratio (PLCR-M) was 30.0% (25.0%, 30.0%).And those values weresignificantly higher than in ITP, ALL, and healthy controls (all P<0.05). Patients with MYH9 gene "head region" mutations had a lower platelet count [24.5 (15.0, 47.5)×10 9/L]than those with "tail region" mutations [69.0 (49.5, 86.3) ×10 9/L]( Z=-3.493, P<0.001), but a higher IPF ( t=2.024, P=0.044).Patients with "extra-hematological involvement had a lower platelet count than those with "isolated hematological manifestations" ( t=-2.015, P=0.043). The optimal cutoff value for diagnosing MYH9-RD with IPF was 26.7%, with a sensitivity of 100% and specificity of 98.7%; the area under the curve was 0.999 (95% CI 0.995-1.000), which was superior toMPV, PLCR and PLCR-M parameters. Conclusion:IPF is superior to other platelet parameters sush as MPV,showing high diagnostic efficacy in distinguishing MYH9-RD from ITP and ALL. It can be used as a simple and effective indicator for early screening of MYH9-RD.

Objective:To observe the efficacy and safety of polatuzumab vedotin(pola)combined with chemotherapy in the treatment of relapsed and refractory diffuse large B-cell lymphoma(R/R DLBCL).Methods:A total of 23 patients with R/R DLBCL treated at the First Affiliated Hospital of Zhejiang University and its Liangzhu Branch from April 2023 to March 2024 were retrospectively collected.All patients were treated with pola combined with chemotherapy regimens such as BR,R-GDP,R-CHOP,or other regimens.Results:All 23 patients were evaluable for efficacy,with 10 achieving complete response(CR),7 partial response(PR),3 stable disease(SD),and 3 progressive disease(PD).The most common adverse events included myelosuppression,fever,and pulmonary infection.No severe adverse events resulted in drug withdrawal.Conclusion:Pola combined with chemotherapy demonstrates promising efficacy and a favorable safety profile in the treatment of R/R DLBCL.

Objective To evaluate the safety and efficacy of valve-in-valve transcatheter mitral valve replacement(ViV-TMVR)in patients with bioprosthetic valve degeneration who are at high surgical risk.Methods This study is a multi-center,retrospective cohort analysis of 20 consecutive patients who underwent transseptal ViV-TMVR using the Edwards SAPIEN 3 transcatheter heart valve(THV).The primary endpoints include technical success and procedural success,both defined according to the Mitral Valve Academic Research Consortium(MVARC)criteria,as well as mortality and functional change assessed based on New York Heart Association(NYHA)classification at 30-days and six months post-procedure.Clinical follow-up assessments are conducted at 30-days and six months.Results From February 2021 to October 2022,a total of 20 patients with symptoms of bioprosthetic valve degeneration were enrolled across nine sites in China.The patients had a mean age of(73.5±5.5)years,with 85.0％being females and 70.0％classified as NYHA class Ⅲ/Ⅳ.The study achieved a 100.0％technical success rate and a 90.0％procedural success rate finally.All patients remained alive during the 30-day follow-up period.However,six months post-intervention,two patients(10.0％)were re-hospitalized due to heart failure,and sadly,one of them(5.0％)died.None of the patients reported any adverse events related to ViV-TMVR during the follow-up period.Notably,there was a significant improvement in NYHA class compared to baseline(P=0.0004)at six-month follow-ups.Conclusions The transseptal ViV-TMVR technique proved to be highly successful and was associated with significant improvement in NYHA class function.These findings strongly suggest that it serves as a safe and efficient treatment alternative for high-risk patients suffering from bioprosthetic valve degeneration.

Objective This study aims to provide theoretical support for the formulation of thyroid nodule prevention and treatment strategies in Northwest China and to investigate the characteristics of thyroid nodules and their association with metabolic indicators in this population.Methods A cross-sectional survey was conducted by retrospectively enrolling healthy individuals who underwent routine health examinations at our hospital between January 1 and December 31,2023.A total of 38 919 participants were included.The detection rate of thyroid nodules was stratified by age and sex.Univariate and multivariate binary logistic regression analyses were performed to identify risk factors for thyroid nodules.Results Among the 38 919 participants,20 395 were men(52.4%)and 18 524 were women(47.6%).The overall detection rate of thyroid nodules was 47.1%(18 317/38 919),including 40.1%(8 187/20 395)in men and 54.7%(10 130/18 524)in women.Across all age groups,women had a significantly higher detection rate than men(P<0.001).The detection rate increased with age in both sexes(χ2trend=1392.867,P<0.001 in men;χ2trend=1521.215,P<0.001 in women).Significant differences were observed between participants with and without thyroid nodules in age,body mass index(BMI),fasting blood glucose(FBG),glycated hemoglobin(HbA1c),total cholesterol(TC),low-density lipoprotein cholesterol(LDL-C),triiodothyronine(T3),thyroid-stimulating hormone(TSH),and anti-thyroglobulin antibodies(TGAB),anti-thyroid microsomal antibodies,the proportion of hypertension and central obesity(all P<0.05).Multivariate binary logistic regression analysis identified female sex(OR=2.158),older age(OR=1.040),central obesity(OR=1.144),elevated FBG(OR=1.039),hypertension(OR=1.095),elevated TGAB(OR=1.008),and elevated T3(OR=1.154)as independent risk factors for thyroid nodules(all P<0.05).Conclusion Women in Northwest China are at higher risk of developing thyroid nodules.Screening and health management should be prioritized for older individuals,those with central obesity,elevated FBG,hypertension,elevated TGAB,and elevated T3 levels.