BackgroundLow medication compliance among patients with severe mental disorders increases the disease burden on both the patients' families and the society. Medication adherence is influenced by numerous factors. Traditional methods such as Logistic regression struggle to quantify the importance of these factors. By introducing Extreme Gradient Boosting （XGBoost） combined with Shapley Additive Explanations （SHAP）， enables the quantification of the relative contribution weights of each factor， providing support for identifying the core influencing factors. ObjectiveTo explore the influencing factors of medication adherence among patients with severe mental disorders in Zhuhai， aiming to provide references for optimizing patient management strategies. MethodsExtract the data of patients with severe mental disorders who were registered on the mental health system platform in Zhuhai City from January 1， 2023 to March 31， 2025. A total of 9 329 patients were finally included for analysis. Influencing factors were screened using univariate analysis and multivariate logistic regression analysis， and an XGBoost model combined with the SHAP algorithm was constructed to quantify the importance of each influencing factor. ResultsAmong 9 329 patients， 8 446 demonstrated medication adherence， yielding an adherence rate of 90.53%. Multivariable analysis identified several risk factors significantly associated with medication non-adherence， being unmarried （OR=1.237， 95% CI： 1.019–1.502） or divorced （OR=1.389， 95% CI： 1.038–1.832）， a diagnosis of mental retardation with psychiatric disorders （OR=3.025， 95% CI： 2.402–3.796） or paranoid psychosis （OR=5.117， 95% CI： 3.086–8.299）， a disease duration of 2–4 years （OR=1.355， 95% CI： 1.085–1.696）， 4–6 years （OR=2.143， 95% CI： 1.671–2.747）， or >6 years （OR=1.681， 95% CI： 1.365–2.079）， lack of guardian subsidies （OR=1.412， 95% CI： 1.099–1.801）， absence of a disability certificate （OR=1.900， 95% CI： 1.588–2.282）， not being enrolled in care and support groups （OR=1.384， 95% CI： 1.183–1.617） or community services （OR=1.313， 95% CI： 1.042–1.645）， and not cohabiting with a guardian （OR=1.257， 95% CI： 1.048–1.501）. Conversely， the enrollment in special outpatient disease programs （OR=0.716， 95% CI： 0.609–0.842） and a family history of mental illness （OR=0.713， 95% CI： 0.503–0.982） were identified as protective factors. The XGBoost model exhibited robust predictive performance， with a sensitivity of 0.433， specificity of 0.944， accuracy of 0.891， Area Under the Curve （AUC） of 0.837， and F1 value of 0.449. Feature importance ranking indicated that the top three factors were disease duration， diagnosis， and the acquisition of disability certificates. ConclusionPolicy-based support （acquisition of disability certificates， special outpatient disease enrollment） and clinical disease characteristics （disease duration， diagnosis type） are key factors affecting medication adherence among patients with severe mental disorders in Zhuhai City. ［Funded by Zhuhai Medical Research Project （number， 2220009000281）］

BackgroundThe middle school stage represents a crucial period for the development of borderline personality features. Negative parenting styles and emotional regulation strategies are associated with the formation of borderline personality features. However， the moderating role of emotional regulation strategies between negative parenting styles and borderline personality features among middle school students remains unclear. ObjectiveTo explore the moderating influence of emotional regulation strategies in the relationship between negative parenting styles and borderline personality features among middle school students， and to provide references for the intervention of borderline personality features. MethodsIn October 2023， a total of 5 965 middle school students from three middle schools in Nanning， Guangxi Zhuang Autonomous Region were selected by cluster sampling， and assessed by the Borderline Personality Features Scale for Children （BPFS-C）， the Egna Minnen Barndoms Uppfostran （EMBU）， and the Emotion Regulation Questionnaire-Chinese Revised Version （ERQ-CRV）. Pearson correlation analysis was used to test the correlation between the scores of each scale， and the model 1 of the Process macro program was used to conduct the moderating effect test. ResultsA total of 5 572 middle school students （93.41%） completed this study， and 1 388 of them （24.91%） were identified as having high borderline personality features. The BPFS-C score of middle school students was positively correlated with the score of the negative parenting style dimension of EMBU （r=0.367， P<0.01）， negatively correlated with the score of the cognitive reappraisal dimension of ERQ-CRV （r=-0.168， P<0.01）， and positively correlated with the score of the expression inhibition dimension of ERQ-CRV （r=0.344， P<0.01）. Cognitive reappraisal played a negative moderating effect between negative parenting styles and borderline personality features （β=-0.072， 95% CI： -0.104–-0.041， P<0.01）， while expressive suppression played a positive moderating effect （β=0.076， 95% CI： 0.055–0.097， P<0.01）. ConclusionCognitive reappraisal strategy may help mitigate the negative influence of negative parenting styles on middle school students' borderline personality features， while expressive suppression may exacerbate the harm of negative parenting styles to the borderline personality features of middle school students.

Objective: To identify the latent class characteristics of health literacy and related factors among junior high school students, so as to provide evidence for developing precise and systematic health literacy promotion strategies. Methods: In November 2024, a two stage random cluster sampling method was used to conduct a questionnaire survey among 8 933 junior high school students in Zhongshan. Health literacy was assessed across six dimensions: health behavior and lifestyle, disease prevention and control, mental health, growth development and puberty health, safety emergency and risk avoidance, and medical knowledge and appropriate healthcare utilization. Latent profile analysis was used to identify distinct health literacy classes, and multinomial Logistic regression was applied to analyze the related factors. Results: Three latent classes of health literacy among junior high school students were identified: the well balanced type(71.7%，6 406), the medical knowledge deficit type(22.3%，1 992), and the overall low literacy type(6.0%，537). Logistic regression analysis showed that girls had lower risks of belonging to the medical knowledge deficit type( OR =0.53, 95% CI =0.48-0.59) and the overall low literacy type( OR =0.27,95% CI =0.22-0.33) compared with boys(both P <0.05). Students in rural schools had the highest risks of belonging to these two profiles above [ OR (95% CI ) =1.89 (1.61-2.21), 3.18 (2.50-4.06),both P <0.05]. Junior high school students having ≥2 siblings were positively associated with belonging to these two profiles, with risks 1.60 (95% CI = 1.35-1.89) and 2.25 times (95% CI =1.66-3.05) higher than those of only children (both P <0.05). Junior high school students with parental education of bachelor s degree or above were associated with lower risk of belonging to the medical knowledge deficit type (father: OR =0.63, 95% CI =0.47-0.84; mother: OR =0.68, 95% CI = 0.52 -0.90,both P <0.05). Junior high school students with receiving health education courses ≥3 times per month were associated with lower risks of belonging to both the medical knowledge deficit type and overall low literacy type ( OR =0.51, 95% CI =0.43- 0.60 ; OR =0.33, 95% CI =0.25-0.42, both P <0.05). Conclusions Three latent classes of health literacy exist among junior high school students in Zhongshan. Targeted interventions should be implemented based on profile characteristics, with an emphasis on strengthening medical knowledge education and providing comprehensive support for vulnerable groups.

This study focuses on the core pathology of sinew wei （痿）， which is mainly characterized by the fai-lure of qi and blood to nourish the sinews. A mechanical-biological response framework is constructed with mitochondria as a key component， explaining the modern interpretation of the disease location of sinew transmitting to qi and blood pathology. Mechanical loading， as a physical stress stimulus applied to the body， manifests primarily as passive loading formed by external forces such as massage， and active loading resulting from voluntary muscle contractions， such as dao yin （导引）. Mechanical loading can regulate mitochondrial function through two pathways， mechanical signal transduction and metabolic demand-driven regulation. Skeletal muscle mitochondrial dysfunction is regarded as the core microscopic basis of qi imbalance in sinew wei， highlighting the intrinsic connection between qi and mitochondrial energy metabolism， as well as between blood and microcirculatory efficiency. Accordingly， distinct regulatory patterns of mechanical loading are identified. Wei associated with qi stagnation may correspond to mitochondrial network fragmentation and can be treated by regulating qi through passive loading， such as tuina， to restore mitochondrial dynamics. In contrast， wei caused by qi deficiency is attributed to insufficient mitochondrial biogenesis and may be treated by tonifying qi through active loading， such as dao yin， to promote mitochondrial biogenesis. This framework reveals the biological differences in mitochondrial regulation induced by distinct mechanical loading modalities and provides a microscopic mechanism-based explanation for the principle of "treating the same disease with different methods" in sinew wei.

ObjectiveTo analyze the research hotspots and development trends of brain-computer interface (BCI) in the treatment for spinal cord injury (SCI). MethodsRelevant literatures on BCI applied in SCI treatment, published from the inception of the Web of Science Core Collection to July, 2025, were retrieved. Visualization analysis was performed using CiteSpace, VOSviewer and Tableau Desktop. ResultsA total of 437 literatures were included, and the annual number of publications showed an overall increasing trend. The United States ranked first in the number of publications; Graz University of Technology was the institution with the highest number of publication; Gernot R Mueller-Putz was the most productive author, while Jonathan R Wolpaw was the most cited author. Brain-computer interface and artificial intelligence were identified as the high-frequency and bursting keywords in this field. The researches were characterized by the cross-integration of five core disciplines: neuroscience and rehabilitation medicine, biomedical engineering, computer science and artificial intelligence, neurophysiology, and materials science. ConclusionResearches on BCI in SCI treatment are accelerating continuously, and technological integration is becoming the core trend.

ObjectiveTo analyze the research hotspots and development trends of brain-computer interface (BCI) in the treatment for spinal cord injury (SCI). MethodsRelevant literatures on BCI applied in SCI treatment, published from the inception of the Web of Science Core Collection to July, 2025, were retrieved. Visualization analysis was performed using CiteSpace, VOSviewer and Tableau Desktop. ResultsA total of 437 literatures were included, and the annual number of publications showed an overall increasing trend. The United States ranked first in the number of publications; Graz University of Technology was the institution with the highest number of publication; Gernot R Mueller-Putz was the most productive author, while Jonathan R Wolpaw was the most cited author. Brain-computer interface and artificial intelligence were identified as the high-frequency and bursting keywords in this field. The researches were characterized by the cross-integration of five core disciplines: neuroscience and rehabilitation medicine, biomedical engineering, computer science and artificial intelligence, neurophysiology, and materials science. ConclusionResearches on BCI in SCI treatment are accelerating continuously, and technological integration is becoming the core trend.

ObjectiveTo analyze the research hotspots and development trends of brain-computer interface (BCI) in the treatment for spinal cord injury (SCI). MethodsRelevant literatures on BCI applied in SCI treatment, published from the inception of the Web of Science Core Collection to July, 2025, were retrieved. Visualization analysis was performed using CiteSpace, VOSviewer and Tableau Desktop. ResultsA total of 437 literatures were included, and the annual number of publications showed an overall increasing trend. The United States ranked first in the number of publications; Graz University of Technology was the institution with the highest number of publication; Gernot R Mueller-Putz was the most productive author, while Jonathan R Wolpaw was the most cited author. Brain-computer interface and artificial intelligence were identified as the high-frequency and bursting keywords in this field. The researches were characterized by the cross-integration of five core disciplines: neuroscience and rehabilitation medicine, biomedical engineering, computer science and artificial intelligence, neurophysiology, and materials science. ConclusionResearches on BCI in SCI treatment are accelerating continuously, and technological integration is becoming the core trend.

Pulmonary fibrosis（PF） is a progressive and life-threatening disease with limited therapeutic options， highlighting the urgent need for innovative drug discovery strategies. To address this challenge， the authors propose the formula-originated rational intelligent screening&translation（FIRST）， a systematic framework for developing anti-fibrotic monomers derived from classical traditional Chinese medicine（TCM）. The strategy integrates three key dimensions， including tissue-oriented intelligent screening of active compounds， structural optimization based on drug-target spatial interactions and plant biosynthetic pathways， and cross-scale validation of drug. We further highlight its applications in discovering tissue-oriented novel drugs from clinically validated TCM， the development and mechanistic elucidation of anti-fibrotic therapeutics， as well as the clinical translation and secondary development of candidate drugs. This strategy paves the way for first-in-class， formula-derived monomeric drugs with defined structures， clarified mechanisms， and proven safety， offering a transformative avenue to meet the urgent therapeutic needs of PF and setting a new paradigm for TCM-based drug innovation.

Pulmonary fibrosis（PF） is a progressive and life-threatening disease with limited therapeutic options， highlighting the urgent need for innovative drug discovery strategies. To address this challenge， the authors propose the formula-originated rational intelligent screening&translation（FIRST）， a systematic framework for developing anti-fibrotic monomers derived from classical traditional Chinese medicine（TCM）. The strategy integrates three key dimensions， including tissue-oriented intelligent screening of active compounds， structural optimization based on drug-target spatial interactions and plant biosynthetic pathways， and cross-scale validation of drug. We further highlight its applications in discovering tissue-oriented novel drugs from clinically validated TCM， the development and mechanistic elucidation of anti-fibrotic therapeutics， as well as the clinical translation and secondary development of candidate drugs. This strategy paves the way for first-in-class， formula-derived monomeric drugs with defined structures， clarified mechanisms， and proven safety， offering a transformative avenue to meet the urgent therapeutic needs of PF and setting a new paradigm for TCM-based drug innovation.

OBJECTIVE To construct a preoperative medication reconciliation model assisted by a localized large language model （LLM） for gastric cancer and evaluate its clinical efficacy. METHODS A total of 249 gastric cancer patients with a history of continuous medication before admission in the Gastric Surgery Department of Jiangsu Cancer Hospital were retrospectively enrolled. Patients were divided into training set （154 cases） and validation set （95 cases） based on the order of time. Based on guidelines， drug package inserts， and other evidence， a standardized medication reconcili ation process and a structured knowledge base were constructed. DeepSeek-V3 LLM was deployed privately in the hospital， combined with retrieval-augmented generation technology， to achieve automated integration of medication information， risk screening， and generation of personalized recommendations. The quality of LLM-generated recommendations was evaluated using automatic metrics （BERT Score and ROUGE-1， 2， L） and manual scoring [seven-dimensional index （7DI） ] . Spearman correlation analysis was performed to explore the correlation between automatic scores and manual scores. Cronbach’s α coefficient was used to test the internal consistency of manual scoring results. The time consumed by manual and LLM-assisted medication reconciliation was compared across tasks of different difficulty levels （simple， moderate， and high）. RESULTS A structured knowledge base covering 8 major drug categories was finally established， covering common and high-risk preoperative medication scenarios and providing structured retrieval support for the LLM. For automatic evaluation， the precision， recall， and F1-score of BERT Score were 0.783±0.033， 0.811±0.038， and 0.796±0.028， respectively. The F1-scores of ROUGE-1， ROUGE-2 and ROUGE-L were 0.566±0.067， 0.338±0.076 and 0.468±0.082， respectively. The 7DI scores from three manual raters ranged from 32.06 to 33.45. The F1-score of automatic scoring was significantly positively correlated with the 7DI score of manual scoring （maximum coefficient of determination＝0.611， P ＜0.001）， and the internal consistency of manual scoring was good （Cronbach’s α ＝ 0.876）. In terms of efficiency， LLM-assisted medication reconciliation reduced time consumption by more than 90% compared with manual reconciliation in the simple， moderate， and high-difficulty groups （ P ＜0.001）. CONCLUSIONS The medication reconciliation model constructed based on a localized LLM and structured knowledge base shows high accuracy， consistency， and clinical applicability in complex preoperative medication scenarios for gastric cancer. It can improve the efficiency of medication reconciliation and reduce potential medication risks.