ObjectiveTo elucidate the critical role of rehabilitation medical records (including electronic records) in rehabilitation medicine's clinical practice and management, comprehensively analyzed the structure, core content and data standards of rehabilitation medical records, to develop a standardized medical record data architecture and core dataset suitable for rehabilitation medicine and to explore the application of rehabilitation data in performance evaluation and payment. MethodsBased on the regulatory documents Basic Specifications for Medical Record Writing and Basic Specifications for Electronic Medical Records (Trial) issued by National Health Commission of China, and referencing the World Health Organization (WHO) Family of International Classifications (WHO-FICs) classifications, International Classification of Diseases (ICD-10/ICD-11), International Classification of Functioning, Disability and Health (ICF), and International Classification of Health Interventions (ICHI Beta-3), this study constructed the data architecture, core content and data standards for rehabilitation medical records. Furthermore, it explored the application of rehabilitation record summary sheets (home page) data in rehabilitation medical statistics and payment methods, including Diagnosis-related Groups (DRG), Diagnosis-Intervention Packet (DIP) and Case Mix Index. ResultsThis study proposed a systematic standard framework for rehabilitation medical records, covering key components such as patient demographics, rehabilitation diagnosis, functional assessment, rehabilitation treatment prescriptions, progress evaluations and discharge summaries. The research analyzed the systematic application methods and data standards of ICD-10/ICD-11, ICF and ICHI Beta-3 in the fields of medical record terminology, coding and assessment. Constructing a standardized data structure and data standards for rehabilitation medical records can significantly improve the quality of data reporting based on the medical record summary sheet, thereby enhancing the quality control of rehabilitation services, effectively supporting the optimization of rehabilitation medical insurance payment mechanisms, and contributing to the establishment of rehabilitation medical performance evaluation and payment based on DRG and DIP. ConclusionStructured rehabilitation records and data standardization are crucial tools for quality control in rehabilitation. Systematically applying the three reference classifications of the WHO-FICs, and aligning with national medical record and electronic health record specifications, facilitate the development of a standardized rehabilitation record architecture and core dataset. Standardizing rehabilitation care pathways based on the ICF methodology, and developing ICF- and ICD-11-based rehabilitation assessment tools, auxiliary diagnostic and therapeutic systems, and supporting terminology and coding systems, can effectively enhance the quality of rehabilitation records and enable interoperability and sharing of rehabilitation data with other medical data, ultimately improving the quality and safety of rehabilitation services.

ObjectiveTo investigate the effect and mechanism of cordycepin in inhibiting fat infiltration after rotator cuff injuries in rats by regulating the Wnt/β-catenin signaling pathway， providing a theoretical basis for clinical treatment of rotator cuff injuries. MethodsFifty SPF-grade female SD rats were used in this study， with 10 randomly selected as the blank group. A rotator cuff injury repair model was established by supraspinatus tendon and suprascapular nerve compression. The successfully modeled rats were randomized into model and low-dose （20 mg·kg^-1）， medium-dose （40 mg·kg^-1）， and high-dose （80 mg·kg^-1） cordycepin groups. After 6 weeks of treatment， the gait analysis was performed to assess the limb function in rats. Oil red O staining and Masson staining were employed to observe pathological changes in the muscle tissue. Enzyme-linked immunosorbent assay （ELISA） was used to measure the levels of interleukin-1β （IL-1β）， interleukin-6 （IL-6）， and tumor necrosis factor-α （TNF-α） in the serum. Immunohistochemistry （IHC） was employed to detect the expression of peroxisome proliferator-activated receptor γ （PPARγ） and CCAAT/enhancer-binding protein α （C/EBPα）， which are markers of adipogenesis. Real-time fluorescence quantitative polymerase chain reaction （Real-time PCR） and Western blot were employed to determine the mRNA and protein levels， respectively， of Wnt3a， Wnt10b， and β-catenin. ResultsCompared with the blank group， the model group showed decreases in stride length and paw print area （P<0.01）， an increase in ratio of wet muscle mass reduction and a decrease in muscle fiber cross-sectional area （P<0.05）， and decreased ratios of fat infiltration area and collagen fiber area （P<0.01）. Additionally， the model group showed elevated levels of IL-1β， IL-6， and TNF-α （P<0.05）， up-regulated protein levels of PPARγ and C/EBPα （P<0.01）， and down-regulated mRNA and protein levels of Wnt3a， Wnt10b， and β-catenin （P<0.05， P<0.01）. Compared with the model group， the low-， medium-， and high-dose cordycepin groups showed increases in stride length and paw print area （P<0.01）， a decrease in ratio of wet muscle mass reduction and an increase in muscle fiber cross-sectional area （P<0.05）， and increases in ratios of fat infiltration area and collagen fiber area （P<0.05， P<0.01）. In addition， cordycepin lowered the serum levels of IL-1β， IL-6， and TNF-α （P<0.05， P<0.01）， down-regulated the protein levels of PPARγ and C/EBPα （P<0.01）， and up-regulated the mRNA and protein levels of Wnt3a， Wnt10b， and β-catenin （P<0.05， P<0.01）. ConclusionCordycepin can improve the limb function， alleviate rotator cuff muscle atrophy， fat infiltration， and fibrosis， and inhibit inflammation in rats by regulating the Wnt/β-catenin signaling pathway.

Objective To construct a predictive model for the survival of transplant kidneys after kidney transplantation. Methods The clinical data of 366 kidney transplant recipients and donors were retrospectively analyzed, and the recipients were divided into low-risk group (n=101), medium-risk group (n=189), and high-risk group (n=76) based on the kidney donor profile index (KDPI). Each group was further divided into Remuzzi score ≤3 group and Remuzzi score >3 group based on time-zero biopsy Remuzzi scores. Kaplan-Meier method was used to analyze the survival of transplant kidneys. Univariate and multivariate Cox regression analyses were performed to identify risk factors affecting long-term survival after kidney transplantation. A predictive model for transplant kidney survival was established and a nomogram was drawn. The predictive performance of the model was evaluated using the receiver operating characteristic (ROC) curve and the area under the curve (AUC). Results The median KDPI was 65%, and the median Remuzzi score was 3. The 5-year survival rate of transplant kidneys was 83.5%. Kaplan-Meier survival curves showed that in the KDPI medium-risk and KDPI high-risk groups, the subgroup with lower Remuzzi score had a higher survival rates of transplant kidneys than the subgroup with higher Remuzzi score. Univariate and multivariate Cox regression analyses showed that KDPI, Remuzzi score, and donor’s age were independent risk factors for transplant kidney loss (all P<0.05). The ROC curve showed that the AUC of the nomogram prediction model established based on independent risk factors for the 1, 3 and 5-year survival rates of transplant kidneys were 0.91, 0.93 and 0.94 for the training set, and 0.89, 0.85 and 0.88 for the validation set. Calibration curves shows good consistency between the training and validation sets of the model. Conclusions The nomogram predictive model based on KDPI, time-zero biopsy Remuzzi score and donor’s age has good predictive value for transplant kidney survival.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective To compare the performance and efficacy of prognostic models constructed by different machine learning algorithms in predicting the survival period of lung transplantation (LTx) recipients. Methods Data from 483 recipients who underwent LTx were retrospectively collected. All recipients were divided into a training set and a validation set at a ratio of 7:3. The 24 collected variables were screened based on variable importance (VIMP). Prognostic models were constructed using random survival forest (RSF) and extreme gradient boosting tree (XGBoost). The performance of the models was evaluated using the integrated area under the curve (iAUC) and time-dependent area under the curve (tAUC). Results There were no significant statistical differences in the variables between the training set and the validation set. The top 15 variables ranked by VIMP were used for modeling and the length of stay in the intensive care unit (ICU) was determined as the most important factor. Compared with the XGBoost model, the RSF model demonstrated better performance in predicting the survival period of recipients (iAUC 0.773 vs. 0.723). The RSF model also showed better performance in predicting the 6-month survival period (tAUC _{6 months} 0.884 vs. 0.809, P = 0.009) and 1-year survival period (tAUC _{1 year} 0.896 vs. 0.825, P = 0.013) of recipients. Based on the prediction cut-off values of the two algorithms, LTx recipients were divided into high-risk and low-risk groups. The survival analysis results of both models showed that the survival rate of recipients in the high-risk group was significantly lower than that in the low-risk group (P<0.001). Conclusions Compared with XGBoost, the machine learning prognostic model developed based on the RSF algorithm may preferably predict the survival period of LTx recipients.

Objective To evaluate the performance of the artificial intelligence (AI)-enabled snail identification system for recognition of Oncomelania hupensis robertsoni and Tricula in schistosomiasis-endemic areas of Yunnan Province. Methods Fifty O. hupensis robertsoni and 50 Tricula samples were collected from Yongbei Township, Yongsheng County, Lijiang City, a schistosomiasis-endemic area in Yunnan Province in May 2024. A total of 100 snail sample images were captured with smartphones, including front-view images of 25 O. hupensis robertsoni and 25 Tricula samples (upward shell opening) and back-view images of 25 O. hupensis robertsoni and 25 Tricula samples (downward shell opening). Snail samples were identified as O. hupensis robertsoni or Tricula by schistosomiasis control experts with a deputy senior professional title and above according to image quality and morphological characteristics. A standard dataset for snail image classification was created, and served as a gold standard for recognition of snail samples. A total of 100 snail sample images were recognized with the AI-enabled intelligent snail identification system based on a WeChat mini program in smartphones. Schistosomiasis control professionals were randomly sampled from stations of schistosomisis prevention and control and centers for disease control and prevention in 18 schistosomiasis-endemic counties (districts, cities) of Yunnan Province, for artificial identification of 100 snail sample images. All professionals are assigned to two groups according the median years of snail survey experiences, and the effect of years of snail survey experiences on O. hupensis robertsoni sample image recognition was evaluated. A receiver operating characteristic (ROC) curve was plotted, and the sensitivity, specificity, accuracy, Youden’s index and the area under the curve (AUC) of the AI-enabled intelligent snail identification system and artificial identification were calculated for recognition of snail sample images. The snail sample image recognition results of AI-enabled intelligent snail identification system and artificial identification were compared with the gold standard, and the internal consistency of artificial identification results was evaluated with the Cronbach’s coefficient alpha. Results A total of 54 schistosomiasis control professionals were sampled for artificial identification of snail sample image recognition, with a response rate of 100% (54/54), and the accuracy, sensitivity, specificity, Youden’s index, and AUC of artificial identification were 90%, 86%, 94%, 0.80 and 0.90 for recognition of snail sample images, respectively. The overall Cronbach’s coefficient alpha of artificial identification was 0.768 for recognition of snail sample images, and the Cronbach’s coefficient alpha was 0.916 for recognition of O. hupensis robertsoni snail sample images and 0.925 for recognition of Tricula snail sample images. The overall accuracy of artificial identification was 90% for recognition of snail sample images, and there was no significant difference in the accuracy of artificial identification for recognition of O. hupensis robertsoni (86%) and Tricula snail sample images (94%) (χ² = 1.778, P > 0.05). There was no significant difference in the accuracy of artificial identification for recognition of snail sample images with upward (88%) and downward shell openings (92%) (χ² = 0.444, P > 0.05), and there was a significant difference in the accuracy of artificial identification for recognition of snail sample images between schistosomiasis control professionals with snail survey experiences of 6 years and less (75%) and more than 6 years (90%) (χ² = 7.792, P < 0.05). The accuracy, sensitivity, specificity and AUC of the AI-enabled intelligent snail identification system were 88%, 100%, 76% and 0.88 for recognition of O. hupensis robertsoni snail sample images, and there was no significant difference in the accuracy of recognition of O. hupensis robertsoni snail sample images between the AI-enabled intelligent snail identification system and artificial identification (χ² = 0.204, P > 0.05). In addition, there was no significant difference in the accuracy of artificial identification for recognition of snail sample images with upward (90%) and downward shell openings (86%) (χ² = 0.379, P > 0.05), and there was a significant difference in the accuracy of artificial identification for recognition of snail sample images between schistosomiasis control professionals with snail survey experiences of 6 years and less and more than 6 years (χ² = 5.604, P_adjusted < 0.025). Conclusions The accuracy of recognition of snail sample images is comparable between the AI-enabled intelligent snail identification system and artificial identification by schistosomiasis control professionals, and the AI-enabled intelligent snail identification system is feasible for recognition of O. hupensis robertsoni and Tricula in Yunnan Province.

OBJECTIVE To provide suggestions for improving the path and system construction of patient engagement in drug regulation in China. METHODS By reviewing initiatives and experiences from the United States （U. S.）， European Union （EU）， and Japan in promoting patient engagement， this study summarizes the roles and contributions of patients in the entire drug regulatory process internationally. Combining China’s current progress and challenges in patient engagement， specific proposals are formulated to refine regulatory pathways and institutional systems. RESULTS & CONCLUSIONS With growing global emphasis on patient engagement as a regulatory strategy， countries or regions such as the U.S.， EU， and Japan have established clear policies， designated oversight agencies， and developed diversified pathways for patient engagement. Patients contribute to regulatory processes through advisory meetings， direct decision-making roles， and leveraging lived experiences and expertise to optimize drug evaluation and monitoring. In contrast， China’s patient engagement remains primarily limited to clinical value- oriented drug development， lacking formal policy guidance. It is recommended that China， based on its existing policy system， further strengthen the construction of a safeguard system for patient engagement， improve the capacity building and pathway models for patient participation in pharmaceutical regulation， and promote the continuous development of patient engagement in pharmaceutical regulation in our country.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.