BACKGROUND:Exercise as a form of active rehabilitation can improve the dysfunction caused by peripheral nerve injury,and different exercise modalities target different lesion sites and recovery mechanisms. OBJECTIVE:To comprehensively analyze the application and mechanisms of different exercise modalities in functional recovery from peripheral nerve injury. METHODS:A computerized search was conducted in PubMed and CNKI databases for relevant literature published before January 2024.The search terms used were"peripheral nerve injury,spinal cord,exercise,cerebral cortex,muscle atrophy,mirror therapy,blood flow restriction training"in both English and Chinese.Finally,77 articles were included for review. RESULTS AND CONCLUSION:Peripheral nerve injury can cause systemic pathological changes such as skeletal muscle atrophy,corresponding spinal cord segmental lesions,and sensorimotor cortex remodeling.Aerobic exercise can improve dysfunction by enhancing the immune response,promoting glial cell polarization,and promoting the release of nerve growth factor.Blood flow restriction exercise can regulate the secretion of muscle growth factor,promote muscle growth and enhance muscle strength.Mirror movement has a good effect in activating the cerebral cortex and reducing cortical remodeling.Different exercise modalities have potential benefits in functional recovery after peripheral nerve injury;however,there are still some problems and challenges,such as the choice of exercise modalities,the control of exercise intensity and frequency,and the detailed analysis of mechanisms.

Clinical practice guidelines represent the best recommendations for patient care. They are developed through systematically reviewing currently available clinical evidence and weighing the relative benefits and risks of various interventions. However, clinical practice guidelines have to go through a long translation cycle from development and revision to clinical promotion and application, facing problems such as scattered distribution, high duplication rate, and low actual utilization. At present, the clinical practice guideline information platform can directly or indirectly solve the problems related to the lengthy revision cycles, decentralized dissemination and limited application of clinical practice guidelines. Therefore, this paper systematically examines different types of clinical practice guideline information platforms and investigates their corresponding challenges and emerging trends in platform design, data integration, and practical implementation, with the aim of clarifying the current status of this field and providing valuable reference for future research on clinical practice guideline information platforms.

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.

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: To construct a hypertension risk prediction model for residents aged 18-79 years, so as to provide an assessment tool for early screening and prevention of hypertension in high-risk groups. Methods: The permanent residents aged 18-79 years from 6 townships (streets) in Fangshan District of Beijing Municipality were selected as the study subjects using a multi-stage stratified random sampling method from March to June 2023. Demographic information, lifestyle, body mass index (BMI), blood pressure, fasting blood glucose, and blood lipid were collected through questionnaire survey, physical examination, and laboratory tests. Subjects were randomly divided into training and validation sets at a 7∶3 ratio. The logistic regression model was used to screen the risk factors of hypertension, and a hypertension risk prediction nomogram was constructed. Receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis were used to verify the discrimination, fit, and clinical application value of the model. Results: A total of 4 438 subjects were included, including 2 365 males (53.29%) and 2 073 females (46.71%), with a mean age of (44.99±14.90) years. The prevalence of hypertension was 35.29% (1 566 cases), and the standardized prevalence was 24.74%. The logistic regression model screened out 9 influencing factors of hypertension. The nomogram was established as ln[p/ (1-p)]= -2.873 + 0.935×40-<50 years + 1.463×50-<60 years + 1.908×60-<70 years + 2.346×70-79 years + 0.298×male-0.675×college degree or above + 0.384×smoking + 0.227×drinking + 0.572×overweight + 1.449×obesity + 0.557×heart rate ≥80 beats/min + 0.428×diabetes + 0.484×dyslipidemia. The area under the ROC curve of the validation set was 0.821 (95%CI: 0.798-0.843), and the calibration curve results showed that the calibration curve fitted the actual curve well. Decision curve analysis showed that the threshold probability was in the range of 0.10 to 0.70, and the model had good predictive value and clinical application value. Conclusion The nomogram based on age, gender, educational level, smoking, drinking, body mass index, heart rate, diabetes, and dyslipidemia can be used to predict the risk of hypertension among residents aged 18-79 years.

ObjectiveThis study aims to understand the recognition of practitioners in traditional Chinese medicine （TCM） hospitals on hospital-based health technology assessment （HB-HTA）， assessment needs， challenges， and suggestions， so as to provide references for the future work. MethodThe convenient sampling method was adopted to survey the relevant practitioners in TCM hospitals. The questionnaire included 39 questions in 4 dimensions and was distributed through the online platform Weijuanxing. ResultA total of 244 questionnaires were recovered， and the obtained data were analyzed in SPSS. The results showed that 137 practitioners were very familiar with HB-HTA and there was no significant difference in the recognition of practitioners in different occupations （F=0.251； P=0.778）. The practitioners in Hong Kong， Macao， and Taiwan had lower recognition than those in other regions. In terms of the assessment needs， 127 practitioners believed that it was very necessary to carry out HB-HTA in TCM hospitals in the future. Chinese patent medicines/Chinese herbal medicine decoction pieces （5.91） and TCM appropriate technology （5.57） had higher assessment priority scores. The assessment needs were high for the effectiveness （235 practitioners） and safety （224 practitioners） of health technology. The lack of specialized organization and standardized evaluation process system and the shortage of talents were considered to be the major challenges for the future development in this field. ConclusionThe stakeholders carrying out the health technology assessment in TCM hospitals had certain awareness of HB-HTA. Most practitioners believed that it was necessary to carry out HB-HTA in TCM hospitals in the future， while the work might face challenges such as the lack of organizations and system and the shortage of talents， which requires policy support.