OBJECTIVE:Nowadays,machine learning algorithms are gradually being applied to predict stroke and cardiovascular disease.Compared with traditional regression models,machine learning can learn from data to achieve high prediction accuracy by exploring the flexible relationship between a large number of predictive features and outcome variables,providing a new method for the formulation of individualized treatment and rehabilitation programs.This study aims to systematically evaluate stroke functional recovery and prognosis prediction models based on machine learning,comprehensively assessing their predictive performance and clinical application potential to provide references for the development,application,and promotion of related predictive models.METHODS:This review was conducted following the PRISMA(Preferred Reporting Items for Systematic Reviews and Meta-Analyses)guidelines.Relevant literature on stroke prognosis prediction using machine learning methods was selected by searching PubMed,EMbase,Web of Science Core Collection,CNKI,WanFang,and the China Biomedical Literature Database,with the search period from January 1,2014,to July 1,2024.Two researchers independently screened the literature and extracted data based on inclusion and exclusion criteria,using the Prediction model Risk Of Bias ASsessment Tool(PROBAST)to assess model quality.RESULTS:(1)A total of 3 126 articles were obtained in the preliminary search.After screening and exclusion,18 articles were finally included.150 prediction models were constructed using 13 machine learning methods.The three most frequently used methods are Logistic Regression,Random Forest,and Extreme Gradient Boosting(XGBoost).Only one study was externally validated.Eight studies reported how the missing data were handled.(2)In terms of outcome indicators,8 studies used the combination of clinical data and imaging data to build models,9 studies only used clinical data to build models,and 1 study only used imaging data to build models.(3)Each of the 18 studies gave the most important characteristics of the study,with the most mentioned being the National Institute of Health Stroke Scale and age.All studies reported area under curve values ranging from 0.74 to 0.96,with the highest area under curve being 0.96.The overall risk of bias in all models was high.The high risk of bias in the field of model analysis was the main reason for the high risk of overall bias in all models.(4)The results of meta-analysis showed that age and National Institute of Health Stroke Scale score had significant influence on stroke prognosis,with age[MD=8.49,95％CI(6.24,10.75),P＜0.01]and National Institute of Health Stroke Scale score[MD=4.78,95％CI(2.56,7.00),P＜0.01].CONCLUSION:This study systematically evaluated the predictive model of functional recovery and prognosis of stroke based on machine learning,and all the models have good predictive potential.However,future studies should increase the sample size of the included model,adopt prospective studies,and add external validation of the model to improve the stability and prediction accuracy of the model,control the risk of bias,and contribute to the validation and promotion of the model in practical clinical applications.At the same time,the interpolation of missing values is more transparent and accurate.Although existing machine learning models show good predictive performance,it is also important to focus on the functionality and usability of the model,and the inclusion of features will reduce ease of use.We should develop easy to use model interfaces and user-friendly clinical tools to enable medical staff to better apply the model for clinical decision.

OBJECTIVE:Nowadays,machine learning algorithms are gradually being applied to predict stroke and cardiovascular disease.Compared with traditional regression models,machine learning can learn from data to achieve high prediction accuracy by exploring the flexible relationship between a large number of predictive features and outcome variables,providing a new method for the formulation of individualized treatment and rehabilitation programs.This study aims to systematically evaluate stroke functional recovery and prognosis prediction models based on machine learning,comprehensively assessing their predictive performance and clinical application potential to provide references for the development,application,and promotion of related predictive models.METHODS:This review was conducted following the PRISMA(Preferred Reporting Items for Systematic Reviews and Meta-Analyses)guidelines.Relevant literature on stroke prognosis prediction using machine learning methods was selected by searching PubMed,EMbase,Web of Science Core Collection,CNKI,WanFang,and the China Biomedical Literature Database,with the search period from January 1,2014,to July 1,2024.Two researchers independently screened the literature and extracted data based on inclusion and exclusion criteria,using the Prediction model Risk Of Bias ASsessment Tool(PROBAST)to assess model quality.RESULTS:(1)A total of 3 126 articles were obtained in the preliminary search.After screening and exclusion,18 articles were finally included.150 prediction models were constructed using 13 machine learning methods.The three most frequently used methods are Logistic Regression,Random Forest,and Extreme Gradient Boosting(XGBoost).Only one study was externally validated.Eight studies reported how the missing data were handled.(2)In terms of outcome indicators,8 studies used the combination of clinical data and imaging data to build models,9 studies only used clinical data to build models,and 1 study only used imaging data to build models.(3)Each of the 18 studies gave the most important characteristics of the study,with the most mentioned being the National Institute of Health Stroke Scale and age.All studies reported area under curve values ranging from 0.74 to 0.96,with the highest area under curve being 0.96.The overall risk of bias in all models was high.The high risk of bias in the field of model analysis was the main reason for the high risk of overall bias in all models.(4)The results of meta-analysis showed that age and National Institute of Health Stroke Scale score had significant influence on stroke prognosis,with age[MD=8.49,95％CI(6.24,10.75),P＜0.01]and National Institute of Health Stroke Scale score[MD=4.78,95％CI(2.56,7.00),P＜0.01].CONCLUSION:This study systematically evaluated the predictive model of functional recovery and prognosis of stroke based on machine learning,and all the models have good predictive potential.However,future studies should increase the sample size of the included model,adopt prospective studies,and add external validation of the model to improve the stability and prediction accuracy of the model,control the risk of bias,and contribute to the validation and promotion of the model in practical clinical applications.At the same time,the interpolation of missing values is more transparent and accurate.Although existing machine learning models show good predictive performance,it is also important to focus on the functionality and usability of the model,and the inclusion of features will reduce ease of use.We should develop easy to use model interfaces and user-friendly clinical tools to enable medical staff to better apply the model for clinical decision.

【Objective:】 To explore the network characteristics of sleep quality and anxiety in first-line medical staff fighting against COVID-19, further understand the relationship between sleep quality and anxiety, and provide basis for intervention. 【Methods:】 Using the convenient sampling method, this paper used the Pittsburgh Sleep Quality Index (PSQI) and Self Rating Anxiety Scale (SAS) to conduct a questionnaire survey on the front-line medical staff who fought against the epidemic during the COVID-19. Network analysis was used to construct sleep quality and anxiety network, and R language was used for statistical analysis and visualization. 【Results:】 In the network of sleep quality and anxiety of first-line medical staff fighting against COVID-19, "sleep disorder" and "sleep quality", "unfortunate premonition" and "inability to sit still", "syncope" and "hand and foot tingling" were highly related. "Fatigue", "dizziness" and "panic" had the highest expected influence. "Sleep quality", "sleep disorder" and "fatigue" had the highest bridge expected influence. The average predictability value of all nodes was 0.778. 【Conclusion:】 This paper used network analysis to explore the sleep quality and anxiety of first-line medical staff fighting against COVID-19 and found that there was a unique correlation path between them. Intervention against core symptoms can ameliorate anxiety and sleep problems to the great extent, and provide guidance for improving the physical and mental health.

OBJECTIVE To exc avate the adverse drug event （ADE）signals of semaglutide and provide reference for its clinical rational use. METHODS The proportional unbalance method was used to mine the signals of all semaglutide ADE reports from FDA Adverse Event Reporting System （FAERS）up to September 2021. The basic situations of the reported cases were analyzed. The corresponding system organ classification （SOC）was mapped and compared with the adverse drug reactions recorded in the drug instructions. Preferred terms （PT）of patients with different indications were analyzed. RESULTS A total of 6 661 semaglutide ADE reports were extracted and 194 valid signals were mined. Among 6 661 cases of ADE ，the proportion of men （43.40％）was lower than women （52.65％）；the age was mainly distributed in ＞40-65 years old （29.00％）and ＞65 years old （22.61％）；the reporting country was mainly the United States （83.88％）；the report year was mainly concentrated in 2021 （40.88％），with an increasing trend year by year ；the main outcome was hospitalization or prolonged hospitalization in serious ADE reports （17.78％）. Semaglutide ADE signal was mapped to the main SOC ，mainly including gastrointestinal diseases ，various injuries，poisoning and operation complications ，metabolic and nutritional diseases ，various examinations. The screening criteria were based on the report odds ratio ＞10 or ADE reported cases ＞50，and 48 new potential adverse drug reactions were added to the drug description. Among the indications with the top two reported cases （type 2 diabetes and obesity ，overweight，weight control），the frequency of gastrointestinal system related ADE reports represented by nausea ，vomiting and diarrhea was higher ， which was similar to the drug instructions. CONCLUSIONS This study supplemented 48 new potential adverse drug reactions based on the drug instructions of semaglutide. At present ，it can be considered that semaglutide is safe.

Objective:To observe the expression of hypoxia-inducible factor 1 alpha (HIF-1α) signaling pathway in the aorta of chronic kidney disease (CKD) rats with vascular calcification and to explore the role of this pathway in aortic calcification of CKD.Methods:Forty 8-week-old male SD rats were randomly divided into control group (CON group, n=15) and CKD with aortic calcification group (CKD+AC group, n=25). The rats were sacrificed at the end of 4 th, 6 th and 8 th week respectively and urine, blood, aorta and kidney samples were collected. The level of serum HIF-1α was tested by ELISA. The pathology changes of kidney were observed by HE staining. The aortic calcification was evaluated by alizarin red staining and calcium content detection. Immunohistochemistry and real-time PCR were applied to detect the protein and mRNA expression of alpha-smooth muscle actin (α-SMA), Runt-related transcription factor 2 (Runx2), HIF-1α, vascular endothelial growth factor A (VEGFA) and Notch1 in the aorta. Results:Compared with CON group, serum urea, creatinine, cystatin C, phosphorus, calcium-phosphorus product and 24-h urine protein were significantly higher in CKD+AC group (all P<0.05). Serum HIF-1α levels were higher at 4 th and 8 th week in CKD+AC group than that in CON group (both P<0.05). There was no significant calcium deposit in the aorta of the CON group at all time points, and calcium deposits were seen in the aorta of the CKD+AC rats at each time point, which gradually increased with time. Compared with CON group, the expressions of aortic α-SMA protein and mRNA were significantly decreased in CKD+AC group at each time point, however the protein and mRNA expressions of Runx2, HIF-1α, VEGFA and Notch1 in the aorta of CKD+AC group rats were markedly increased at each time point (all P<0.05). Correlation analysis showed that the aortic calcium content was positively correlated with serum HIF-1α ( r=0.706, P<0.001) and the protein expressions of HIF-1α ( r=0.852, P<0.001), VEGFA ( r=0.747, P<0.001) and Notch1 ( r=0.813, P<0.001) in aorta. Conclusion:The HIF-1α-VEGFA-Notch1 signaling pathway is activated during aortic calcification in CKD rats, suggesting that this signaling pathway might be involved in the vascular calcification in CKD, and serum HIF-1α is expected to be one of serum markers for CKD vascular calcification.