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 analyze the characteristics of brain functional networks in patients with post-stroke aphasia(PSA)during a Chinese word-picture matching task.Methods A total of 18 PSA patients in Dongzhimen Hospital from January,2018 to December,2021 were enrolled as PSA group,and nine healthy controls matched for sex,age and education were included as the control group.The Chinese Rehabilitation Research Center Aphasia Examination(CRRCAE)and task-state electroencephalogram(EEG)data based on a Chinese word-picture matching paradigm were collected.Source-space reconstruction was applied to EEG signals to construct functional connectivity matrices.Graph-theoretical analysis was used to com-pute global network properties,and network-based statistics were used to identify subnetwork differences be-tween groups.Correlations between global network properties and CRRCAE subscales were further analyzed.Results The global properties of each frequency band were higher in the control group(unmatched)than in PSA group and the control group(matched)(P<0.01).Subnetworks connections enhanced in the alpha band in the frontal,temporal,parietal,occipital lobes and limbic system(23 nodes,31 edges,P<0.05),and weakened connections in the frontal,temporal,limbic system and basal ganglia(20 nodes,26 edges,P<0.01)in PSA group;in the beta band,subnetwork connections enhanced in the frontal,temporal lobes,basal ganglia and limbic system(15 nodes,23 edges,P<0.01);in the theta band,subnetwork connections weakened in the left frontal and temporal lobes(10 nodes,11 edges,P<0.05),and enhanced in the right frontal and temporal lobes(7 nodes,7 edges,P<0.05).Under the matched condition,the global properties of the alpha and beta bands in PSA group were positive-ly correlated with reading(r=0.511 to 0.650,all P<0.05),and the local efficiency and average clustering coeffi-cient of the theta band were negatively correlated with repetition(r=-0.500 to-0.505,all P<0.05);under the unmatched condition,the local efficiency and average clustering coefficient of the alpha and beta bands in PSA group were positively correlated with reading(r=0.522 to 0.642,all P<0.05),and the global efficiency and lo-cal efficiency of the alpha band were positively correlated with auditory comprehension(r=0.486 to 0.496,all P<0.05).The comparison between the task state and the resting state further revealed that compensatory connec-tion enhanced in the alpha band in the frontal lobe and limbic system(6 nodes,5 edges,P<0.05)and in the beta band in the frontal,temporal,parietal,occipital lobes and limbic system(38 nodes,52 edges,P<0.01)in PSA pa-tients.Conclusion The core network damage characteristics of PSA patients are abnormally decreased global properties across multiple frequency bands and frequency-specific abnormal network reorganization,suggesting the overall brain network efficiency decline coexists with local compensation.This provides an objective basis for the neuroelec-trophysiological evaluation and targeted intervention of PSA.

Objective To understand the current situation of healthcare-associated infection(HAI)in China,pro-vide data support and decision-making basis for formulating scientific and effective strategies for HAI prevention and control.Methods A nationwide cross-sectional survey on HAI was conducted among various types and levels of medical institutions in China according to a unified protocol of bedside surveys and case investigations.Results In 2024,a total of 5 736 medical institutions and 2 751 765 patients were surveyed.Among them,34 889 HAI cases were identified,with a prevalence rate of 1.27％.The number of HAI episodes was 38 032,and case prevalence rate was 1.38％.The prevalence rate of HAI in medical institutions in different regions of China ranged from 0.66％to 2.35％.Among medical institutions of different scales,those with a bed capacity of ≥900 had the high-est incidence of HAI,reaching 1.65％.The most common infection site was the lower respiratory tract(44.66％),followed by the urinary tract(12.94％),surgical site(9.32％),upper respiratory tract(7.02％),and bloodstream infection(5.78％).The top 3 departments with the highest HAI rates were the general intensive care unit(10.02％),department of neurosurgery(5.51％),and department(group)of hematology(5.34％).A total of 23 238 strains of HAI pathogens were detected,with 10 714 strains(46.10％)from lower respiratory tract speci-mens.The top 5 detected strains were Klebsiella pneumoniae(14.76％),Pseudomonas aeruginosa(13.33％),Escherichia coli(12.79％),Acinetobacter baumannii(9.23％),and Staphylococcus aureus(7.88％).231 944 pa-tients underwent class Ⅰ incision surgery were monitored,with 1 647 cases experienced surgical site infection,and the prevalence rate of surgical site infection was 0.71％.The number of patients who should undergo pathogen de-tection(patients receiving therapeutic and therapeutic combined prophylactic antimicrobial agents)was 715 179,while the actual number was 480 492,with a pathogen detection rate of 67.18％.425 225 patients received patho-genic detection before treatment,with a detection rate of 59.46％.Conclusion The overall HAI prevalence in Chi-na is lower,showing disparities among medical institutions of different regions and scales.Therefore,precise imple-mentation of measures is necessary for HAI prevention and control,with a focus on high-risk institutions and high-risk departments,key areas,and critical procedures.All levels of medical institutions should continuously reduce the incidence of HAI by strengthening monitoring,standardizing the use of antimicrobial agents,and reinforcing basic HAI prevention and control measures.

Objective:To investigate the intervention effect of five elements music of traditional Chinese medicine and western classical music on the sleep quality of stroke patients, and to compare the difference between the two, to provide a reference for the clinical care measures to improve the sleep quality of stroke patients.Methods:By adopting a randomized controlled trial, 75 stroke patients who were hospitalized in the rehabilitation hospital of the Second Affiliated Hospital of Shandong University of Traditional Chinese Medicine from July 2022 to July 2023 were selected by convenience sampling method as the study subjects, and the patients were randomly divided the conventional treatment group, the five-element music group, and the classical music group according to the method of randomized numerical table with 25 cases in each group. On the basis of general treatment, the conventional treatment group took sleep health education, the five elements music group were given five-element music intervention based on conventional treatment, and the classical music group were given classical music intervention based on conventional treatment. The changes of Athens Insomnia Scale (AIS), Pittsburgh Sleep Quality Index (PSQI) and polysomnography (PSG) examination parameters in the 3 groups were compared.Results:There were 2 cases of shedding in the conventional treatment group, and 23 cases were finally included, 15 males and 8 females, aged (54.02 ± 7.80) years; there were no cases of shedding in the five-element music group, and 25 cases were finally included, 12 males and 13 females, aged (53.69 ± 6.02) years; and there was 1 case of shedding in the classical music group, and 24 cases were finally included, 10 males and 14 females, aged (52.34 ± 7.08) years. Comparison of AIS, PSQI and PSG scores among the 3 groups of patients before intervention showed no statistically significant differences (all P>0.05). After the intervention, the AIS scores and PSQI scores of the 3 groups were (9.48 ± 1.53) and (12.22 ± 2.94), (6.76 ± 1.36) and (7.64 ± 2.08), (7.46 ± 1.38) and (10.33 ± 2.82), respectively, and the differences were statistically significant ( F=23.21, 18.44, both P<0.05). PSG sleep structure parameters showed that the total sleep time, REM latency time and sleep efficiency of the patients in the five-element music group after the intervention were (399.89 ± 51.76) min, (136.26 ± 25.36) min, (78.87 ± 8.21)%, higher than (368.45 ± 47.88) min, (124.46 ± 26.25) min, (73.36 ± 7.86)% in the classical music group and (345.48 ± 38.69) min, (111.37 ± 23.23) min, (69.44 ± 7.88)% in the conventional treatment group, and the differences between three groups were statistically significant ( F=8.27, 5.93, and 8.49, all P<0.05); the sleep latency time, awakening time, and number of awakenings in the five-element music group were (28.86 ± 17.68) min, (54.37 ± 25.15) min, and (2.72 ± 1.19) times, respectively, which were lower than those in the classical music group (35.67 ± 16.99) min, (64.28 ± 29.34) min, and (3.67 ± 1.12) times and (42.38 ± 18.96) min, (78.38 ± 37.26) min, (4.87 ± 1.46) times in the conventional treatment group, and the differences between three groups were statistically significant ( F=3.51, 3.66, and 17.56, all P<0.05). The results of the PSG sleep progression showed that the duration of the N1 stage of sleep in the five-element music group after the intervention was (95.71 ± 15.23) min, which was higher than (83.20 ± 18.34) min in the classical music group and (80.93 ± 16.47) min in the conventional treatment group, and the difference between three groups was statistically significant ( F=5.53, P<0.01); the N3 stage sleep time and the sleep percentage of the five-element music group after the intervention were respectively (84.23 ± 20.98) min and (23.98 ± 5.89)%, which were higher than (65.33 ± 18.82) min and (18.34 ± 3.78)% in the classical music group and (45.87 ± 18.65) min and (15.03 ± 5.56)% in the conventional treatment group, and the differences between three groups were statistically significant ( F=23.08, 18.50, both P<0.05). Conclusions:Both five elements music and classical music can improve the sleep quality of stroke patients, and the effect of five elements music to improve sleep is more significant.

Objective To investigate the relationship between the threonine and tyrosine kinase(TTK)gene and eyelid basal cell carcinoma(BCC).Methods Bioinformatics methods were used to screen the core gene(namely,TTK)associated with the occurrence and development of BCC from the Gene Expression Omnibus(GEO)database.Surgically removed eyelid BCC tissue specimens(BCC cells were divided into BBC Grade Ⅰ,Ⅱ and Ⅲ groups by tumor grade)and be-nign tumor tissue specimens(Control group)were collected from Ziyang Central Hospital for subsequent experiments.Cel-lular immunofluorescence assay(CIA)was used to detect the expression of the TTK gene in benign and malignant eyelid tumor cells.After knocking down TTK in BCC cells through transfection with lentiviruses(the cells transfected with LV-TTK-shRNA were taken as the TTK-shRNA group,and those transfected with LV-BBC-shRNA were taken as the BBC nega-tive control group),CIA was used to detect the expression of key proteins Bcl-2 and Bax in the apoptotic signaling pathway of each group of cells.Results The bioinformatics analysis showed that the TTK gene was the core gene associated with the occurrence and development of eyelid BCC.CIA detection results revealed that the fluorescence signal intensities in the tumor cytoplasm of Control,BCC Grade Ⅰ,Ⅱ,and Ⅲ groups were 1.03±0.07,1.28±0.11,1.58±0.13 and 1.92±0.17,respectively.The fluorescence signal intensity gradually increased,and the difference in fluorescence signal intensity among the four groups was statistically significant(all P＜0.05).Compared with that in the Control group(1.02±0.05),the cell fluorescence intensity was increased in the BCC negative control group(1.74±0.12)and decreased in the TTK-shRNA group(1.31±0.09)(P＜0.05).The difference in cell fluorescence intensity was significant among the Control,BCC nega-tive control and TTK-shRNA groups(all P＜0.05).The fluorescence intensity of the anti-apoptotic protein Bcl-2 was 1.04±0.12 in the Control group,2.12±0.23 in the BCC negative control group,and 1.43±0.15 in the TTK-shRNA group.The fluorescence intensity of the pro-apoptotic protein Bax was 1.02±0.08 in the Control group,0.64±0.11 in the BCC negative control group,and 1.47±0.16 in the TTK-shRNA group.After TTK knockdown,the expression level of BcL-2 in BCC cells decreased,and that of Bax increased.The fluorescence intensities of BcL-2 and Bax were significantly different among the Control,BCC negative control and TTK-shRNA groups(all P＜0.05).Conclusion The TTK gene plays a role in the regulation of eyelid BCC cell proliferation,and this effect is closely related to the PI3K-AKT-Bcl-2/Bax signaling path-way.

Objective:To evaluate the early outcomes of total hip arthroplasty (THA) and discuss the revisions post THA in the treatment for Kashin-Beck disease (KBD) with severe hip problems.Methods:This retrospective cohort study enrolled 50 patients (64 hips) with a mean age of 52.4±8.7 years, including 25 male patients and 25 female patients (36 left hips and 28 right hips), who were diagnosed as KBD with hip problems and received THA at Arthritis Clinical and Research Centre, Peking University People's Hospital from October 2019 to January 2024. The leg length discrepancy (LLD), femoral offset (FO), abduction angle and anteversion angle were calculated preoperatively and one week post-operation. The postoperative radiological indexes and the functional outcomes in the last follow-up were compared with the preoperative assessment.Results:The surgical duration was 105(80, 120) min and the bleeding amount was 300(200, 400) ml. All the cases were followed up for an average of 37 months (ranging from 21 to 44 months). Significant differences were found on postoperative radiological images, with LLD improving to 0.50±0.78 cm from a preoperative value of -1.36±0.79 cm, and FO increasing to 3.28±1.01 cm from 2.72±0.83 cm ( P<0.05). The mean postoperative abduction angle and anteversion angle were 42.5°±7.7° and 15.1°±5.9°, respectively. A total of 71.8% and 95.3% hips fell within the Lewinnek safe zones of abduction angle and anteversion angle, respectively. In terms of functional outcomes, the average range of motion improved significantly to 185°(173°, 210°) from a preoperative value of 99°(76°, 123°), and the Harris Hip Score increased from 35(26, 43) preoperatively to 70(63, 80) postoperatively ( P<0.05). During the follow-up, there were complications for two cases of femoral stem loosening, one case of periprosthetic femoral fracture, one case of hip dislocation, and one case of acetabular component loosening with hip subluxation. Additionally, seven patients exhibited Trendelenburg gait. A total of five hips required revision surgery due to severe complications, including two cases of femoral stem loosening, one case of periprosthetic femoral fracture, one case of hip dislocation, and one case of acetabular component loosening with subluxation. Conclusions:Patients with KBD demonstrated significant early improvements in both radiological and functional outcomes following THA.

Objective:To explore the independent association between handgrip strength and chronic kidney disease (CKD) in adult residents in Anhui Province using data from the China Adult Chronic Disease and Risk Factor Surveillance (2023).Methods:A multi-stage stratified cluster random sampling method was used to select residents aged ≥18 years for surveys, physical measurements, and laboratory tests. Relevant covariates were adjusted, and a multivariable logistic regression model was established to infer the association between handgrip strength and CKD, followed by subgroup analysis.Results:A total of 7 295 participants were included in the study, with age of (61.5±13.2) years, and 55.6% of the study participants were women. The results of the multivariate logistic regression analysis showed that with each 1.0 kg increase in handgrip strength, the risk for CKD decreased by 1.3% ( OR=0.987, 95% CI: 0.978-0.997). Compared with those with low handgrip strength, the people with moderate hasdgrip strength ( OR=0.818, 95% CI: 0.694-0.964) and high handgrip strength ( OR=0.729, 95% CI: 0.598-0.989) had lower risk for CKD. In the subgroup analysis, the association between handgrip strength and risk for CKD remained unchanged regardless age, sex, smoking status, and alcohol consumption statuys, and the prevalence of hypertension and hyperlipidemia (interaction P>0.05), except BMI and diabetes. Conclusion:The decline in handgrip strength is associated with an increased risk for CKD in adult residents in Anhui.