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 establish a prediction model for high-risk cardiovascular disease (CVD) among residents aged 35 to 75 years, so as to provide the basis for improving CVD prevention and control measures. Methods: Permanent residents aged 35 to 75 years were selected from Dongcheng District, Beijing Municipality using the stratified random sampling method from 2018 to 2023. Demographic information, lifestyle, waist circumference and blood biochemical indicators were collected through questionnaire surveys, physical examinations and laboratory tests. Influencing factors for high-risk CVD among residents aged 35 to 75 years were identified using a multivariable logistic regression model, and a prediction model for high-risk CVD was established. The predictive effect was evaluated using the receiver operating characteristic (ROC) curve. Results: A total of 6 968 individuals were surveyed, including 2 821 males (40.49%) and 4 147 females (59.51%), and had a mean age of (59.92±9.33) years. There were 1 155 high-risk CVD population, with a detection rate of 16.58%. Multivariable logistic regression analysis showed that gender, age, smoking, central obesity, systolic blood pressure, fasting blood glucose, triglyceride and low-density lipoprotein cholesterol were influencing factors for high-risk CVD among residents aged 35 to 75 years (all P<0.05). The area under the ROC curve of the established prediction model was 0.849 (95%CI: 0.834-0.863), with a sensitivity of 0.693 and a specificity of 0.863, indicating good discrimination. Conclusion The model constructed by eight factors including demographic characteristics, lifestyle and blood biochemical indicators has good predictive value for high-risk CVD among residents aged 35 to 75 years.

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.

Aim To preliminarily evaluate the effects of hypobaric hypoxia on organism damage in rats with post-traumatic stress disorder(PTSD),with a view to laying a foundation for drug research in plateau PTSD.Methods The rats were randomly divided into four groups,namely,the control(Control)group,the sin-gle-prolonged stress(SPS)group,the hypobaric hy-poxia(HH)group and the single-prolonged stress combined with hypobaric hypoxia(SPS+HH)group.The PTSD model was firstly constructed using the SPS method for rats in the SPS and SPS+HH groups.On the second day,rats in the HH group and SPS+HH group were placed in a low-pressure hypoxia chamber at a simulated altitude of 6000 m for 14 days.General condition,behavior,blood tests,and histomorphology were examined in order to evaluate the damage caused by low pressure hypoxia in PTSD rats.Results The body mass of rats in the SPS+HH group was signifi-cantly reduced;the feces were partly hard and lumpy,and some of them were seen to have high viscosity.Anxiety-like and depression-like behaviors were ob-served in all groups except in the control group,in which hypobaric hypoxia aggravated the behavioral ab-normalities in SPS rats.Rats in both the SPS and SPS+HH groups had coagulation dysfunction and abnor-mally increased blood viscosity,which was significantly abnormal in the SPS+HH group;erythrocytes,hemo-globin,and erythrocyte specific volume in whole blood of rats in the SPS+HH group were significantly in-creased compared with those of rats in the SPS group;and serum TP,LDH and GLU levels were abnormal in rats in the SPS+HH group.Dilated and congested blood vessels were seen in hippocampal tissue,conges-ted central veins were seen in hepatic tissue,and dilat-ed and congested liver sinusoids with mild granuloma-tous degeneration of hepatocytes were seen in rats of the SPS+HH group.Conclusion Hypobaric hypoxia exacerbates depression-like and anxiety-like behaviors in PTSD rats,as well as hematological indices and his-tomorphometric abnormalities in PTSD rats.

Objective To summarize the reasons and management strategies of reoperation after oblique lateral interbody fusion(OLIF),and put forward preventive measures.Methods From October 2015 to December 2019,23 patients who under-went reoperation after OLIF in four spine surgery centers were retrospectively analyzed.There were 9 males and 14 females with an average age of(61.89±8.80)years old ranging from 44 to 81 years old.The index diagnosis was degenerative lumbar intervertebral dics diseases in 3 cases,discogenie low back pain in 1 case,degenerative lumbar spondylolisthesis in 6 cases,lumbar spinal stenosis in 9 cases and degenerative lumbar spinal kyphoscoliosis in 4 cases.Sixteen patients were primarily treated with Stand-alone OLIF procedures and 7 cases were primarily treated with OLIF combined with posterior pedicle screw fixation.There were 17 cases of single fusion segment,2 of 2 fusion segments,4 of 3 fusion segments.All the cases underwent reoperation within 3 months after the initial surgery.The strategies of reoperation included supplementary posterior pedicle screw instrumentation in 16 cases;posterior laminectomy,cage adjustment and neurolysis in 2 cases,arthroplasty and neuroly-sis under endoscope in 1 case,posterior laminectomy and neurolysis in 1 case,pedicle screw adjustment in 1 case,exploration and decompression under percutaneous endoscopic in 1 case,interbody fusion cage and pedicle screw revision in 1 case.Visu-al analogue scale(VAS)and Oswestry disability index(ODI)index were used to evaluate and compare the recovery of low back pain and lumbar function before reoperation and at the last follow-up.During the follow-up process,the phenomenon of fusion cage settlement or re-displacement,as well as the condition of intervertebral fusion,were observed.The changes in in-tervertebral space height before the first operation,after the first operation,before the second operation,3 to 5 days after the second operation,6 months after the second operation,and at the latest follow-up were measured and compared.Results There was no skin necrosis and infection.All patients were followed up from 12 to 48 months with an average of(28.1±7.3)months.Nerve root injury symptoms were relieved within 3 to 6 months.No cage transverse shifting and no dislodgement,loosening or breakage of the instrumentation was observed in any patient during the follow-up period.Though the intervertebral disc height was obviously increased at the first postoperative,there was a rapid loss in the early stage,and still partially lost after reopera-tion.The VAS for back pain recovered from(6.20±1.69)points preoperatively to(1.60±0.71)points postoperatively(P＜0.05).The ODI recovered from(40.60±7.01)％preoperatively to(9.14±2.66)％postoperatively(P＜0.05).Conclusion There is a risk of reoperation due to failure after OLIF surgery.The reasons for reoperation include preoperative bone loss or osteoporosis the initial surgery was performed by Stand-alone,intraoperative endplate injury,significant subsidence of the fusion cage after surgery,postoperative fusion cage displacement,nerve damage,etc.As long as it is discovered in a timely manner and handled properly,further surgery after OLIF surgery can achieve better clinical results,but prevention still needs to be strengthened.

Objective To explore the clinical efficacy and safety of one-stage posterior lesion removal and internal spinal fixation in patients with lumbar Brucellosis spondylitis.Methods The clinical data of 24 patients admitted from October 2017 to October 2022 were retrospectively analyzed,2 patients were lost to follow-up at 10 months after surgery,at the final 22 cases were included in the study,including 13 males and 9 females with an average age of(52.00±6.89)years old,were treated with one-stage posterior lesion removal and internal spinal fixation.The operation time,intraoperative bleeding,follow-up time,ery-throcyte sedimentation rate(ESR)and C-reactive protein(CRP)before and after operation were recorded.The pain visual ana-logue scale(VAS),Oswestry disability index(ODI),the Japanese Orthopaedic Association(JOA)score for neurofunction,American Spinal Injury Association(ASIA)spinal cord injury grade and modified MacNab criteria were ussed to evaluate the efficacy.Results All patients were followed up from 12 to 30 months with an average of(17.41±4.45)months.The operation time was 70 to 155 min with an average of(1 16.59±24.32)min;the intraoperative bleeding volume was 120 to 520 ml with an average of(275.00±97.53)ml.CRP and ESR levels decreased more significantly at 1 week and at the final follow-up than pre-operative levels(P＜0.05).VAS,JOA score and ODI at 1 week and at the latest follow-up were more significantly improved than preoperative results(P＜0.05).There was no significant difference between ASIA preoperative and 1 week after operation(P＞0.05),and a significant difference between preoperative and last follow-up(P＜0.05).In the final follow-up,21 patients had ex-cellent efficacy,1 patient had fair,and there was no recurrence during the follow-up.Conclusion One-stage transpedicular le-sion removal and internal spinal fixation,with few incisions and short operation time,helps the recovery of neurological func-tion,and the prognosis meets the clinical requirements,which can effectively control Brucella spondylitis.

Objective:Integrating genes and GEO database related cuproptosis chips,to analyze connection between cupropto-sis genes,immune infiltration and myocardial infarction(MI),construct risk prediction model,predict Western medicine and Chi-nese medicine,analyze miRNA-mRNA regulatory network,and to provides a new research direction for the future study of MI-related cuproptosis mechanism and immune infiltration.Methods:By GEO database retrieval of MI related chips,standardized processing and MI-related cuproptosis genes screening were performed,and immunoinfiltration analysis and quantification were performed based on the treated gene expression matrix,correlation between immune infiltrating cells and function was analyzed,as well as their differ-ences in MI group and the control group.Cuproptosis genes that most related to MI in immune infiltration were screened out,and the risk model was constructed to analyze the risk probability of cuproptosis genes in MI.The Enrichr website and Coremine Medical data-base were used to predict cuproptosis-related genes in MI in Western medicine and traditional Chinese medicine.Finally,the up-stream mirnas of FDX1 and SLC31A1 were predicted by miRTarBase,Starbase and Targetscan databases,and miRNA-mRNA regula-tory networks were constructed.Results:Correlation of immune infiltration showed that Tfh cells and B cells had the strongest positive correlation(r=0.68),while regulatory T cells and iDC had the strongest negative correlation(r=-0.63);the difference analysis of im-mune infiltration showed that the differences among mast cells,NK cells and Th1 cells in the MI group at the cellular level were the most significant(P<0.001);and the differences in APC co-inhibition and MHCⅠ at the functional level were the most significant(P<0.001).Six genes with the highest correlation between immune cells and immune function were screened out:ATP7A,DLD,FDX1,LIAS,LIPT1 and SLC31A1.Results of the risk model showed that the high levels of FDX1 and SLC31A1 were negatively correlated with the risk prediction of MI.A total of 21 Western medicines and 30 traditional medicines were predicted by database comparison.miRTarBase,Starbase and Targetscan databases predicted 9 upstream miRNAs of cuproptosis-related genes in MI,including has-miR-122-5p.Conclusion:Tfh cells,B cells,para-inflammatory,typeⅠinterferon response and other related immune cells and functions may play important roles in pathogenesis and prognosis of MI.FDX1 and SLC31A1 as the key genes of cuproptosis process,are negatively correlated with MI.A total of 30 kinds of traditional Chinese medicines including Spirulina,sheep liver,Wen ezhu,Pian jianghuang and Yujin may have potential value in treatment of MI.Finally,9 miRNAs including has-miR-122-5p may play an important role in the regulation of cuproptosis in myocardial infarction.