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.

Aim To investigate the regulatory mecha-nism of arrhythmia of sodium channel ubiquitination af-ter MI and to study the electrophysiological remodeling mechanism of lncRNA-CCRR after MI for the preven-tion and treatment of arrhythmia after MI.Methods LncRNA-CCRR transgenic mice and C57BL/6 mice injected with lncRNA-CCRR overexpressed adeno-asso-ciated virus were used.Four weeks after infection,the left anterior descending branch of the coronary artery was ligated for 12 h to establish a mouse acute myocar-dial infarction model,and the incidence of arrhythmia was detected by programmed electrical stimulation.Ln-cRNA-CCRR overexpression/knockdown adeno-associ-ated virus and negative control were transfected into neonatal mouse cardiomyocytes(NMCMs),and the model was prepared by hypoxia for 12 h.LncRNA-CCRR expression was detected by FISH,Nav1.5 and UBA6 protein and Nav.1.5 mRNA expression were de-tected by Western blot and real-time quantitative poly-merase chain reaction(qRT-PCR),Nav1.5 and UBA6 expressions were detected by immunofluores-cence,and the relationship between lncRNA-CCRR and UBA6 was detected by RIP.INa current density af-ter CCRR overexpression and knockdown was detected by Whole-cell clamp patch.Results In MI mice,the expression of lncRNA-CCRR decreased,the incidence of arrhythmia increased,the expression of CCRR and Nav1.5 mRNA was down-regulated,the protein ex-pression of Nav1.5 was down-regulated,and the pro-tein expression of UBA6 was up-regulated compared with sham group.Overexpression of CCRR could re-verse the above changes.AAV-CCRR could reverse the down-regulated CCRR and Nav1.5 mRNA levels af-ter hypoxia,and improve the expression of Nav1.5 and UBA6 protein.The direct relationship between ln-cRNA-CCRR and UBA6 was identified by RIP analy-sis.The INa density increased after transfection with AAV-CCRR.The INa density decreased after transfec-tion with AAV-si-CCRR.Conclusions The expres-sion of lncRNA-CCRR decreases after MI,and ln-cRNA-CCRR can improve arrhythmia induced by MI by inhibiting UBA6 to increase the protein expression level of Nav1.5 and the density of INa.

Aim To investigate the effect of total glu-cosides of paeony on inflammatory injury and TLR4/NF-κB/NLRP3 pathway in autoimmune thyroiditis(AIT)rats.Methods The experiment was divided into control group,model group,total glucosides of pae-ony(TGP),TLR4 inhibitor group and TGP+TLR4 ag-onist group,with 10 animals in each group.Except for the control group,the rats in other groups were subcu-taneously injected with thyroglobulin and Freund's ad-juvant to induce the AIT rat model.After six weeks of administration,thyroid histopathological changes were observed using hematoxylin-eosin(HE)staining;ser-um levels of TPOAb,TgAb,TSH,T3,T4,TNF-α,INF-γ,IL-1 β and IL-1 β were detected by enzyme-linked immunosorbent assay(ELISA);TLR4/NF-κB/NLRP3 pathway mRNAs and proteins expression in thyroid tis-sues were detected by RT-qPCR and Western blot.Re-sults Compared with the control group,the thyroid follicular epithelium of rats was significantly damaged,and the serum levels of TPOAb,TgAb,TSH,T3,T4,TNF-α,INF-γ,IL-1 β and IL-1 β increased(P＜0.01).The expression of TLR4/NF-κB/NLRP3 path-way mRNAs and proteins increased in the model group(P＜0.01).Compared with the model group,the damage of thyroid follicular epithelium was alleviated,and the serum levels of TPOAb,TgAb,TSH,T3,T4,TNF-α,INF-γ,IL-1 β and IL-1 β were reduced(P＜0.01),the expression of TLR4/NF-κB/NLRP3 path-way mRNAs and proteins were down-regulated in the TGP group and TLR4 inhibitor group(P＜0.01).Compared with TGP group,the damage of thyroid follic-ular epithelium was aggravated,and the levels of serum TPOAb,TgAb,TSH,T3,T4,TNF-α,INF-γ,IL-1 β and IL-1 β were elevated(P＜0.05 or P＜0.01),the pro-tein expressions of TLR4/NF-κB/NLRP3 pathway mR-NAs and proteins were up-regulated in TGP+TLR4 ag-onist group(P＜0.05 or P＜0.01).Conclusions TGP may play a protective role in thyroid by inhibiting the TLR4/NF-κB/NLRP3 pathway and improving the inflammatory injury of thyroid tissues.

Objective To verify the safety and efficacy of a new portable extracorporeal membrane oxygenation(ECMO)system(Xijing Advanced Life Support System JC-Ⅲ)in large animals.Methods A total of 10 healthy small fat-tail sheep underwent veno-arterial extracorporeal membrane oxygenation(VA-ECMO)support by carotid arterial-jugular catheterization to evaluate the performance of the JC-Ⅲ ECMO system.Systemic anticoagulation was achieved by continuous infusion of heparin.Active coagulation time(ACT)was recorded every 2 hours during the experiment,and the ACT was maintained between 200-250 s.Centrifugal pump speed is set at 3 000-3 500 r/min.The changes of hemoglobin,blood cell counts,hematocrit,liver and kidney function were monitored before and 24 h after ECMO initiation,respectively.After the experiment,the pump and oxygenator were dissected to probe the thrombosis.Results The success rate of VA-ECMO operation was 100％,and there was no hemolysis,pump thrombosis and oxygenator thrombosis after 24 h of ECMO.Before and after the operation,there were no significant changes in indicators such as hemoglobin content,white blood cell counts,platelet counts,alanine aminotransferase concentration,aspartate aminotransferase concentration,urea,creatinine,high-sensitivity troponin Ⅰ,and N-terminal pro-brain natriuretic peptide(all P＞0.05).Conclusions This in vivo study confirms that Xijing Advanced Life support System JC-Ⅲ is safe and effective.

The aim of this study was to investigate the impact of overexpressing 70-kD heat shock pro-teins(Hsp70)on glycolysis in C2C12 cells during myogenesis and adipogenesis.Using C2C12 cells as the research material,adenovirus was used to overexpress the Hsp70 gene,and changes in the expression of glycolytic genes were detected using fluorescence quantitative PCR and Western blotting techniques.The study indicated that during C2C12 cell myogenic differentiation,the expression trend of the Hsp70 gene was consistent with that of Gsk3β,Pkm,Prkag3,Pfkm,and Hk-2 genes,suggesting a relationship between Hsp70 and the glycolytic pathway during myogenic differentiation.Overexpression of Hsp70 in the later stages of myogenic differentiation significantly upregulated the expression of Gsk3β,Pkm,Prk-ag3,and Pfkm genes(P＜0.05),with no significant impact on Hk-2 gene expression(P＞0.05).Dur-ing C2C12 cell adipogenic induction,the expression trend of the Hsp70 gene was similar to that of Gsk3β,Pkm,Prkag3,Pfkm,and Hk-2 genes,indicating a relationship between Hsp70 and the glycolytic path-way during adipogenic induction.Following Hsp70 overexpression,in the later stages of adipogenic in-duction,the number of lipid droplets was significantly higher compared to the control group,with a sig-nificant upregulation of Gsk3β,Pkm,Prkag3,and Pfkm gene expression(P＜0.05),while Hk-2 gene expression was not significantly affected(P＞0.05).In conclusion,Hsp70 in C2C12 cells in myogenic and adipogenic states promoted the breakdown of glycogen into 6-phospho-glucose,thereby enhancing the glycolytic pathway,providing insights into the functional role of the Hsp70 gene in glycolysis in C2C12 cells.

Objective:To compare the short-term effect and adverse reaction of venetoclax(VEN)combined with azacitidine(AZA)versus"7+3"regimen in newly diagnosed elder patients with acute myeloid leukemia(AML).Methods:From January 2021 to January 2022,the clinical data of seventy-nine newly diagnosed elder patients with AML at the Second Hospital of Shanxi Medical University and the Shanxi Bethune Hospital were retrospectively analyzed,including VEN+AZA group(41 cases)and"7+3"group(38 cases).The propensity score matching(PSM)method was used to balance confounding factors,then response,overall survival(OS),progression-free survival(PFS)and adverse reactions between the two groups were compared.Results:The ORR of VEN+AZA group and"7+3"group was 68％and 84％,respectively,and the CRc was 64％and 72％,respectively,the differents were not statistically significant(P＞0.05).In the VEN+AZA group,there were 5 non-remission(NR)patients,4 with chromosome 7 abnormality(7q-/-7),and 1 with ETV6 gene mutation.Median followed-up time between the two groups was 8 months and 12 months,respectively,and the 6-months OS was 84％vs 92％(P=0.389),while 6-months PFS was 84％vs 92％(P=0.258).The main hematological adverse reactions in two groups were stage Ⅲ-Ⅳmyelosuppression,and the incidence rate was not statistically different(P＞0.05).The median time of neutrophil recovery in two groups was 27(11-70)d,25(14-61)d(P=0.161),and platelet recovery was 27(11-75)d,25(16-50)d(P=0.270),respectively.The infection rate of VEN+AZA group was lower than that of"7+3"group(56％vs 88％,P=0.012).The rate of lung infections of two groups was 36％and 64％,respectively,the difference was statistically significant(P=0.048).Conclusion:The short-term effect of VEN+AZA group and"7+3"regimens in eldrly AML patients are similar,but the VEN+AZA regimen had a lower incidence of infection.The presence of chromosome 7 abnormality(7q-/-7)may be a poor prognostic factor for elderly AML patients treated with VEN+AZA.