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 study the active ingredients and mechanism of action of Xinkeshu tablets against myo-cardial ischemia by network pharmacology and ze-brafish model.Methods The anti-myocardial ische-mia activity of Xinkeshu tablets was evaluated by iso-prenaline hydrochloride(ISO)-induced zebrafish myo-cardial ischemia model and H2O2-induced H9c2 dam-age model.The active ingredients of Xinkeshu tablets were retrieved using databases such as TCMSP.The potential targets were predicted by PharmaMapper data-base.Myocardial ischemic disease targets were searched by OMIM database.The potential therapeutic targets of Xinkeshu tablets against myocardial ischemia were analyzed.GO and KEGG enrichment analysis were conducted on core targets.The active ingredients were verified by zebrafish and cell model.qRT-PCR was used to detect the expression of key targets.Re-sults Xinkeshu tablets could significantly alleviate ISO-induced pericardial edema and bradycardia.It al-so could increase sinus venous-bulb aortic(SV-BA)distance and improve the cell viability.The 30 poten-tial active ingredients of Xinkeshu tables mainly acted on 30 core targets,including ALB,AKT1 and MAPK1,to regulate 627 GO items,including protein phosphorylation,negative regulation of apoptosis and positive regulation of PI3K signal transduction.KEGG results showed that 117 signaling pathways,including PI3K/Akt,FOXO and Ras,exerted anti-myocardial ischemia effect.Salvianolic acid A,lithospermic acid,rosmarinic acid,salvianolic acid D,salvianolic acid B,ginsenoside Rg2,hyperoside,3'-methoxypuerarin,3'-hydroxypuerarin and ginsenoside Rg1 could alleviate ISO-induced zebrafish myocardial ischemia and im-prove the cell viability.Xinkeshu tablets could upregu-late the expression of genes such as ras and akt1,and downregulate the expression of genes such as mapk1 and mapk8.Conclusion The active ingredients,in-cluding salvianolic acid A in Xinkeshu tablets,exert anti-myocardial ischemia effects by targeting targets,such as AKT1,MAPK1,and regulating signaling path-ways,such as PI3K/Akt,MAPK and Ras.

Methamphetamine abuse is a major public health problem in the world,and in recent years,methamphetamine is also the most abused synthetic drug in China.The neurotoxic or addiction mechanism of methamphetamine has not been fully clarified,and there is still a lack of specific withdrawal methods and drugs for methamphetamine abuse.Mitochondria are not on-ly the organelles to which methamphetamine directly produces toxic effects,but also participate in regulating the neurotoxic damage process of methamphetamine.Mitochondrial quality is the regulatory basis for maintaining mitochondrial homeostasis and is regulated by three main mechanisms,which are mitochon-drial biogenesis,mitochondrial dynamic,and mitophagy.This review summarizes the research progress of mitochondrial quality control in methamphetamine-induced neurotoxicity,which may provide theoretical support for further research on the mechanism of methamphetamine neurotoxicity and development the mito-chondria-targeting drugs.

Aim To study the effect of Guben Yanling pills on liver aging in aging mice and the related mech-anism.Methods The mice were randomly divided in-to blank control group,model group,vitamin E group(0.1 g·kg-1)and low,medium and high dose groups(0.59,1.17,2.34 g·kg-1)of Guben Yan-ling pills.The aging mouse model was established by subcutaneous injection of D-galactose(150 mg·kg-1)into the back of neck.At the same time of mod-eling,the corresponding drugs were given by gavage once a day for six weeks.The main organ indexes were calculated.HE staining was used to observe the mor-phology of liver tissue.Colorimetry was used to detect the activity of β-galactosidase in liver.ELISA was used to detect the content of TNF-α,IL-1 β,IL-6,IL-4,IL-10.Western blot was used to detect the protein relative expression level of IKKβ,Iκ Bα,NF-κB p65.Immunofluorescence was used to detect the expression level of NF-κB p65.Results Compared with the blank control group,the organ index of the brain,liv-er,kidney,spleen,and thymus in the model group decreased(P＜0.05,P＜0.01),the activity of β-galactosidase increased(P＜0.01),liver tissue mor-phology and structure were significantly damaged,the content of TNF-α,IL-1 β and IL-6 increased(P＜0.01),the content of IL-4 and IL-10 decreased(P＜0.01),the levels of IKKβ,NF-κB p65 in-creased(P＜0.01),the levels of IKBα decreased(P＜0.01),and the levels of NF-κB p65 in nucleus increased(P＜0.01).Compared with the model group,the organ indexes of brain,liver,kidney,spleen,and thymus in each dose group of Guben Yan-ling pills increased(P＜0.05,P＜0.01),the activity of β-galactosidase decreased(P＜0.01),the morpho-logical and structural damage of liver tissue was signifi-cantly improved,the content of TNF-α,IL-1 β and IL-6 decreased(P＜0.01),the content of IL-4 and IL-10 increased(P＜0.01),the levels of IKKβ,NF-κB p65 decreased(P＜0.01),the levels of IκBα in-creased(P＜0.01),and the levels of NF-κB p65 in nucleus decreased(P＜0.01).Conclusions Guben Yanling pills can antagonize liver aging in mice,and its mechanism may be related to inhibiting the activa-tion of NF-κB signaling pathway in liver,downregulat-ing downstream pro-inflammatory factor levels,upregu-lating anti-inflammatory factor levels,and alleviating inflammation in liver.

Aim To investigate the expression level of CBX4 in esophageal squamous cell carcinoma(ESCC)and the effect of CBX4 on ESCC proliferation and un-derlying molecular mechanisms.Methods The ex-pression of CBX4 in different cancers was analyzed in Pan-cancers.The expression level of CBX4 in ESCC was analyzed by t-test based on Gene Expression Omni-bus(GEO)data.The viability of CBX4-overex-pressed/knockdown ESCC cells was detected by MTT assay,colony formation assay and flow cytometry assay.Furthermore,the tumor volumn,tumor weight and Ki67 expression were measured by mouse xenograft assay and immunohistochemistry.The mRNA and protein ex-pression levels of apoptosis-related genes PARP、Bcl-2、Bax were determined by qRT-PCR and Western blot,respectively.In addition,the underlying molecular mechanism of CBX4 in ESCC was revealed by qRT-PCR and Western blot.Results CBX4 was upregulat-ed in various cancers.The expression level of CBX4 in ESCC was higher than that in normal tissues(P＜0.05)based on Gene Expression Omnibus(GEO)da-ta.Compared with the normal group,the proliferation of CBX4 knockdown ESCC cells was significantly in-hibited and the apoptosis was promoted(P＜0.05).Meanwhile,the mRNA and protein expression levels of cleaved PARP and Bax were upregulated while that of Bcl-2 was downregulated.In CBX4 overexpression group,tumor volume in vivo increased(P＜0.05).Immunohistochemical results also showed an increase in Ki67 expression.Furthermore,the results of RNA-seq,bioinformatics analysis and qRT-PCR experiments indicated that CBX4 probably regulated the prolifera-tion and apoptosis of ESCC through p38 MAPK signa-ling pathway.Conclusion CBX4 is highly expressed in ESCC and plays as an oncogene role,which might regulate cell proliferation through the p38 MAPK signa-ling pathway.

Objective To explore clinical efficacy of autologous bone grafts and bone substitute for the treatment of tibial plateau fractures by Meta analysis.Methods Controlled clinical studies on autogenous bone transplantation and bone substitutes in treating tibial plateau fractures published on PubMed,Web of Science,CNKI,Wanfang and other databases from January 2005 to August 2022 were searched by computer.Literature screening and data extraction were performed according to random-ized controlled trial(RCT),and the quality of RCT were evaluated by using intervention meta-analysis criteria in Cochrane man-ual.Meta-analysis of joint depression,secondary collapse rate of articular surface,blood loss,operative time and infection rate between two methods were performed by Rev Man 5.3 software.Results Seven RCT studies(424 patients)were included,296 patients in bone replacement group and 128 patients in autograft group.Operative time[MD=-16.79,95％CI(-25.72,-7.85),P=0.000 2]and blood loss[MD=-70.49,95％CI(-79.34,-61.65),P＜0.000 01]between two groups had statistically differ-ences,while joint depression[MD=-0.17,95％CI(-0.91,0.58),P=0.66],secondary collapse rate of joint surface[RR=-0.74,95％CI(0.35,1.57),P=0.43],infection rate[RR=1.21,95％CI(0.31,4.70),P=0.78]between two groups had no differences.Conclusion The effects of bone substitute and autograft for the treatment of tibial plateau fracture have similar effects in terms of joint depression,secondary articular surface collapse rate and infection rate.However,compared with autologous bone trans-plantation,bone replacement could reduce blood loss and shorten operation time.