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.

ObjectiveTo evaluate the clinical efficacy of Huayu Tongluo Formula （化瘀通络方， HTF） in patients with mid-stage diabetic kidney disease of blood stasis syndrome and explore its potential mechanisms. MethodsA multi-center， randomized， double-blind， placebo-controlled clinical trial was conducted. Ninety patients of mid-stage diabetic kidney disease of blood stasis syndrome were divided into a control group of 46 cases and a treatment group of 44 cases. Both groups received conventional western medicine treatment， the treatment group additionally taking HTF， while the control group taking a placebo of the formula. The treatment was administered once daily for 24 weeks. The primary outcomes included 24-hour urine total protein （24 h-UTP）， serum albumin （Alb）， glycated hemoglobin （HbA1c）， and serum creatinine （Scr）.The secondary outcomes included changes in levels of endothelin-1 （ET-1）， nitric oxide （NO）， vascular endothelial growth factor （VEGF）， and traditional Chinese medicine （TCM） syndrome scores before and after treatment. Clinical efficacy was evaluated based on TCM syndrome scores and overall disease outcomes. Adverse reactions and endpoint events were recorded. ResultsIn the treatment group after treatment， 24 h-UTP， ET-1， and VEGF levels significantly decreased （P＜0.05）， Alb and NO levels significantly increased （P＜0.05）； while the TCM syndrome scores for edema， lumbar pain， numbness of limbs， dark purple lips， dark purple tongue or purpura， and thin， rough pulse all significantly decreased （P＜0.05）. In the control group， no significant changes were observed in any of the indicators after treatment （P＞0.05）.Compared with the control group， the treatment group showed significant reductions in 24 h-UTP， ET-1， and VEGF levels， and increases in Alb and NO levels （P＜0.05）. The TCM syndrome scores for edema， lumbar pain， dark purple tongue or purpura， and thin， rough pulse were all lower in the treatment group than in the control group （P＜0.05）. The total effective rate of TCM syndrome in the treatment group was 59.09% （26/44）， and the overall clinical effective rate was 45.45% （20/44）. In the control group， these rates were 15.22% （7/46） and 8.7% （4/46）， respectively， with the treatment group showing significantly better outcomes （P＜0.05）. A total of 7 adverse events occurred across both groups， with no significant difference （P＞0.05）. No endpoint events occurred during the study. ConclusionOn the basis of conventional treatment of Western medicine， HTF can further reduce urinary protein levels and improve clinical symptoms in patients with mid-stage diabetic kidney disease of blood stasis syndrome. The mechanism may be related to its effects on endothelial function.

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 assess the situation and influencing factors of occupational burnout among the staff at the Center for Disease Control and Prevention (CDC) in Sichuan Province. Methods A total of 1 038 CDC staff members in Sichuan Province were selected as the study subjects using the stratified random sampling method. Occupational burnout of the staff was assessed using the Maslach Burnout Inventory General Survey via an online questionnaire. Results The detection rate of occupational burnout was 42.3% (439/1 038). Binary logistic regression analysis result showed that, after controlling for confounding factors such as education level and alcohol consumption, CDC staffs aged at 20-<31, 31-<41, and 41-<51 years were at higher risk of occupational burnout compared with those ≥51 years (all P<0.05). CDC staffs with 5-<10 or ≥10 years of service had higher occupational burnout risk compared with those with <5 years (both P<0.05). CDC staffs with poor or fair health status, irregular diet, and poor sleep quality had higher risk of occupational burnout compared with those healthy, have regular diet, and good sleep quality (all P<0.05). The risk of occupational burnout increased with higher overtime frequency (all P<0.05). Conclusion Occupational burnout among CDC staffs in Sichuan Province is relatively high. Age, years of service, health status, diet, sleep quality, and overtime frequency are key influencing factors.

Objective:To determine the content of the released nickel ion through the 7 extraction media to extract the Ni-Ti wires and to plot the curve of the released nickel ion so as to identify a leaching medium that can be substituted for blood for in vitro Ni release evaluation. Methods:The release of Ni through microwave digestion/inductively coupled plasma mass spectrometry (ICP-MS) in the goat serum was determined. Because of the high content of Ni release, it could be determined by diluting the extraction medium, and other extraction media could be determined directly. Ni release standard curves were plotted by the release amount and different time point variables. Though the different extraction media Ni release curves confirm the specificity of extraction media instead of blood.Results:By analyzing the Ni release curves of seven leaching media, it was found that none of these seven extraction media was suitable for the evaluation of Ni release in in vitro leaching media. Considering the safety of the leaching medium and the simplicity of preparation, hydrochloric acid solution was chosen as the leaching medium, but the concentration needed to be diluted accordingly. Finally, a hydrochloric acid solution was created as an alternative to blood for the in vitro study of Ni release from Ni-Ti alloy cardiovascular products, with a volume fraction of 0.005%. Conclusions:The in vitro leaching medium that can replace blood was found to be hydrochloric acid for the time being, but its concentration was too high, resulting in too much Ni release as well, which deviated from the actual situation. Therefore, the hydrochloric acid solution was diluted step by step, and the Ni release curve was examined until it was close to the clinical release level, and the actual concentration was determined, thus laying a solid foundation for the subsequent evaluation of the safety and risk.

Aim To explore the effect of circ_0038467 on nerve cell damage induced by hypoxia-glucose dep-rivation(OGD)and its possible mechanism.Methods Rat cortical nerve cells were isolated and cultured,and then induced by OGD to establish a cell injury model.si-NC,si-circ_0038467,miR-NC,and miR-940 mimics were transfected into rat cortical nerve cells and treated with OGD for 6 h.si-circ_0038467 and an-ti-miR-NC or anti-miR-940 were co-transfected into rat cortical neurons,followed by OGD treatment for 6 h.qRT-PCR was used to detect the expression levels of circ_0038467 and miR-940.CCK-8 method and flow cytometry were used to examine cell proliferation and apoptosis.LDH method was used to detect cell dam-age.The dual luciferase reporter experiment was used to detect the targeting relationship between circ_0038467 and miR-940.Western blot was employed to detect cleaved caspase-3 and cleaved caspase-9 protein levels.Results Circ_0038467 expression increased and miR-940 expression decreased in OGD-induced nerve cells(P＜0.01).After transfection with si-circ_0038467 or miR-940 mimics,cell survival rate in-creased(P＜0.01),while LDH release rate,apopto-sis rate,and the protein levels of cleaved caspase-3 and cleaved caspase-9 decreased(P＜0.01).Circ_0038467 could target miR-940.Compared with the OGD+si-circ_0038467+anti-miR-NC group,cell survival rate in OGD+si-circ_0038467+anti-miR-940 group was down-regulated(P＜0.01),while LDH re-lease rate,apoptosis rate and cleaved caspase-3,cleaved caspase-9 levels were up-regulated(P＜0.01).Conclusion Interference of circ_003 8467 and could protect nerve cells from OGD-induced oxidative stress and apoptosis by up-regulating miR-940.

Aim To explore the improvement effect of Bazi Bushen capsule on thymic degeneration in SAMP6 mice and the possible mechanism.Methods Twenty 12 week old male SAMP6 mice were randomly divided into the model group(SAMP6)and the Bazi Busheng capsule treatment group(SAMP6+BZBS).Ten SAMR1 mice were assigned to a homologous control group(SAMR1).The SAMP6+BZBS group was oral-ly administered Bazi Bushen capsule suspension(2.8 g·kg-1)daily,while the other two groups were orally administered an equal amount of distilled water.After nine weeks of administration,the morphology of the thymus in each group was observed and the thymus in-dex was calculated;HE staining was used to observe the structural changes of thymus tissue;SA-β-gal stai-ning was used to detect thymic aging;flow cytometry was used to detect the proportion of thymic CD3+T cells in each group;Western blot was used to detect the levels of p16,Bax,Bcl-2,and cleaved caspase-3 proteins in thymus;immunofluorescence was applied to detect the proportion of cortical thymic epithelial cells in each group;ELISA was employed to detect IL-7 lev-els in thymus.Results Compared with the SAMP6 group,the thymic index of the SAMP6+BZBS group significantly increased(P＜0.05);the disordered thy-mic structure was significantly improved;the positive proportion of SA-β-gal staining significantly decreased(P＜0.01);the proportion of CD3+T cells apparently increased(P＜0.05);the level of p16 protein signifi-cantly decreased(P＜0.05);the level of Bcl-2 pro-tein significantly increased(P＜0.05),while the lev-el of cleaved caspase-3 protein markedly decreased(P＜0.05);the proportion of cortical thymic epithelial cells evidently increased;the level of IL-7 significantly increased(P＜0.01).Conclusions Bazi Bushen capsule can delay thymic degeneration,inhibit cell ap-optosis in thymus and promote thymic cell development in SAMP6 mice,which may be related to increasing the proportion of cortical thymic epithelial cells and promoting IL-7 secretion.