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.

italic>Salmonella has emerged as a promising tumor-targeting strategy in recent years due to its good tumor targeting ability and certain safety. In order to further optimize its therapeutic effect, scientists have tried to modify Salmonella, including its attenuation and drug loading. This paper summarizes the mechanism and research progress of Salmonella-mediated targeted tumor therapy, and introduces the strategies and related progress of its modification and optimization. At the same time, the advantages, current challenges and future development directions of Salmonella-mediated tumor therapy are summarized.

OBJECTIVE: To assess the risk of aristolochic acid (AA)-associated cancer in patients with AA nephropathy (AAN). METHODS: A retrospective study was conducted on patients diagnosed with AAN at Peking University First Hospital from January 1997 to December 2014. Long-term surveillance and follow-up data were analyzed to investigate the influence of different factors on the prevalence of cancer. The primary endpoint was the incidence of liver cancer, and the secondary endpoint was the incidence of urinary cancer during 1 year after taking AA-containing medication to 2014. RESULTS: A total of 337 patients diagnosed with AAN were included in this study. From the initiation of taking AA to the termination of follow-up, 39 patients were diagnosed with cancer. No cases of liver cancer were observed throughout the entire follow-up period, with urinary cancer being the predominant type (34/39, 87.17%). Logistic regression analysis showed that age, follow-up period, and diabetes were potential risk factors, however, the dosage of the drug was not significantly associated with urinary cancer. CONCLUSIONS No cases of liver cancer were observed at the end of follow-up. However, a high prevalence of urinary cancer was observed in AAN patients. Establishing a direct causality between AA and HCC is challenging.
Humans ; Retrospective Studies ; Incidence ; Carcinoma, Hepatocellular ; Liver Neoplasms/epidemiology* ; Kidney Diseases/chemically induced* ; Aristolochic Acids/adverse effects*

Objective To explore the effect of scutellarin on lipopolysaccharide(LPS)induced neuroinflammation in BV-2 microglia cells.Methods BV-2 microglia were cultured and randomly divided into 6 groups:control group(Ctrl),cyclic GMP-AMP synthetase(cGAS)inhibitor RU320521 group(RU.521 group),LPS group,LPS+RU.521 group,LPS+scutellarin pretreatment group(LPS+S)and LPS+S+RU.521 group.The expressions of cGAS,stimulator of interferon gene(STING),nuclear factor kappa B(NF-κB),phosphorylated NF-κB(p-NF-κB),neuroinflammatory factors PYD domains-containing protein 3(NLRP3)and tumor necrosis factor α(TNF-α)in BV-2 microglia were detected by Western blotting and immunofluorescent double staining(n= 3).Results Western blotting and immunofluorescent double staining showed that compared with the control group,the expression of cGAS,STING,p-NF-κB,NLRP3 and TNF-α in BV-2 microglia increased significantly after LPS induction(P<0.05),while the expression of cGAS,STING,p-NF-κB,NLRP3 and TNF-α in LPS+S group were significantly lower than those in LPS group(P<0.05).Treatment with cGAS pathway inhibitor RU.521 showed similar effects as the pre-treatment group with scutellarin.In addition,the change of NF-κB in each group was not statistically significant(P>0.05).Conclusion Scutellarin inhibits the neuroinflammation mediated by BV-2 microglia cells,which may be related to cGAS-STING signaling pathway.

We aimed to identify the infectious source of a case of melioidosis,to provide evidence for the prevention and control of melioidosis in Shanxi Province,China.The patient developed repeated fever,fatigue,diarrhea,and other symptoms after being caught in the rain while traveling in Hainan Province.The blood culture was positive,and the bacterial strain was i-dentified as Burkholderia thayensis and sent to the provincial Center for Disease Control and Prevention for further evaluation.MALDI-TOF MS and biochemical identification were used to identify the strain,whole genome sequencing was performed after nucleic acid extraction,MLST type and drug-resistance genes were analyzed,and a phylogenetic tree was constructed.The iso-lated strain was identified as Burkholderia pseudomallei by MALDI-TOF MS and biochemistry,and the MLST type was 366.The whole gene sequencing analysis indicated a close evolutionary relationship with the three isolates in Hainan Province,with high homology.This case of melioidosis was indeed imported from Hainan Province,according to molecular tracing analysis and epidemiological investigation,thus suggesting that medical institutions and disease control departments should strengthen understanding of melioidosis,and improve the diagnosis and treatment ability.

Objective:To investigate the status quo of training and domestic use of 707 clinical nutrition support specialty nurses from 21 provinces,cities,and autonomous regions in China. And to analyze their influencing factors and provide reference for improving the training system of clinical nutrition support specialty nurses,selection and development of specialist nurses in clinical nutrition support. Methods:From October to November 2023,a cross-sectional survey was conducted on 707 clinical nutrition support specialty nurses from 21 provinces,cities,and autonomous regions across China was conducted using a convenience sampling method based on a questionnaire about the training and home use of clinical nutrition support nurses. Univariate and multiple linear regression analysis was used to examine the use status and application of clinical nutrition support specialty nurses in five aspects:clinical nursing practice,nursing education,nursing management,coordination,nursing research and consultation. Results:The use of specialist clinical nutrition support nurses is not ideal,with 75.67％ of specialist nurses scoring less than 208 points (i.e. less than 80％ of the total score). Among the use of different dimensions,the clinical nursing practice dimension received the highest score (54.17±10.26),followed by the nursing education dimension (36.98±8.00). The results of multiple linear regression analysis show that hospital level and professional title are independent influencing factors influencing the use and development of specialist nurses. Conclusion:There is a need to further improve the utilisation of clinical nutrition support nurses. It is recommended that links and cooperation between hospitals at all levels,communities,and families be strengthened. For specialist nurses with higher professional titles,encourage them to fully play their roles,strengthen training in weak areas,continuously optimize the professional ability of clinical nutrition support nursing teams,comprehensively improve the quality of clinical nutrition support specialist nursing,and promote their high-level development.