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.

Objective:To assess the association between body mass index (BMI) and major adverse cardiovascular and cerebrovascular events (MACCE) among patients with acute coronary syndrome (ACS).Methods:This was a multicenter prospective cohort study, which was based on the Improving Care for Cardiovascular Disease in China (CCC) project. The hospitalized patients with ACS aged between 18 and 80 years, registered in CCC project from November 1, 2014 to December 31, 2019 were included. The included patients were categorized into four groups based on their BMI at the time of admission: underweight (BMI<18.5 kg/m 2), normal weight (BMI between 18.5 and 24.9 kg/m 2), overweight (BMI between 25.0 and 29.9 kg/m 2), and obese (BMI≥30.0 kg/m 2). Multivariate logistic regression models was used to analyze the relationship between BMI and the risk of in-hospital MACCE. Results:A total of 71 681 ACS inpatients were included in the study. The age was (63.4±14.7) years, and 26.5% (18 979/71 681) were female. And the incidence of MACCE for the underweight, normal weight, overweight, and obese groups were 14.9% (322/2 154), 9.5% (3 997/41 960), 7.9% (1 908/24 140) and 7.0% (240/3 427), respectively ( P<0.001). Multivariate logistic regression analysis showed a higher incidence of MACCE in the underweight group compared to the normal weight group ( OR=1.30, 95% CI 1.13-1.49, P<0.001), while the overweight and obese groups exhibited no statistically significant difference in the incidence of MACCE compared to the normal weight group (both P>0.05). Conclusion:ACS patients with BMI below normal have a higher risk of in-hospital MACCE, suggesting that BMI may be an indicator for evaluating short-term prognosis in ACS patients.

Humans ; Obesity/epidemiology*

Objective To explore the difference in efficacy of metoprolol versus ivabradine in the treatment of postural orthostatic tachycardia syndrome(POTS)in the elderly after COVID-19 infection.Methods A total of 110 patients diagnosed with POTS at our department from Decem-ber 1,2022 to January 31,2023 were included.According to their drug regimen,they were divided into metoprolol group(62 patients)and ivabradine group(48 patients).On the 28th day of out-patient follow-up,the resting heart rate,heart rate of 10 min of standing,symptom disappearance rate,hospitalization rate,and mortality rate were compared between the two groups.Results On the 28th day of treatment,the resting heart rate and postural heart rate for 10 min were decreased in both groups when compared with the levels at initial diagnosis(P＜0.01).And there were no significant differences in the two types of heart rate between the two groups on the 28th day(71.0±7.0 vs 72.1±7.0,P=0.401;76.5±7.2 vs 77.4±7.6,P=0.573).No obvious differences were observed between the two groups in symptom disappearance rate,hospitalization rate,or mortality rate(88.7％vs 89.6％,3.2％vs2.1％,0％vs 0％,P＞0.05).Conclusion Metoprolol and ivabradine can effectively treat POTS in the elderly patients after COVID-19 infection.

Objective To investigate the clinical significance of combined cerebrospinal fluid(CSF)and routine blood parameter analysis in differentiating between multiple cerebral glioma(MCG)and primary central nervous system lymphoma(PCNSL).Methods We Rretrospectively analyzed the clinical data,CSF and routine blood indicators levels of 62 MCG patients and 56 PCNSL patients admitted to Beijing Tiantan Hospital,Capital Medical University from November 2017 to March 2023.Additionally,we assessed the diagnostic value of individual meaningful indicators as well as their combinations in distinguishing between MCG and PCNSL.Results The levels of CSF total cell count,CSF white cell count,CSF:pro,lactate,routine bloodperipheral neutrophil count,and neu-trophil percentage were significantly higher in the MCG group than in the PCNSL group(P<0.05);while the levels of CSF:Glu,CSF:cl,routine blood lymphocyte count,eosinophil,lymphocyte percentage,and eosinophil percent-age were significantly higher in the PCNSL group than in the MCG group(P<0.05).The AUCs of CSF cell count,CSF white cell count,CSF:pro,lactate,routine blood neutrophil count,neutrophil percentage for differentiating MCG from PCNSL were 0.900,0.899,0.797,0.867,0.828 and 0.772 respectively;sensitivities were 72.4%,77.6%,63.8%,67.2%,72.4%,82.8%,77.6%and 81%,with sensitivities of 97.1%,100%,88.2%,91.2%,88.2%,64.7%,100%and 94.1%,respectively.In addition,the combined detection of CSF total cell count,CSF white cell count,CSF:pro,routine blood neutrophil count and neutrophil percentage in CSF had an AUC of 0.919 for differentiating MCG from PCNSL,with a sensitivity and specificity of 77.6%and 100%,respectively.Conclusions Combined detection of CSF indicators including CSF total cell count,CSF white cell count,CSF:pro,along with routine blood markers such as neutrophil count and neutrophil percentage,holds significant clinical utility for differ-entiating between MCG and PCNSL.

Humans ; Obesity/epidemiology*