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.

With the completion of the "Human Genome Project" and the smooth progress of the "Herbal Genome Project", the research wave of RNAomics is gradually advancing, opening the research gateway for the modernization of traditional Chinese medicine (TCM) and initiating the post-genome era of medicinal plant RNA research. Therefore, this article proposes for the first time the concept of HerbRNomes, which involves constructing databases of medicinal plant, medicinal fungus, and medicinal animal RNA at different stages, from different origins, and in different organs. This research aims to explore the role of HerbRNA in self-genetic information transmission, functional regulation, as well as cross-species regulation functional mechanisms and key technologies. It also investigates application scenarios, providing a theoretical basis and research ideas for the resistance of TCM or medicinal plants to adversity and stress, molecular assistant breeding, and the development of small nucleic acid drugs. This article reviews recent research progress in elucidating the molecular mechanisms of the transmission and expression of genetic information, self-regulation and cross-species regulation of herbs at the RNA level, along with key technologies. It proposes a development strategy for small nucleic acid drugs based on HerbRNomes, providing theoretical support and guidance for the modernization of TCM based on HerbRNomes research.

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.

Background::Early detection of esophageal squamous cell carcinoma (ESCC) can considerably improve the prognosis of patients. Aberrant cell-free DNA (cfDNA) methylation signatures are a promising tool for detecting ESCC. However, available markers based on cell-free DNA methylation are still inadequate. This study aimed to identify ESCC-specific cfDNA methylation markers and evaluate the diagnostic performance in the early detection of ESCC.Methods::We performed whole-genome bisulfite sequencing (WGBS) for 24 ESCC tissues and their normal adjacent tissues. Based on the WGBS data, we identified 21,469,837 eligible CpG sites (CpGs). By integrating several methylation datasets, we identified several promising ESCC-specific cell-free DNA methylation markers. Finally, we developed a dual-marker panel based on methylated KCNA3 and OTOP2, and then, we evaluated its performance in our training and validation cohorts. Results::The ESCC diagnostic model constructed based on KCNA3 and OTOP2 had an AUC of 0.91 [95% CI: 0.85–0.95], and an optimal sensitivity and specificity of 84.91% and 94.32%, respectively, in the training cohort. In the independent validation cohort, the AUC was 0.88 [95% CI: 0.83–0.92], along with an optimal sensitivity of 81.5% and specificity of 92.9%. The model sensitivity for stage I–II ESCC was 78.4%, which was slightly lower than the sensitivity of the model (85.7%) for stage III–IV ESCC. Conclusion::The dual-target panel based on cfDNA showed excellent performance for detecting ESCC and might be an alternative strategy for screening ESCC.

Objective To analyze the risk factors of gastrointestinal bleeding in patients with type A aortic dissection(TAAD)after Sun's operation.Methods The clinical data of 87 patients who underwent TAAD Sun's operation in our hospital from March 2021 to June 2022 were retrospectively analyzed.They were divided into the bleeding group and the non-bleeding group according to whether there was gastrointestinal bleeding after operation.The clinical data of patients in the two groups was compared and analyzed.The binary Logistic regression analysis was used to analyze the risk factors of gastrointestinal bleeding.The clinical predictor of postoperative gastrointestinal bleeding was analyzed by receiver operating characteristic(ROC)curve.Results In this study,there were 40 cases of postoperative gastrointestinal bleeding(the bleeding group)and 47 cases of non-bleeding(the non-bleeding group).Compared with the non-bleeding group,the bleeding group had a shorter onset time,a higher proportion of patients with hypertension history,a higher preoperative creatinine abnormality rate,more intraoperative blood loss,longer postoperative mechanical ventilation time,higher postoperative infection rate,and higher poor prognosis rate,with statistically significant differences(P<0.05).There was no statistically significant difference in the gender,age,gastrointestinal diseases history,smoking history,preoperative platelets,preoperative international normalized ratio(INR),preoperative alanine aminotransferase(ALT),preoperative aspartate aminotransferase(AST),preoperative γ-glutamyl transpeptidase(GGT),preoperative dissection involving abdominal aorta,operation time,intraoperative cardiopulmonary bypass time,intraoperative circulatory arrest time,intraoperative aortic occlusion time or intraoperative blood transfusion rate.Logistic regression analysis showed that hypertension history(OR=2.468,95%CI:0.862 to 7.067,P=0.037),preoperative creatinine>105 μmol/L(OR=3.970,95%CI:1.352 to 11.659,P=0.011),long postoperative mechanical ventilation time(OR=1.015,95%CI:0.094 to 1.018,P=0.041)and postoperative infection(OR=3.435,95%CI:0.991 to 11.900,P=0.012)were the independent risk factors for postoperative gastrointestinal bleeding in TAAD patients.ROC curve showed that the postoperative mechanical ventilation time exceeding 64 hours were the clinical predictor of postoperative gastrointestinal bleeding in TAAD patients.Conclusion The prognosis of TAAD patients with postoperative gastrointestinal bleeding after Sun's operation is poor.Hypertension history,preoperative acute renal insufficiency,long postoperative mechanical ventilation time and postoperative infection are closely related to postoperative gastrointestinal bleeding in TAAD patients after operation,which should be paid more attention to,and corresponding evaluation,early identification and early intervention should be made to improve the prognosis of patients.

Objective To investigate the feasibility of establishing an animal model of rectovaginal fistula in rabbits by magnetic compression technique.Methods A magnetic device suitable for preparing rabbit rectovaginal fistula model was self-designed.Eight New Zealand female rabbits were used as experimental subjects.They were placed in a supine position after auricular intravenous anesthesia,and the magnets were placed on both sides of the rectovaginal septum through the vagina and anus,respectively,and the magnets were made to attract together.The operation time was recorded,the general state of the experimental rabbits was observed after operation,and the time of magnet discharge was recorded.The experimental rabbits were killed 1 week after operation to obtain rectovaginal fistula specimens,and the formation of rectovaginal fistula was observed by naked eye.Results The animal model of rectovaginal fistula was successfully established in 8 experimental rabbits.The operation process was smooth,with an average time of(1.63±0.70)minutes.The rabbits were generally in good condition after operation.The magnet was discharged from the body at(4.63±0.99)day after operation,and the rectovaginal fistula specimens were obtained 1 week after operation,and the rectovaginal fistula was well formed by naked eye observation.Conclusion The establishment of rabbit rectovaginal fistula model by magnetic compression technique has the advantages of simple operation and high success rate of model preparation.

Objective We aimed to compare the efficacy and prognosis of percutaneous coronary intervention(PCI)in complex and high-risk patients with coronary heart disease(CHD)treated with extracorporeal membrane oxygenation(ECMO)combined with intra-aortic balloon pump(IABP)assistance,and explore the application value of combined use of mechanical circulatory support(MCS)devices in complex PCI.Methods A total of patients who met the inclusion criteria and underwent selective PCI supported by MCS at the Department of Cardiology,the First Affiliated Hospital of the Air Force Medical University from January 2018 to December 2022 were continuously enrolled.According to the mechanical circulatory support method,the patients were divided into ECMO+IABP group and IABP group.Clinical characteristics,angiographic features,in-hospital outcomes,and complications were collected.The intra-hospital outcomes and major adverse cardiovascular events(MACE)at one month and one year after the procedure were observed.The differences and independent risk factors between the two groups in the above indicators were analyzed.Results A total of 218 patients undergoing elective PCI were included,of which 66 patients were in the ECMO+IABP group and 152 patients were in the IABP group.The baseline characteristics of the two groups of patients were generally comparable,but the ECMO+IABP group had more complex lesion characteristics.The proportion of patients with atrial fibrillation(6.1％vs.0.7％,P=0.030),left main disease(43.9％vs.27.0％,P=0.018),triple vessel disease(90.9％vs.75.5％,P=0.009),and RCA chronic total occlusion disease(60.6％vs.35.5％,P＜0.001)was higher in the ECMO+IABP group compared to the IABP group.The proportion of patients with previous PCI history was higher in the IABP group(32.9％vs.16.7％,P=0.014).There was no statistically significant difference in the incidence of in-hospital complications between the two groups(P=0.176),but the incidence of hypotension after PCI was higher in the ECMO+IABP group(19.7％vs.9.2％,P=0.031).The rates of 1-month MACE(4.5％vs.2.6％,P=0.435)and 1-year MACE(7.6％vs.7.9％,P=0.936)were comparable between the two groups.Multivariate analysis showed that in-hospital cardiac arrest(OR 7.17,95％CI 1.27-40.38,P=0.025)and after procedure hypotension(OR 3.60,95％CI 1.10-11.83,P=0.035)were independent risk factors for the occurrence of 1-year MACE.Conclusions Combination use of ECMO+IABP support can provide complex and high-risk coronary heart disease patients with an opportunity to achieve coronary artery revascularization through PCI,and achieve satisfactory long-term prognosis.