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: Patients with severe symptomatic aortic stenosis are assessed for coronary artery disease (CAD) prior to transcatheter aortic valve implantation (TAVI) with treatment implications. Invasive coronary angiography (ICA) is the recommended modality but is associated with peri-procedural complications. Integrating machine learning (ML)-based computed tomography-derived fractional flow reserve (CT-FFR) into existing TAVI-planning CT protocol may aid exclusion of significant CAD and thus avoiding ICA in selected patients. Materials and Methods: A single-center, retrospective study was conducted, 41 TAVI candidates with both TAVI-planning CT and ICA performed were analyzed. CT datasets were evaluated by a ML-based CT-FFR software. Beta-blocker and nitroglycerin were not administered in these patients. The primary outcome was to identify significant CAD. The diagnostic performance of CT-FFR was compared against ICA. Results: On per-patient level, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and diagnostic accuracy were 89%, 94%, 80%, 97% and 93%, respectively. On per-vessel level, the sensitivity, specificity, PPV, NPV and diagnostic accuracy were 75%, 94%, 67%, 96% and 92%, respectively. The area under the receiver operative characteristics curve per individual coronary vessels yielded overall 0.90 (95% confidence interval 85%–95%). ICA may be avoided in up to 80% of patients if CT-FFR results were negative. Conclusion ML-based CT-FFR can provide accurate screening capabilities for significant CAD thus avoiding ICA. Its integration to existing TAVI-planning CT is feasible with the potential of improving the safety and efficiency of pre-TAVI CAD assessment.

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 compare the clinical efficacy and safety of remimazolam and propofol in general anesthesia induction and maintenance for elderly patients undergoing thoracoscopic lobectomy. METHODS A total of 86 elderly lung cancer patients who underwent thoracoscopic lobectomy at Chongqing University Three Gorges Hospital from February to July 2024 were selected and divided into the propofol group and the remimazolam group according to the randomized numerical table method， with 43 cases in each group. During anesthesia induction， patients in the propofol group and the remimazolam group were intravenously administered 2 mg/kg of Propofol medium- and long-chain fat emulsion injection or 0.25 mg/kg of Remimazolam tosilate for injection， respectively； during anesthesia maintenance， the two groups received intravenous infusion of 6-10 mg/（kg·h） of Propofol medium- and long- chain fat emulsion injection or 1-3 mg/（kg·h） of Remimazolam tosilate for injection， respectively. The anesthesia effects， anesthesia-related indicators， intraoperative opioid and muscle relaxant dosages， Ramsay sedation score， numerical rating scale （NRS） score， and hemodynamic parameters were compared between the two groups， and the occurrence of adverse drug reactions was recorded. RESULTS A total of 41 patients in the propofol group and 43 patients in the remimazolam group completed the trial. The proportion of patients with grade Ⅰ anesthesia effect in the remimazolam group was significantly higher than that in the propofol group， while the proportion of patients with grade Ⅱ anesthesia effect was significantly lower than that in the propofol group （P＜0.05）. In this group， the disappearance time of eyelash reflex， the time taken for the bispectral index to drop to 60， and the Ramsay sedation scores （2 and 6 hours after operation） were all significantly prolonged or increased， while the recovery time， NRS scores （2 and 6 hours after operation）， and the incidence of intraoperative hypotension were all significantly shortened or reduced； moreover， the improvements of the above sedation/NRS scores exhibited a time-dependent pattern within 2 to 24 hours after operation （P＜0.05）. Compared with before anesthesia induction （T0）， the heart rate [except at 2 min after medication （T1）， 60 min after anesthesia （T4）， and at the end of surgery （T5） in the remimazolam group] and mean arterial pressure [except at T1 in the remimazolam group] of patients in both groups significantly decreased at T1， 5 min after medication （T2）， at the start of surgery （T3）， T4， and T5 （P＜0.05）. Meanwhile， regional cerebral oxygen saturation significantly increased in both groups. Furthermore， the heart rate and mean arterial pressure of patients in the remimazolam group were significantly higher than those in the propofol group at T1， T2 and T4 （P＜0.05）. No statistically significant differences were observed between the two groups in terms of postanesthesia care unit stay time， dosage of opioids and muscle relaxants， regional cerebral oxygen saturation， or peripheral oxygen saturation at various time points （P＞0.05）. CONCLUSIONS Compared to propofol， remimazolam demonstrates superior anesthesia effects when used for the induction and maintenance of general anesthesia in elderly patients undergoing thoracoscopic lobectomy. It not only provides more stable intraoperative hemodynamics and shortens the postoperative recovery time but also effectively reduces the incidence of intraoperative hypotension.

Aim To screen and study the expression of long non-coding RNA (IncRNA) in rats with middle cerebral artery occlusion (MCAO) with MCAO treated with Tao Hong Si Wu decoction (THSWD) and determine the possible molecular mechanism of THSWD in treating MCAO rats. Methods Three cerebral hemisphere tissue were obtained from the control group, MCAO group and MCAO + THSWD group. RNA sequencing technology was used to identify IncRNA gene expression in the three groups. THSWD-regulated IncRNA genes were identified, and then a THSWD-regu-lated IncRNA-mRNA network was constructed. MCODE plug-in units were used to identify the modules of IncRNA-mRNA networks. Gene ontology (GO) and kyoto encyclopedia of genes and genomes (KEGG) were used to analyze the enriched biological functions and signaling pathways. Cis- and trans-regulatory genes for THSWD-regulated IncRNAs were identified. Reverse transcription real-time quantitative pol-ymerase chain reaction (RT-qPCR) was used to verify IncRNAs. Molecular docking was used to identify IncRNA-mRNA network targets and pathway-associated proteins. Results In MCAO rats, THSWD regulated a total of 302 IncRNAs. Bioinformatics analysis suggested that some core IncRNAs might play an important role in the treatment of MCAO rats with THSWD, and we further found that THSWD might also treat MCAO rats through multiple pathways such as IncRNA-mRNA network and network-enriched complement and coagulation cascades. The results of molecular docking showed that the active compounds gallic acid and a-mygdalin of THSWD had a certain binding ability to protein targets. Conclusions THSWD can protect the brain injury of MCAO rats through IncRNA, which may provide new insights for the treatment of ischemic stroke with THSWD.

Objective To investigate different doses of the analgesic effects of dezocine comparing with sufentanil after laryngectomy.Methods A total of 129 patients who underwent elective partial laryngectomy from Feb 2022 to Jan 2023 were randomly assigned to dezocine 0.5 mg/kg group(group D1),dezocine 0.6 mg/kg group(group D2)and sufentanil 2 μg/kg group(group S).Twenty-four hours amount of drugs,the visual analogue scale(visual analogue scale,VAS)and 48 h total pressing times of PCA(patient-controlled intravenous analgesia,PCIA)were compared among the three groups at 6,12,24 and 48 h after operation,and the postoperative adverse reactions(nausea,vomiting,dizziness,urinary retention and respiratory depression)were recorded.Results There was no significant difference in 24 h amount of drugs among the three groups.The VAS score of group D1 was higher than that of group S at 6 h postoperatively(P＜0.05),but did not differ significantly among the three groups at 12,24 and 48 h.There was no significant differences in the number of compressions and postoperative adverse reactions among the three groups.Conclusion Compared with sufentanil,0.6 mg/kg dezocine can provide the same degree of analgesic effect.However,no advantage was found to reduce adverse reactions.

Tricuspid regurgitation(TR)cases are widely distributed in China.Poor clinical drug efficacy,high surgical risk,and poor prognosis for right heart failure are found in patients with moderate or severe TR.In recent years,with the innovation of valve instruments and the development of technology,transcatheter tricuspid valve treatment could be a new choice for high-risk TR patients in surgery.Many TR animal models have emerged these years for the research of the mechanism of TR and for the clinical verification of instruments.Therefore,this review focuses on how to develop an animal model of TR and discusses the advantages and disadvantages of these techniques.