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: and Purpose Symptomatic intracranial hemorrhage (sICH) following endovascular thrombectomy (EVT) is a severe complication associated with adverse functional outcomes and increased mortality rates. Currently, a reliable predictive model for sICH risk after EVT is lacking. Methods: This study used data from patients aged ≥20 years who underwent EVT for anterior circulation stroke from the nationwide Taiwan Registry of Endovascular Thrombectomy for Acute Ischemic Stroke (TREAT-AIS). A predictive model including factors associated with an increased risk of sICH after EVT was developed to differentiate between patients with and without sICH. This model was compared existing predictive models using nationwide registry data to evaluate its relative performance. Results: Of the 2,507 identified patients, 158 developed sICH after EVT. Factors such as diastolic blood pressure, Alberta Stroke Program Early CT Score, platelet count, glucose level, collateral score, and successful reperfusion were associated with the risk of sICH after EVT. The TREAT-AIS score demonstrated acceptable predictive accuracy (area under the curve [AUC]=0.694), with higher scores being associated with an increased risk of sICH (odds ratio=2.01 per score increase, 95% confidence interval=1.64–2.45, P<0.001). The discriminatory capacity of the score was similar in patients with symptom onset beyond 6 hours (AUC=0.705). Compared to existing models, the TREAT-AIS score consistently exhibited superior predictive accuracy, although this difference was marginal. Conclusions The TREAT-AIS score outperformed existing models, and demonstrated an acceptable discriminatory capacity for distinguishing patients according to sICH risk levels. However, the differences between models were only marginal. Further research incorporating periprocedural and postprocedural factors is required to improve the predictive accuracy.

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: and Purpose Symptomatic intracranial hemorrhage (sICH) following endovascular thrombectomy (EVT) is a severe complication associated with adverse functional outcomes and increased mortality rates. Currently, a reliable predictive model for sICH risk after EVT is lacking. Methods: This study used data from patients aged ≥20 years who underwent EVT for anterior circulation stroke from the nationwide Taiwan Registry of Endovascular Thrombectomy for Acute Ischemic Stroke (TREAT-AIS). A predictive model including factors associated with an increased risk of sICH after EVT was developed to differentiate between patients with and without sICH. This model was compared existing predictive models using nationwide registry data to evaluate its relative performance. Results: Of the 2,507 identified patients, 158 developed sICH after EVT. Factors such as diastolic blood pressure, Alberta Stroke Program Early CT Score, platelet count, glucose level, collateral score, and successful reperfusion were associated with the risk of sICH after EVT. The TREAT-AIS score demonstrated acceptable predictive accuracy (area under the curve [AUC]=0.694), with higher scores being associated with an increased risk of sICH (odds ratio=2.01 per score increase, 95% confidence interval=1.64–2.45, P<0.001). The discriminatory capacity of the score was similar in patients with symptom onset beyond 6 hours (AUC=0.705). Compared to existing models, the TREAT-AIS score consistently exhibited superior predictive accuracy, although this difference was marginal. Conclusions The TREAT-AIS score outperformed existing models, and demonstrated an acceptable discriminatory capacity for distinguishing patients according to sICH risk levels. However, the differences between models were only marginal. Further research incorporating periprocedural and postprocedural factors is required to improve the predictive accuracy.

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: and Purpose Symptomatic intracranial hemorrhage (sICH) following endovascular thrombectomy (EVT) is a severe complication associated with adverse functional outcomes and increased mortality rates. Currently, a reliable predictive model for sICH risk after EVT is lacking. Methods: This study used data from patients aged ≥20 years who underwent EVT for anterior circulation stroke from the nationwide Taiwan Registry of Endovascular Thrombectomy for Acute Ischemic Stroke (TREAT-AIS). A predictive model including factors associated with an increased risk of sICH after EVT was developed to differentiate between patients with and without sICH. This model was compared existing predictive models using nationwide registry data to evaluate its relative performance. Results: Of the 2,507 identified patients, 158 developed sICH after EVT. Factors such as diastolic blood pressure, Alberta Stroke Program Early CT Score, platelet count, glucose level, collateral score, and successful reperfusion were associated with the risk of sICH after EVT. The TREAT-AIS score demonstrated acceptable predictive accuracy (area under the curve [AUC]=0.694), with higher scores being associated with an increased risk of sICH (odds ratio=2.01 per score increase, 95% confidence interval=1.64–2.45, P<0.001). The discriminatory capacity of the score was similar in patients with symptom onset beyond 6 hours (AUC=0.705). Compared to existing models, the TREAT-AIS score consistently exhibited superior predictive accuracy, although this difference was marginal. Conclusions The TREAT-AIS score outperformed existing models, and demonstrated an acceptable discriminatory capacity for distinguishing patients according to sICH risk levels. However, the differences between models were only marginal. Further research incorporating periprocedural and postprocedural factors is required to improve the predictive accuracy.

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::Although overnight fasting is recommended prior to endoscopic retrograde cholangiopancreatography (ERCP), the benefits and safety of high-carbohydrate fluid diet (CFD) intake 2 h before ERCP remain unclear. This study aimed to analyze whether high-CFD intake 2 h before ERCP can be safe and accelerate patients’ recovery.Methods::This prospective, multicenter, randomized controlled trial involved 15 tertiary ERCP centers. A total of 1330 patients were randomized into CFD group ( n = 665) and fasting group ( n = 665). The CFD group received 400 mL of maltodextrin orally 2 h before ERCP, while the control group abstained from food/water overnight (>6 h) before ERCP. All ERCP procedures were performed using deep sedation with intravenous propofol. The investigators were blinded but not the patients. The primary outcomes included postoperative fatigue and abdominal pain score, and the secondary outcomes included complications and changes in metabolic indicators. The outcomes were analyzed according to a modified intention-to-treat principle. Results::The post-ERCP fatigue scores were significantly lower at 4 h (4.1 ± 2.6 vs. 4.8 ± 2.8, t = 4.23, P <0.001) and 20 h (2.4 ± 2.1 vs. 3.4 ± 2.4, t= 7.94, P <0.001) in the CFD group, with least-squares mean differences of 0.48 (95% confidence interval [CI]: 0.26–0.71, P <0.001) and 0.76 (95% CI: 0.57–0.95, P <0.001), respectively. The 4-h pain scores (2.1 ± 1.7 vs. 2.2 ± 1.7, t = 2.60, P = 0.009, with a least-squares mean difference of 0.21 [95% CI: 0.05–0.37]) and positive urine ketone levels (7.7% [39/509] vs. 15.4% [82/533], χ2 = 15.13, P <0.001) were lower in the CFD group. The CFD group had significantly less cholangitis (2.1% [13/634] vs. 4.0% [26/658], χ2 = 3.99, P = 0.046) but not pancreatitis (5.5% [35/634] vs. 6.5% [43/658], χ2 = 0.59, P = 0.444). Subgroup analysis revealed that CFD reduced the incidence of complications in patients with native papilla (odds ratio [OR]: 0.61, 95% CI: 0.39–0.95, P = 0.028) in the multivariable models. Conclusion::Ingesting 400 mL of CFD 2 h before ERCP is safe, with a reduction in post-ERCP fatigue, abdominal pain, and cholangitis during recovery.Trail Registration::ClinicalTrials.gov, No. NCT03075280.

Objective To explore clinical effect of distal tibial tubercle-high tibial osteotomy(DTT-HTO)in treating knee osteoarthritis(KO A)with medial meniscus posterior root tear(MMPRT).Methods A retrospective analysis was performed on 21 patients with varus KOA with MMPRT from May 2020 to December 2021,including 3 males and 18 females,aged from 49 to 75 years old with an average of(63.81±6.56)years old,the courses of disease ranged from 0.5 to 18.0 years with an average of(5.9±4.2)years,and 4 patients with grade Ⅱ,14 patients with grade Ⅲ,and 3 patients with grade Ⅳ according to Kellgren-Lawrence;14 patients with type 1 and 7 patients with type 2 according to MMPRT damage classification.The distance of medi-al meniscusextrusion(MME)and weight-bearing line ratio(WBLR)of lower extremity were compared before and 12 months after operation.Visual analogue scale(V AS),Western Ontarioand and McMaster Universities(WOMAC)osteoarthritis index,and Lysholm knee score were used to evaluate knee pain and functional improvement before operation,1,6 and 12 months after operation,respectively.Results Twenty-one patients were followed up for 12 to 18 months with an average of(13.52±1.72)months.MME distance was improved from(4.99±1.05)mm before operation to(1.87±0.76)mm at 12 months after operation(P＜0.05).WBLR was increased from(15.49±7.04)％before operation to(62.71±2.27)％at 12 months after operation(P＜0.05).VAS was decreased from(7.00±1.14)before operation to(2.04±0.80),(0.90±0.62)and(0.61±0.50)at 1,6 and 12 months after operation.WOMAC were decreased from preoperative(147.90±9.88)to postoperative(103.43±8.52),(74.00±9.54)and(47.62±9.53)at 1,6 and 12 months,and the difference were statistically significant(P＜0.05).Lysholm scores were increased from(46.04±7.34)before oepration to(63.19±8.93),(81.10±6.41)and(89.29±3.04)at 1,6 and 12 months after operation(P＜0.05).Conclusion For the treatment of varus KOA with MMPRT,DTT-HTO could reduce medial meniscus pro-trusion distance,improve the ratio of lower limb force line,and effectively reduce knee pain and improve knee joint function.