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.

ObjectiveTo observe the clinical effectiveness and safety of Xiaozhong Zhitong Mixture （消肿止痛合剂） combined with antibiotic bone cement in the treatment of diabetic foot ulcer （DFU） with damp-heat obstructing syndrome. MethodsA total of 72 DFU patients with damp-heat obstructing syndrome were randomly assigned to treatment group （36 cases） and the control group （36 cases）. Both groups received standard treatment and topical antibiotic bone cement for ulcer wounds， while the treatment group received oral Xiaozhong Zhitong Mixture （50 ml per time， three times daily） in additionally. Both groups underwent daily wound dressing changes for 21 consecutive days. Ulcer healing rate， serum levels of tumor necrosis factor-alpha （TNF-α）， interleukin-1 beta （IL-1β）， malondialdehyde （MDA）， superoxide dismutase （SOD）， C-reactive protein （CRP）， and white blood cell （WBC） count were observed before and after treatment， and visual analog scale （VAS） scores for wound pain， traditional Chinese medicine （TCM） syndrome scores， and the DFU Healing Scale （DMIST scale） were also compared. Liver and kidney function were evaluated before and after treatment， and adverse events such as allergic reactions， worsening ulcer pain were recorded. ResultsTotally 35 patients in the treatment group and 33 in the control group were included in the final analysis. The ulcer healing rate in the treatment group was （87.93±9.34）%， significantly higher than （81.82±12.02）% in the control group （P = 0.035）. Compared to pre-treatment levels， both groups showed significant reductions in serum CRP， WBC， MDA， IL-1β， and TNF-α levels， with an increase in SOD level （P＜0.05）. TCM syndrome scores， VAS， and DMIST scores also significantly decreased in both groups （P＜0.05）， with greater improvements in the treatment group （P＜0.05）. No significant adverse reactions were observed in either group during treatment. ConclusionXiaozhong Zhitong Mixture combined with antibiotic bone cement has significant advantages in promoting DFU healing， reducing inflammatory response， and alleviating oxidative stress in DFU patients with damp-heat obstructing syndrome， with good safety for DFU patients with damp-heat obstructing syndrome.

AIM: To investigate the correlation of serum Silent mating-type information regulation 2 homolog 1(SIRT1)and Vasostatin-2 content with pathological changes in diabetic retinopathy(DR)patients.METHODS: A total of 104 DR patients(104 eyes)admitted to our hospital from April 2021 to April 2024 were included as the DR group. According to different disease stages, they were assigned into a non-proliferative DR(NPDR)group of 44 cases(44 eyes)and a proliferative DR(PDR)group of 60 cases(60 eyes). Meantime, 104 patients(104 eyes)with simple diabetes were treated as non-DR group. ELISA was applied to detect the levels of SIRT1 and Vasostatin-2 in serum. The diagnostic value of serum SIRT1 and Vasostatin 2 in DR was analyzed by ROC curve. Multivariate Logistic regression was applied to analyze the factors that affected the occurrence of DR. Pearson correlation was applied to analyze the relationship between the levels of SIRT1 and Vasostatin-2 in the serum of DR patients and angiogenesis indicators(VEGF, Ang-2).RESULTS: Compared with the non-DR group, the levels of SIRT1 and Vasostatin-2 in the serum of the DR group were significantly decreased(P<0.05). Compared with the NPDR group, the levels of SIRT1 and Vasostatin-2 in the serum of the PDR group were significantly decreased(P<0.05). Compared with the non-DR group, the levels of VEGF and Ang-2 in the serum of the DR group were obviously higher(P<0.05). Compared with the single detection of serum SIRT1 and Vasostatin-2 levels, combined detection significantly increased the AUC in the diagnosis of DR(Z=4.180, 5.128, all P<0.05). Multivariate Logistic regression analysis showed that HOMA-IR(OR=3.455), fasting blood glucose(OR=1.467), SIRT1(OR=0.836), Vasostatin-2(OR=0.767), VEGF(OR=2.564), and Ang-2(OR=1.834)levels were the influencing factors on the occurrence of DR(all P<0.05). Pearson correlation analysis showed that the levels of SIRT1 and Vasostatin-2 in the serum of DR patients were negatively correlated with VEGF and Ang-2(rSIRT1 vs VEGF=-0.395, rSIRT1 vs Ang-2=-0.474, rVasostatin-2 vs VEGF=-0.323, rVasostatin-2 vs Ang-2=-0.583, all P<0.001).CONCLUSION: The abnormal decrease of serum SIRT1 and Vasostatin 2 levels in DR patients is closely related to the stage of DR lesions and angiogenesis.

Background: s/Aims: Non-invasive models stratifying clinically significant portal hypertension (CSPH) are limited. Herein, we developed a new non-invasive model for predicting CSPH in patients with compensated cirrhosis and investigated whether carvedilol can prevent hepatic decompensation in patients with high-risk CSPH stratified using the new model. Methods: Non-invasive risk factors of CSPH were identified via systematic review and meta-analysis of studies involving patients with hepatic venous pressure gradient (HVPG). A new non-invasive model was validated for various performance aspects in three cohorts, i.e., a multicenter HVPG cohort, a follow-up cohort, and a carvediloltreating cohort. Results: In the meta-analysis with six studies (n=819), liver stiffness measurement and platelet count were identified as independent risk factors for CSPH and were used to develop the new “CSPH risk” model. In the HVPG cohort (n=151), the new model accurately predicted CSPH with cutoff values of 0 and –0.68 for ruling in and out CSPH, respectively. In the follow-up cohort (n=1,102), the cumulative incidences of decompensation events significantly differed using the cutoff values of <–0.68 (low-risk), –0.68 to 0 (medium-risk), and >0 (high-risk). In the carvediloltreated cohort, patients with high-risk CSPH treated with carvedilol (n=81) had lower rates of decompensation events than non-selective beta-blockers untreated patients with high-risk CSPH (n=613 before propensity score matching [PSM], n=162 after PSM). Conclusions Treatment with carvedilol significantly reduces the risk of hepatic decompensation in patients with high-risk CSPH stratified by the new model.

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: s/Aims: Non-invasive models stratifying clinically significant portal hypertension (CSPH) are limited. Herein, we developed a new non-invasive model for predicting CSPH in patients with compensated cirrhosis and investigated whether carvedilol can prevent hepatic decompensation in patients with high-risk CSPH stratified using the new model. Methods: Non-invasive risk factors of CSPH were identified via systematic review and meta-analysis of studies involving patients with hepatic venous pressure gradient (HVPG). A new non-invasive model was validated for various performance aspects in three cohorts, i.e., a multicenter HVPG cohort, a follow-up cohort, and a carvediloltreating cohort. Results: In the meta-analysis with six studies (n=819), liver stiffness measurement and platelet count were identified as independent risk factors for CSPH and were used to develop the new “CSPH risk” model. In the HVPG cohort (n=151), the new model accurately predicted CSPH with cutoff values of 0 and –0.68 for ruling in and out CSPH, respectively. In the follow-up cohort (n=1,102), the cumulative incidences of decompensation events significantly differed using the cutoff values of <–0.68 (low-risk), –0.68 to 0 (medium-risk), and >0 (high-risk). In the carvediloltreated cohort, patients with high-risk CSPH treated with carvedilol (n=81) had lower rates of decompensation events than non-selective beta-blockers untreated patients with high-risk CSPH (n=613 before propensity score matching [PSM], n=162 after PSM). Conclusions Treatment with carvedilol significantly reduces the risk of hepatic decompensation in patients with high-risk CSPH stratified by the new model.

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: s/Aims: Non-invasive models stratifying clinically significant portal hypertension (CSPH) are limited. Herein, we developed a new non-invasive model for predicting CSPH in patients with compensated cirrhosis and investigated whether carvedilol can prevent hepatic decompensation in patients with high-risk CSPH stratified using the new model. Methods: Non-invasive risk factors of CSPH were identified via systematic review and meta-analysis of studies involving patients with hepatic venous pressure gradient (HVPG). A new non-invasive model was validated for various performance aspects in three cohorts, i.e., a multicenter HVPG cohort, a follow-up cohort, and a carvediloltreating cohort. Results: In the meta-analysis with six studies (n=819), liver stiffness measurement and platelet count were identified as independent risk factors for CSPH and were used to develop the new “CSPH risk” model. In the HVPG cohort (n=151), the new model accurately predicted CSPH with cutoff values of 0 and –0.68 for ruling in and out CSPH, respectively. In the follow-up cohort (n=1,102), the cumulative incidences of decompensation events significantly differed using the cutoff values of <–0.68 (low-risk), –0.68 to 0 (medium-risk), and >0 (high-risk). In the carvediloltreated cohort, patients with high-risk CSPH treated with carvedilol (n=81) had lower rates of decompensation events than non-selective beta-blockers untreated patients with high-risk CSPH (n=613 before propensity score matching [PSM], n=162 after PSM). Conclusions Treatment with carvedilol significantly reduces the risk of hepatic decompensation in patients with high-risk CSPH stratified by the new model.

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.