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 analyze the efficacy and safety of ustekinumab(UST)in patients with moderate-to-severe Crohn's disease(CD),and to identify factors influencing clinical outcomes.Methods Data were retrospectively collected from patients with moderate-to-severe CD treated with UST in the First Affiliated Hospital of Zhejiang Chinese Medical University and Hangzhou Hospital of Traditional Chinese Medicine between November 2020 and May 2023.Patients were categorized into first-line(not treated with biologic agents,n=68)and second-line(treated with biologic agents,n=66)treatment groups based on prior use of biologic agents.Baseline characteristics,including age,sex,smoking status,disease duration,age at diagnosis,lesion site,disease behavior,perianal disease,history of intestinal surgery,and CD-related drug use,were compared between the two groups.Crohn's disease activity indices(CDAI)were recorded at baseline,week 14,and week 52 to assess the clinical efficacy at weeks 14 and 52.Endoscopic evaluations were performed at baseline and week 52 to evaluate endoscopic efficacy at week 52.The 52-week drug persistence rate and safety profile were also analyzed.Influencing factors related to clinical outcomes were evaluated using univariate and multivariate logistic regression.Results A total of 134 patients with moderate-to-severe CD treated with UST were included.At week 14,clinical response and remission rates were 75.4%(101/134)and 33.6%(45/134),respectively,with no significant difference in clinical efficacy between first-line and second-line groups(clinical response rate:77.9%vs.72.7%,P=0.484;clinical remission rate:38.2%vs.28.8%,P=0.247).At week 52,clinical response and remission rates were 79.9%(107/134)and 56.0%(75/134),respectively.The rates of endoscopic response and remission were 70.9%(95/134)and 38.8%(52/134),respectively.There were no significant differences in clinical efficacy(clinical response rate:80.9%vs.78.8%,P=0.763;clinical remission rate:60.3%vs.51.5%,P=0.306)and endoscopic efficacy(endoscopic response rate:76.5%vs.65.2%,P=0.149;endoscopic remission rate:42.6%vs.34.8%,P=0.354)between the two groups.The 52-week drug persistence rate was 85.8%(115/134),and the adverse reaction rate was 4.5%(6/134).Compared with first-line treatment group,biologic-experienced patients had a significantly higher proportion of dose-optimized therapy in second-line treatment group(45.5%vs.22.1%,P=0.004).Multivariate logistic regression showed that the 14-week clinical response was a significant predictor of 52-week clinical remission,while perianal disease and intestinal surgery history were significant factors associated with treatment failure(P<0.05).Conclusions UST demonstrates significant efficacy in improving clinical and endoscopic outcomes for moderate-to-severe CD patients,with a favorable safety profile.Clinical response at 14 weeks is strongly predictive of clinical remission at 52 weeks.Patients with perianal disease or a history of intestinal surgery were at higher risk of treatment failure.

Atherosclerosis(AS)is an inflammatory cardiovascular disease characterized by plaque accumulation in the arterial wall,leading to increased morbidity and mortality of related cardiovascular disorders.The main pathological mechanisms of AS include lipid deposition,oxidative stress,and chronic inflammation,with disease progression involving endothelial cell dysfunction,macrophage polarization,foam cell formation,and smooth muscle cell proliferation or apoptosis.Mitochondria are essential organelles that provide energy for cellular metabolism,and the mitochondrial quality control(MQC)system is the fundamental mechanism maintaining mitochondrial functional homeostasis.MQC dysfunction can induce vascular phenotype changes through pathways such as oxidative stress,apoptosis,and inflammation,thereby promoting the progression of AS.Therefore,targeting MQC to regulate mitochondrial function may become a new direction for the treatment of AS.This review summarizes the molecular mechanisms of MQC,including mitochondrial biogenesis,mitochondrial dynamics,and mitochondrial autophagy(mitophagy),and further elucidates the role of abnormal MQC in the pathological processes of AS,aiming to provide a scientific basis for identifying potential targets to delay the progression of AS and developing related drugs.

Transcranial temporal interference stimulation (tTIS) is a novel non-invasive neuromodulation technique with the potential to precisely target deep brain structures. This study explores the neural and behavioral effects of tTIS on the superior colliculus (SC), a region involved in eye movement control, in mice. Computational modeling revealed that tTIS delivers more focused stimulation to the SC than traditional transcranial alternating current stimulation. In vivo experiments, including Ca²⁺ signal recordings and eye movement tracking, showed that tTIS effectively modulates SC neural activity and induces eye movements. A significant correlation was found between stimulation frequency and saccade frequency, suggesting direct tTIS-induced modulation of SC activity. These results demonstrate the precision of tTIS in targeting deep brain regions and regulating eye movements, highlighting its potential for neuroscientific research and therapeutic applications.
Animals ; Superior Colliculi/physiology* ; Transcranial Direct Current Stimulation/methods* ; Eye Movements/physiology* ; Male ; Mice ; Mice, Inbred C57BL

Bovine rotavirus(BRV)is an important pathogen causing diarrhea in calves.To understand the genomic charac-teristics and genetic variations in bovine rotavirus,and to further enrich data on the biological characteristics of rotavirus,we aimed to amplify 11 gene segments of the isolated and cultured G6P[1]bovine rotavirus BLL strain,perform whole genome se-quencing,and analyze the molecular characteristics.MEGA7.0 and DNAMAN software were used for homology and typing a-nalysis,and the whole genome phylogenetic tree was constructed to analyze genetic evolution relationships.The complete geno-type of the BLL strain was G6-P[1]-I2-R2-C2-M2-A3-N2-T6-E2-H3.Phylogenetic analysis of the VP7 and VP4 genes of the BLL strain showed that the VP7 gene had the highest homology with RVA/Cow-wt/HB01/China/2021,and the VP4 gene of the BLL strain was in the same branch as RVA/Human-tc/ISR/Ro8059/1995.From the sequence alignment of VP8*amino acids,the sialic acid domain of the BLL strain was found to be similar to that in other P[1]strains,but different from those in other types of strains,except for residue 189,which was the same as that in Ro8059 but different from that in other strains.The results suggested that the BLL strain might potentially infect humans.Therefore,continued monitoring and study of the biological characteristics of this strain are necessary to provide more information and evidence supporting further research on the cross-species transmission of group A rotavirus in China.

Objective:To develop and externally validate a nomogram prediction model for assessing the risk of treatment dropout in allergic rhinitis (AR) patients undergoing sublingual immunotherapy (SLIT).Methods:Between February 2016 and December 2019, data from 358 and 259 AR patients undergoing SLIT were collected from Guizhou Provincial People′s Hospital and Huangshi Central Hospital, respectively. The data included general patient information, dust mite sIgE levels, allergen types, and 22 other clinical variables. Data from Guizhou Provincial People′s Hospital were used as the training set, while data from Huangshi Central Hospital were served as the external validation set. A multivariable Cox regression model was used to identify independent factors associated with SLIT dropout and to develop a nomogram prediction model.Results:Multivariate Cox regression analysis identified several significant factors influencing SLIT dropout, including dust mite sIgE levels (Grade Ⅱ-Ⅳ; HR=1.48, 95% CI: 1.16-1.88), presence of other allergic diseases ( HR=0.47, 95% CI: 0.37-0.61), Rhinoconjunctivitis Quality of Life Questionnaire (RQLQ) score ( HR=0.98, 95% CI: 0.97-1.00), WeChat management ( HR=0.77, 95% CI: 0.60-0.98), treatment efficacy ( HR=0.72, 95% CI: 0.56-0.92), age (5-17 years, HR=0.50, 95% CI: 0.36-0.71;≥60 years, HR=1.42, 95% CI: 1.08-1.87), household income (<5 000 CNY, HR=1.44, 95% CI: 1.09-1.90;>20 000 CNY, HR=0.66, 95% CI: 0.44-0.99), allergen types (single dust mite, HR=0.70, 95% CI: 0.49-0.93; and combined pollen or mold, HR=1.45, 95% CI: 1.02-2.04), and time to efficacy <3 months ( HR=0.73, 95% CI: 0.56-0.94), all P<0.05. At the third-year follow-up, the area under curve (AUC) for the nomogram model was 0.913 (95% CI: 0.881-0.943) in the training group and 0.886 (95% CI: 0.838-0.933) in the validation group. Calibration and decision curve analyses demonstrated the model′s consistency with actual dropout rates and clinical benefit in both groups. Additionally, a Brier score of 0.29 further confirmed the model′s predictive accuracy. Conclusion:We successfully develop a nomogram-based prediction model for SLIT dropout in AR patients, which could assist healthcare professionals in effectively identifying high-risk patients and facilitate the development of more personalized and timely treatment plans aimed at enhancing patient compliance.

Objective:To investigate the prevalence and severity of symptom burden among long-term survivors of oropharyngeal cancer after radiotherapy, to identify core symptom clusters, and to explore their correlation with quality of life.Methods:A previous retrospective study was conducted by the Cancer Hospital, Chinese Academy of Medical Sciences on patients with oropharyngeal cancer who underwent radiotherapy between January 2010 and December 2020. Patients who were still alive as of December 2023 were further followed and analyzed. From December 2023 to August 2024, symptom burden and quality of life were assessed using the Chinese version of the MD Anderson Symptom Inventory–Head and Neck Module (MDASI-HN) and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ). Exploratory factor analysis (principal component analysis with Promax rotation) were used to identify symptom clusters. Spearman correlation analysis was performed to explore the relationship between total symptom cluster scores and standardized domain scores of quality of life. Multivariate linear regression analysis was further employed to determine the relationship between identified symptom clusters and overall quality of life.Results:A total of 273 patients were included, with a median follow-up duration of 6.2 years (range: 3.5-14.5 years) and a median age of 61 years (range: 27-88 years) at follow-up. The top 5 incidence rates of symptom reported by patients were mucus problems in the mouth or throat (147 cases, 53.8%), dental or gum issues (143 cases, 52.4%), xerostomia (140 cases, 51.3%), difficulty swallowing or chewing (95 cases, 34.8%), and taste disturbance (79 cases, 28.9%). Among them, xerostomia was the most serious symptom. The most frequently reported interference was impact on work (including household chores) (55 cases, 20.1%). Exploratory factor analysis identified 3 symptom clusters: fatigue-nausea cluster, eating-voice cluster, and xerostomia-sleep cluster, all of which were significantly correlated with lower overall quality of life of patients (all P<0.001). Conclusion:Long-term survivors of oropharyngeal cancer after radiotherapy experience substantial symptom burden. The fatigue-nausea, eating-voice, and xerostomia-sleep clusters are the core symptom clusters impacting quality of life.