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 explore the material basis of the "interaction of blood stasis and toxin" mechanism in cerebral ischemia-reperfusion injury， as well as the protective role of Huoxue Huayu Jiedu prescription （HXHYJDF） against ferroptosis. MethodsSixty SPF-grade male SD rats were randomly divided into six groups： sham group， model group， deferoxamine （DFO） group （100 mg·kg^-1）， low-dose HXHYJDF group （4.52 g·kg^-1）， medium-dose HXHYJDF group （9.04 g·kg^-1）， and high-dose HXHYJDF group （18.07 g·kg^-1）， with ten rats in each group. Except for the sham group， the other groups were used to replicate the model of focal cerebral ischemia-reperfusion in the middle cerebral artery of rats by the reforming Longa method. Neurological function was assessed at 1^st， 3^rd， 5^th， and 7^th days post-reperfusion using the modified neurological severity scores （m-NSS）. Brain tissue pathology and the morphology of mitochondria were observed using hematoxylin-eosin （HE） staining and transmission electron microscopy. The contents of malondialdehyde （MDA）， glutathione （GSH）， divalent iron ions （Fe²⁺）， and reactive oxygen species （ROS） in the ischemic cerebral tissue were detected using enzyme-linked immunosorbent assay （ELISA）. Immunohistochemistry and Western blot （WB） were used to detect the expression of iron death marker proteins glutathione peroxidase 4 （GPX4）， ferroportin-1 （FPN1）， transferrin receptor protein 1 （TfR1）， and ferritin mitochondrial （FtMt） in brain tissue. ResultsCompared with the sham group， the mNSS score of the model group was significantly increased （P<0.01）. HE staining showed that the number of neurons in the cortex of brain tissue was seriously reduced， and the intercellular space was widened. The nucleus was fragmented， and the cytoplasm was vacuolated. The results of transmission electron microscopy showed that the mitochondria in the cytoplasm contracted and rounded， and the mitochondrial cristae decreased. The matrix was lost and vacuolated， and the density of the mitochondrial bilayer membrane increased. The results of ELISA showed that the content of GSH decreased significantly （P<0.01）， and the contents of MDA， Fe²⁺， and ROS increased significantly （P<0.01）. The results of immunohistochemistry and WB showed that the expression of GPX4 and FPN1 proteins was significantly decreased （P<0.01）， and the expression of FtMt and TfR1 proteins was significantly increased （P<0.01）. Compared with those of the model group， the m-NSS scores of the high-dose and medium-dose HXHYJDF groups began to decrease on the 3^rd and 5^th days， respectively （P<0.05， P<0.01）. The results of HE and transmission electron microscopy showed that the intervention of HXHYJDF improved the pathological changes of neurons and mitochondria. The results of ELISA showed that the content of GSH in the medium-dose and high-dose HXHYJDF groups increased significantly （P<0.01）， and the contents of MDA， Fe²⁺， and ROS decreased significantly （P<0.05， P<0.01）. The content of GSH in the low-dose HXHYJDF group increased significantly （P<0.01）， and the contents of MDA and ROS decreased significantly （P<0.01）. The results of immunohistochemistry showed that the expression of GPX4 and FPN1 in the high-dose HXHYJDF group increased significantly （P<0.01）， and the expression of FtMt and TfR1 decreased significantly （P<0.01）. The expression of GPX4 and FPN1 in the medium-dose HXHYJDF group increased significantly （P<0.05）， and the expression of TfR1 decreased significantly （P<0.01）. WB results showed that the expression levels of FPN1 and GPX4 proteins in the high-dose， medium-dose， and low-dose HXHYJDF groups were significantly up-regulated （P<0.01）， and the expression levels of FtMt and TfR1 proteins were significantly down-regulated （P<0.01）. ConclusionHXHYJDF can significantly improve neurological dysfunction symptoms in rats with cerebral ischemia-reperfusion injury， improve the pathological morphology of the infarcted brain tissue， and protect the brain tissue of rats with cerebral ischemia-reperfusion injury to a certain extent. Neuronal ferroptosis is involved in cerebral ischemia-reperfusion injury， with increased levels of MDA， Fe²⁺， ROS， and TfR1 and decreased levels of FtMt， FPN1， GPX4， and GSH potentially constituting the material basis of the interaction of blood stasis and toxin mechanism in cerebral ischemia-reperfusion injury. HXHYJDF may exert brain-protective effects by regulating iron metabolism-related proteins， promoting the discharge of free iron， reducing brain iron deposition， alleviating oxidative stress， and inhibiting ferroptosis.

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.

Background Schizophrenic patients have metabolic disorders,impaired language and information processing function.Inflammatory factors may play an important role in the occurrence and development of schizophrenia.Objective To explore the relationship of the inflammatory factor levels with metabolic levels,language fluency and information processing function in patients with schizophrenia,so as to provide references for clinical understanding of the neuropathological mechanisms of schizophrenia.Methods A total of 96 patients with schizophrenia were included in the study group,who were hospitalized in the Fourth People's Hospital of Hefei from January 2021 to December 2022 as well as met the diagnostic criteria of Diagnostic and Statistical Manual of Mental Disorders,fifth edition(DSM-5)and Mini-International Neuropsychiatric Interview(MINI)6.0.Meanwhile,population who underwent physical examination at the same hospital were included in the control group(n=42).A high-sensitivity multi factor electrochemiluminescence analyzer was used to detect the levels of inflammatory factors IL-4,IL-5,IL-7,IL-8,IL-10 and IL-13.A fully automated biochemical analyzer was used to detect the levels of metabolic indicators such as fasting blood glucose,triglycerides,high-density lipoprotein,apolipoprotein A,creatinine and urea nitrogen.Verbal fluency and information processing function of all participants were assessed by using Verbal Fluency Test(VFT)and Stroop Color Word Test(SCWT).Results There were statistically significant differences in the levels of IL-4,IL-5,IL-7,IL-8,IL-10,IL-13 and IL-15 between the study group and the control group(P<0.05).There were statistically significant differences in BMI,waist circumference,fasting blood glucose,triglycerides,high-density lipoprotein,urea nitrogen,apolipoprotein A and creatinine levels between the two groups(P<0.05).The differences in the correct number of household appliances,animals,fruits,vegetables,names starting with"water"and"self"in VFT between the two groups were statistically significant(P<0.05).The differences in point reaction time,character reaction time and character color reaction time in SCWT between the two groups were statistically significant(P<0.05).Correlation analysis showed that except for creatinine levels,the levels of IL-4 and IL-5 in patients with schizophrenia were correlated with other indicators(P<0.05).IL-7 levels were correlated with creatinine levels,household appliances,animals,fruits,correct number of names starting with"water"in VFT,point reaction time and word reaction time in SCWT(P<0.05).IL-8 levels were correlated with triglyceride levels,household appliances,animals,fruits,vegetables,correct number of names starting with"water"and"self"in VFT and word reaction time in SCWT(P<0.05).Except for creatinine levels and the correct number of names starting with"self",IL-10 levels were correlated with all other indicators(P<0.05).Except for creatinine and urea nitrogen levels,IL-13 levels were correlated with other indicators(P<0.05).Conclusion The levels of inflammatory factors in patients with schizophrenia may be related to their metabolic levels,language fluency and information processing function.

Objective To observe the alteration of thoracic and lumbar physiological curvature in adolescent idiopathic scoliosis(AIS)and the difference of physiological curvature between different types of scoliosis.Methods A retrospective analysis was conducted on 305 adolescent patients taken full spine X-ray in our hospital from January 2017 to December 2021.The patients were divided into normal group and scoliosis group.The normal group was composed of 179 patients,79 males and 100 females,aged 10 to 18 years old with an average of(12.84±2.10)years old,with cobb agle less than 10 degrees.The scol-iosis group was composed of 126 patients,33 males and 93 females,aged 10 to 18 years old with an average of(13.92±2.20)years old.The gender,age,Risser sign,thoracic kyphosis(TK)and lumbar lordosis(LL)in 2 groups were compared,and the TK and LL were also compared between different genders,different degrees of scoliosis and different segments of scoliosis.Re-sults The female ratio(P=0.001)and age(P＜0.001)in scoliosis group were higher than them in normal group;the ratio of low-grade ossification was higher in normal group than in scoliosis group(P=0.038).TK was significantly smaller in scoliosis group than in normal group(P＜0.001),but there was no significant difference in LL between the 2 groups(P=0.147).There were no significant difference in TK and LL between male and female.The TK was significantly bigger in mild AIS patients than in moderate AIS patients(P＜0.05),but there was no significant difference in LL between mild and moderate patients(P＞0.05).The TK and LL in different segments scoliosis were not found significant difference.Conclusion The physiological curvature of thoracic and lumbar spine is independent of gender.The thoracic physiological curvature becomes smaller in AIS patients,but lumbar curvature remains unchanged.The thoracic physiological curvature in mild AIS patients is greater than that in moderate AIS patients,but the lumbar curvature is almost unchanged between mild and moderate scoliosis and is similar with that in normal adolescent.The alteration of thoracic and lumbar physiological curvature in AIS patients may be related to relative an-terior spinal overgrowth,and the specific detailed mechanism needs to be further studied.

Objective:To explore the clinical and electrophysiological characteristics of peripheral neuropathy in prediabetic patients.Methods:Subjects aged 20-65 years with high-risk factors of impaired glycemia enrolled in Beijing Tiantan Hospital, Capital Medical University from 2019 to 2022 were recruited to conduct oral glucose tolerance test, after excluding other causes of neuropathy or radiculopathy. Patients with impaired fasting glucose or impaired glucose tolerance were defined by American Diabetes Association criteria. These patients were divided into clinical polyneuropathy (PN) and clinical non-PN groups, according to the 2010 Toronto consensus criteria and the presence of PN symptoms and signs or not. Nerve conduction studies (NCS), F wave, sympathetic skin response (SSR), R-R interval variation (RRIV) and current perception thresholds (CPT) were performed and the abnormal rate was compared between different electrodiagnostic methods and between clinical subgroups.Results:Among the 73 prediabetic patients ultimately enrolled, only 20 (27.4%) can be diagnosed as clinical PN according to the Toronto consensus criteria. The abnormal rate of CPT (68.5%, 50/73) was significantly higher than those of F wave (2.7%, 2/73), lower limb NCS (0, 0/73), upper limb NCS changes of carpal tunnel syndrome (26.0%, 19/73), SSR (6.8%, 5/73) and RRIV (5.5%, 4/73; McNemar test, all P<0.001). With sinusoid-waveform current stimuli at frequencies of 2 000 Hz, 250 Hz and 5 Hz, the CPT device was used to measure cutaneous sensory thresholds of large myelinated, small myelinated and small unmyelinated sensory fibers respectively. CPT revealed a 21.9% (16/73) abnormal rate of unmyelinated C fiber in the hands of prediabetic patients, significantly higher than that of large myelinated Aβ fibers [8.2% (6/73), χ2=5.352, P=0.021]. Both abnormal rates of small myelinated Aδ [42.5% (31/73)] and unmyelinated C fibers [39.7% (29/73)] in the feet of prediabetic patients were significantly higher than that of large myelinated Aβ fibers [11.0% (8/73), χ2=18.508, 15.965, both P<0.001]. Compared with the clinical non-PN group, the abnormal rates of CPT [90.0% (18/20) vs 60.4% (32/53), χ2=5.904, P=0.015] and SSR [20.0% (4/20) vs 1.9% (1/53), P=0.016) were significantly higher in the clinical PN group. Conclusions:Peripheral neuropathies in prediabetic patients are usually asymptomatic or subclinical, and predispose to affect unmyelinated and small myelinated sensory fibers. Selective electrodiagnostic measurements of small fibers help to detect prediabetic neuropathies in the earliest stages of the disease.

Objective To explore the efficacy and safety of recombinant human anti-severe acute respiratory syn-drome coronavirus 2(anti-SARS-CoV-2)monoclonal antibody injection(F61 injection)in the treatment of patients with coronavirus disease 2019(COVID-19)combined with renal damage.Methods Patients with COVID-19 and renal damage who visited the PLA General Hospital from January to February 2023 were selected.Subjects were randomly divided into two groups.Control group was treated with conventional anti-COVID-19 therapy,while trial group was treated with conventional anti-COVID-19 therapy combined with F61 injection.A 15-day follow-up was conducted after drug administration.Clinical symptoms,laboratory tests,electrocardiogram,and chest CT of pa-tients were performed to analyze the efficacy and safety of F61 injection.Results Twelve subjects(7 in trial group and 5 in control group)were included in study.Neither group had any clinical progression or death cases.The ave-rage time for negative conversion of nucleic acid of SARS-CoV-2 in control group and trial group were 3.2 days and 1.57 days(P=0.046),respectively.The scores of COVID-19 related target symptom in the trial group on the 3rd and 5th day after medication were both lower than those of the control group(both P＜0.05).According to the clinical staging and World Health Organization 10-point graded disease progression scale,both groups of subjects improved but didn't show statistical differences(P＞0.05).For safety,trial group didn't present any infusion-re-lated adverse event.Subjects in both groups demonstrated varying degrees of elevated blood glucose,elevated urine glucose,elevated urobilinogen,positive urine casts,and cardiac arrhythmia,but the differences were not statistica-lly significant(all P＞0.05).Conclusion F61 injection has initially demonstrated safety and clinical benefit in trea-ting patients with COVID-19 combined with renal damage.As the domestically produced drug,it has good clinical accessibility and may provide more options for clinical practice.