Objective To explore the value of deep learning(DL)model based on grayscale ultrasound for predicting asymptomatic advanced chronic liver disease(cACLD).Methods Totally 258 patients with asymptomatic compensatory chronic liver diseases were retrospectively included,among them 117 with F3 or F4 stage liver fibrosis were classified into cACLD group,while 141 with F1 or F2 stage liver fibrosis were taken as non-cACLD group.The patients were divided into training set(n=180,including 82 cases of cACLD and 98 cases of non-cACLD)and validation set(n=78,including 35 cases of cACLD and 43 cases of non-cACLD)at the ratio of 7∶3.Univariate and multivariate logistic regression were used to screen independent clinical predictors of cACLD and construct a clinical model.Based on liver grayscale ultrasound,optimal DL features were extracted and screened,and Resnet50 network was adopted as framework,na?ve Bayes classifier was used to construct DL model,and a combined model was constructed based on clinical model and DL model.The efficacy and clinical value of each model for predicting asymptomatic cACLD were evaluated.Results Age,gamma-glutamyl transferase and platelet count were all independent clinical predictors of cACLD,and a clinical model was constructed.Totally 38 optimal DL features were screened to build a DL model.The AUC of combined model in training set and validation set was 0.950 and 0.740,of DL model was 0.944 and 0.737,respectively,being not significantly different(both P＞0.05)but all higher than that of clinical model(0.667 and 0.573,all P＜0.05).Taken 0.59-0.90 as the threshold,the net benefits of combined model in both training and validation sets were higher than that of other models.Conclusion DL model based on grayscale ultrasound could be used to effectively predict asymptomatic cACLD.Combining with clinical characteristics might improve clinical net benefit of this model.

Chronic pain causes physical suffering and can have major psychological impacts in patients.Chronic pain can induce depressive disorder,and clinical studies have consistently shown that chronic pain and depression frequently co-occur,suggesting the possibility of shared pathogenic mechanisms underlying these conditions.Acupuncture,as an alternative therapy,has been widely used for analgesia and to treat depression,with demonstrated clinical efficacy.The therapeutic mechanism of acupuncture is related to neural and endocrine regulation.This review considers the mechanism of chronic pain accompanied by depression,in relation to the brain regions and neural circuits affected by acupuncture treatment.This review provides a new approach for the treatment of depression caused by chronic pain.

Chronic pain causes physical suffering and can have major psychological impacts in patients.Chronic pain can induce depressive disorder,and clinical studies have consistently shown that chronic pain and depression frequently co-occur,suggesting the possibility of shared pathogenic mechanisms underlying these conditions.Acupuncture,as an alternative therapy,has been widely used for analgesia and to treat depression,with demonstrated clinical efficacy.The therapeutic mechanism of acupuncture is related to neural and endocrine regulation.This review considers the mechanism of chronic pain accompanied by depression,in relation to the brain regions and neural circuits affected by acupuncture treatment.This review provides a new approach for the treatment of depression caused by chronic pain.

Objective To explore the value of deep learning(DL)model based on grayscale ultrasound for predicting asymptomatic advanced chronic liver disease(cACLD).Methods Totally 258 patients with asymptomatic compensatory chronic liver diseases were retrospectively included,among them 117 with F3 or F4 stage liver fibrosis were classified into cACLD group,while 141 with F1 or F2 stage liver fibrosis were taken as non-cACLD group.The patients were divided into training set(n=180,including 82 cases of cACLD and 98 cases of non-cACLD)and validation set(n=78,including 35 cases of cACLD and 43 cases of non-cACLD)at the ratio of 7∶3.Univariate and multivariate logistic regression were used to screen independent clinical predictors of cACLD and construct a clinical model.Based on liver grayscale ultrasound,optimal DL features were extracted and screened,and Resnet50 network was adopted as framework,na?ve Bayes classifier was used to construct DL model,and a combined model was constructed based on clinical model and DL model.The efficacy and clinical value of each model for predicting asymptomatic cACLD were evaluated.Results Age,gamma-glutamyl transferase and platelet count were all independent clinical predictors of cACLD,and a clinical model was constructed.Totally 38 optimal DL features were screened to build a DL model.The AUC of combined model in training set and validation set was 0.950 and 0.740,of DL model was 0.944 and 0.737,respectively,being not significantly different(both P＞0.05)but all higher than that of clinical model(0.667 and 0.573,all P＜0.05).Taken 0.59-0.90 as the threshold,the net benefits of combined model in both training and validation sets were higher than that of other models.Conclusion DL model based on grayscale ultrasound could be used to effectively predict asymptomatic cACLD.Combining with clinical characteristics might improve clinical net benefit of this model.

OBJECTIVE To identify the influence factors associated with 28-day fatality among COVID-19 hospitalizations and to analyze the interaction between the disease severity at admission and the use of hospital preparations.METHODS A retrospective review of records from COVID-19 hospitalizations aged 18 to 90 who were admitted to the Jiangsu Province Hospital of Chinese Medi-cine from December 15,2022 to January 15,2023 were conducted.Patients who died or were lost to follow-up within 48 h of admis-sion were excluded.Patients were divided into survival and death groups based on their 28-day fatality status.Descriptive statistics were used to characterize the two groups and multivariate logistic regression models were employed to identify factors influencing 28-day fatality risk.The interaction between the severity of illness at admission and the use of hospital preparations was explored through cross-over analysis and hierarchical regression analysis.RESULTS Significant differences were observed between the survival and death groups in terms of severity of illness at admission,hospital preparations usage,steroid therapy,age,platelet count,lactate dehydro-genase levels,and urea(P<0.05.Crossover analysis and hierarchical logistic regression analysis revealed a significant antagonistic interaction between the severity of illness at admission and the use of hospital formulations both before adjustment(RERI=-20.678,95%CI:-33.703～-7.652;APAI=-2.301,95%CI:-4.027～-0.575 and after adjusting for gender,age,clinical characteristics and further adjusting for laboratory parameters(RERI=-5.972,95%CI:-10.564～-1.380;APAI=-2.205,95%CI:-4.131～-0.279,and it was an antagonistic interaction both before(SI=0.279,95%CI:0.157～0.493 and after adjustment(SI=0.222,95%CI:0.095～0.523.CONCLUSION The use of hospital preparations significantly reduces the 28-day fatality risk among COV-ID-19 hospitalizations and a clear antagonistic interaction was observed between the disease severity at admission and the use of hospi-tal preparations.

Objective:To investigate the correlation between the level of thyrotropin receptor antibody(TRAb) and bone turnover markers(BTMs) in the patients with newly-diagnosed Graves′ disease(GD).Methods:The clinical data of GD patients who were newly-diagnosed in the First Affiliated Hospital of Zhengzhou University from October 2016 to June 2021 were collected, including free triiodothyronine(FT 3), free thyroxine(FT 4), thyroid stimulating hormone, thyroid related antibodies, N-terminal procollagen of type I collagen(PINP), N-terminal osteocalcin(N-MID), β-cross-linked C-telopeptide of type I(β-CTX), blood lipid and renal function, etc. Results:There were 618 GD patients with an average age of(43.7±13.2) years(male∶female=1∶1.99). The PINP and β-CTX level in male GD patients were significantly higher than those in female(all P<0.05). Spearman correlation analysis showed that PINP, N-MID and β-CTX were positively correlated with FT 3, FT 4, TRAb, serum calcium and serum phosphorus; and negatively correlated with body mass index and low density lipoprotein cholesterol(all P<0.05). Linear regression analysis showed that TRAb was positively correlated with lg-PINP, lg-N-MID and sqrt-β-CTX in the univariate model of total GD patients( β were 0.006, 0.005, and 0.006, respectively; all P<0.001); positive correlation remained after adjusting for thyroid function(all β=0.004, all P<0.001); and for multiple confounding factors(model 3 and 4, all P<0.05). Results of univariate and adjusted thyroid function models with GD in different genders were consistent with the total patients(all P<0.05). Conclusion:TRAb is a risk factor for accelerated bone turnover in GD patients which is independent of thyroid function.

Objective:To explore independent predictors for poor outcome at 3 months in elderly patients with acute cerebral infarction(ACI)treated with intravenous thrombolysis(IVT), and to develop a nomogram-based predictive model.Methods:This was a retrospective cohort study.Clinical, laboratory and imaging data of 346 elderly patients with ACI treated with IVT from January 2016 to April 2021 in our hospital were collected.Poor outcome was defined as a modified Rankin Scale(mRS)score >2 at 3 months after the stroke.Logistic regression analysis was used to screen for independent factors predicting poor outcome in elderly ACI patients treated with IVT, and a corresponding nomogram model was developed using the R software.The ROC curve, calibration plots and decision curve analysis were used to evaluate discrimination, calibration and clinical application value of the nomogram model.Results:Among 346 candidates, 109 developed a poor outcome, representing a rate of 31.5%.Logistic regression analysis showed that symptomatic hemorrhagic transformation( OR=15.647, 95% CI: 8.913-27.454), stroke severity(moderate stroke, OR=3.322, 95% CI: 1.414-7.811; moderate-severe stroke, OR=8.169, 95% CI: 4.102-16.258; severe stroke, OR=9.653, 95% CI: 5.440-17.121), stroke-associated pneumonia( OR=2.239, 95% CI: 1.134-4.420), and heart failure( OR=2.758, 95% CI: 1.424-5.336)were independent predictors for poor outcome at 3 months in elderly ACI patients treated with intravenous thrombolysis(all P<0.05). With the area under curve(AUC-ROC)value at 0.85(95% CI: 0.80-0.89), the nomogram model, which was composed of the above four predictors, demonstrated good discrimination.On the calibration plot, the mean absolute error was 0.020, indicating that the model had good calibration.Decision curve analysis revealed that the model had good clinical application value. Conclusions:The nomogram model composed of symptomatic hemorrhagic transformation, stroke severity, stroke-associated pneumonia and heart failure may predict poor outcome at 3 months in elderly ACI patients treated with IVT, with high prediction accuracy and high clinical application value.

Objective:To analyze the distribution and drug resistance of wound pathogenic microorganisms in outpatients of wound healing center so as to provide a basis for the standardized construction of wound healing centers.Methods:A retrospective case series study was used to analyzed the data of 365 outpatients treated at Ruijin Hospital, Shanghai Jiaotong University School of Medicine from December 2017 to October 2019. There were 220 males and 145 females, aged (58.8±18.9)years (range, 18-98 years). The patients included 92 first-visit patients and 273 re-visit patients. The culture results (positive rate of pathogenic microorganisms, bacterial species, bacterial distribution) and drug sensitivity results of the wound secretions were compared and analyzed.Results:(1) Among 365 samples of wound secretions, 198 patients were positive for pathogenic microorganisms with a positive rate of 54.3%. A total of 107 strains (51.0%) of Gram-positive bacteria were detected, mainly Staphylococcus aureus (70 strains, 33.3%); 95 strains (45.2%) of Gram-negative bacteria were detected, mainly Escherichia coli (20 strains, 9.5%), followed by Pseudomonas aeruginosa (17 strains, 8.1%); 8 strains (3.8%) of fungi were detected. (2) A total of 26 (28.3%) first-visit patients were positive for pathogenic microorganisms, and 172 (63.0%) re-visit patients were positive for pathogenic microorganisms. The rate of positive microorganism detection had significant differences between first-visit and re-visit patients ( P<0.05). (3) A total of 29 strains were detected in first-visit patients, including 16 strains (55.2%) of Gram-positive bacteria, 11 strains (37.9%) of Gram-negative bacteria and 2 strains (6.9%) of fungi. A total of 181 strains were detected in re-visit patients, including 91 strains (50.3%) of Gram-positive bacteria, 84 strains (46.4%) of Gram-negative bacteria and 6 strains (3.3%) of fungi. The microbial distribution was significantly different between first-visit and re-visit patients ( P<0.05). (4) Compared with first-visit patients, the resistance of Staphylococcus aureus isolated from the re-visit patients to spenicillin, oxacillin, ciprofloxacin, tetracycline, clindamycin, moxifloxacin, erythromycin, and levofloxacin were increased variably. No vancomycin-resistant Staphylococcus aureus was detected, indicating that the staphylococcus aureus presented in the wound was highly sensitive to vancomycin. Conclusions:Staphylococcus aureus is the most common microorganism in wound secretions in outpatients of wound healing center. The rate of positive pathogenic microorganisms in wound secretions of re-visit patients is significantly higher than that of first-visit patients, and the distribution of pathogenic microorganisms of first-visited and revisited patients differs significantly. The Staphylococcus aureus detected in re-visit patients has a higher resistance to common antibiotics compared with first-visit patients. It is suggested that timely detection of pathogenic microorganisms in outpatients and effective control and supervision of outpatient infections are important contents that cannot be ignored in the construction of wound healing center.