Objective To explore the diagnostic efficacy of magnetic resonance imaging(MRI)combined with serum carbohydrate antigen 125(CA125)and angiopoietin like protein 2(ANGPTL2)in breast cancer.Methods From February 2020 to June 2022,152 patients with breast diseases diagnosed initially in our hospital were collected as the subjects of this study.According to the results of pathological tissue examination(breast cancer modified radical mastectomy operation or needle biopsy),they were grouped into breast cancer group(90 cases)and benign group(62 cases).All patients underwent MRI examination.Measurement of serum CA125 and ANGPTL2 levels;receiver Operating Characteristic curve(ROC)were analysed for the critical diagnostic points of breast cancer by MRI combined with serum CA125 and ANGPTL2;and four grid table was applied to analyze the diagnostic value of MRI combined with serum CA125 and ANGPTL2 in breast cancer.Results The ring enhancement in breast cancer group was obviously higher than that in benign group(P＜0.05),and the uniform enhancement in benign group was obviously higher than that in breast cancer group(P＜0.05).The levels of serum CA125 and ANGPTL2 in breast cancer group were obviously higher than those in benign group(P＜0.05).According to the ROC,the AUC of serum CA125 in the diagnosis of breast cancer was 0.870(95％CI:0.815～0.924)and the cut-off value was 29.574 U/ml,the AUC of serum ANGPTL2 CA125 in the diagnosis of breast cancer was 0.893(95％CI:0.843～0.942),the cut-off value was 6.085 ng/ml,and the AUC of MRI diagnosis of breast cancer was 0.891(95％CI:0.832～0.950).The accuracy of MRI,CA125,ANGPTL2 and combination of three indexes in the diagnosis of breast cancer were 89.47％,84.87％,82.89％,93.42％respectively.Conclusion MRI combined with serum CA125 and ANGPTL2 can improve the diagnostic efficacy of breast cancer.

Objective To explore the role and value of thromboelastography(TEG)combined with white matter hyperintensity(WMH)in predicting the functional prognosis of patients with non-cardiogenic acute ischemic stroke(AIS)through machine learning.Methods This study included 130 patients with non-cardiogenic AIS from August 2022 to February 2024.General clinical data,TEG and WMH information of all patients were collected.Three months later,functional outcomes were followed up using the modified Rankin scale(mRS),with an mRS score of≥2 indicating a poor prognosis.The prediction models were divided into four feature sets according to different ranges of predictors:set A(general clinical data+TEG indicators+WMH score),set B(general clinical data+TEG indicators),set C(general clinical data+WMH score),and set D(general clinical data).For each feature set,three machine learning algorithms,traditional logistic regression(LR)model,random forests(RF),neural network(NNET),and K-nearest neighbors(KNN),were used to construct models for predicting the 3-month neurological function outcome of patients with non-cardiogenic AIS.Bootstrap resampling internal validation was used to compare the performance of prediction models.Results The training and testing of the model were performed on 130 patient samples,and the AUC value and its confidence interval of the model were corrected by the 0.632+method(optimism correction).For the LR,NNET,and KNN models,the corrected AUC values of feature set A were significantly better than those of feature set D(DeLong test,P<0.05).For all models,the corrected AUC value of feature set A was higher than that of other feature sets.For feature set A,the corrected AUC value(0.830)of the NNET model was higher than that of other models.Among the 19 features of feature set A,six features with important associations with functional prognosis were selected including National Institute of Health stroke scale(NIHSS)score,stroke history,small artery occlusion subtype,periventricular white matter hyperintensities(PWMH)score,and TEG indicators maximum amplitude(MA)and LY30.Conclusion Combining TEG indicators and WMH information on the basis of general clinical data can significantly improve the accuracy of predicting poor functional prognosis in patients with non-cardiogenic AIS.The prediction models established by machine learning-based NNET and KNN algorithms have high predictive value.

Objective To explore the role and value of thromboelastography(TEG)combined with white matter hyperintensity(WMH)in predicting the functional prognosis of patients with non-cardiogenic acute ischemic stroke(AIS)through machine learning.Methods This study included 130 patients with non-cardiogenic AIS from August 2022 to February 2024.General clinical data,TEG and WMH information of all patients were collected.Three months later,functional outcomes were followed up using the modified Rankin scale(mRS),with an mRS score of≥2 indicating a poor prognosis.The prediction models were divided into four feature sets according to different ranges of predictors:set A(general clinical data+TEG indicators+WMH score),set B(general clinical data+TEG indicators),set C(general clinical data+WMH score),and set D(general clinical data).For each feature set,three machine learning algorithms,traditional logistic regression(LR)model,random forests(RF),neural network(NNET),and K-nearest neighbors(KNN),were used to construct models for predicting the 3-month neurological function outcome of patients with non-cardiogenic AIS.Bootstrap resampling internal validation was used to compare the performance of prediction models.Results The training and testing of the model were performed on 130 patient samples,and the AUC value and its confidence interval of the model were corrected by the 0.632+method(optimism correction).For the LR,NNET,and KNN models,the corrected AUC values of feature set A were significantly better than those of feature set D(DeLong test,P<0.05).For all models,the corrected AUC value of feature set A was higher than that of other feature sets.For feature set A,the corrected AUC value(0.830)of the NNET model was higher than that of other models.Among the 19 features of feature set A,six features with important associations with functional prognosis were selected including National Institute of Health stroke scale(NIHSS)score,stroke history,small artery occlusion subtype,periventricular white matter hyperintensities(PWMH)score,and TEG indicators maximum amplitude(MA)and LY30.Conclusion Combining TEG indicators and WMH information on the basis of general clinical data can significantly improve the accuracy of predicting poor functional prognosis in patients with non-cardiogenic AIS.The prediction models established by machine learning-based NNET and KNN algorithms have high predictive value.

Objective:To explore the risk factors of fatty liver occurrence in postoperative breast cancer patients after endocrine therapy, and establish a risk prediction model.Methods:A total of 120 breast cancer patients who received endocrine therapy after surgery in Huanggang Central Hospital from June 2014 to June 2016 were retrospectively selected, and another 120 breast cancer patients who did not receive endocrine therapy after surgery in the same period were selected as the control group. The difference of prognosis between patients treated with endocrine therapy or not was compared. According to the occurrence of fatty liver after endocrine therapy, the patients were divided into fatty liver group (63 cases) and non-fatty liver group (57 cases). Multivariate logistic regression was used to analyze the risk factors of fatty liver occurrence after endocrine therapy. Based on the risk factors, R 3.3.2 software was used to establish a nomogram prediction model. The Harrell consistency index and receiver operating characteristic (ROC) curve (with imageological diagnosis as the "gold standard") were used to analyze the effect of the model on predicting the occurrence of fatty liver, and the calibration curve was used to evaluate the consistency between the model prediction and the actual situation.Results:The recurrence and metastasis rate and mortality rate of patients with endocrine therapy were lower than those of patients without endocrine therapy, and the 3-year and 5-year disease-free survival rates and overall survival rates were higher than those of patients without endocrine therapy (all P < 0.05). Compared with the non-fatty liver group, the levels of alanine transaminase (ALT), aspartate transaminase (AST), total bilirubin (TBIL), total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C) and body mass index (BMI) in the fatty liver group increased (all P < 0.05), while the level of high-density lipoprotein cholesterol (HDL-C) decreased ( P < 0.05). Multivariate logistic regression analysis showed that increased ALT [ OR = 4.680 (95% CI 3.621-5.738)], AST [ OR = 4.862 (95% CI 3.809-5.914)], TBIL [ OR = 3.808 (95 % CI 2.754-4.861)], TC [ OR = 4.294 (95% CI 3.320-5.267)], TG [ OR = 3.401 (95% CI 2.442-4.359)], LDL-C [ OR = 2.976 (95% CI 2.037-3.916)], BMI [ OR = 4.082 (95% CI 3.118-5.045)] and decreased HDL-C [ OR = 0.930 (95% CI 0.876-0.983)] were independent risk factors for fatty liver occurrence after endocrine therapy (all P < 0.05). The consistency index of the nomogram model was 0.792 (95% CI 0.721-0.863), and the area under the ROC curve (AUC) of the nomogram model to judge the occurrence of fatty liver was 0.810 (95% CI 0.734-0.886), indicating that the model had a good discrimination between fatty liver and non-fatty liver. The evaluation of calibration curve showed that the nomogram model for prediction of fatty liver had a good consistency with the actual occurrence of fatty liver. Conclusions:Increased ALT, AST, TBIL, TC, TG, LDL-C, BMI and decreased HDL-C are risk factors for fatty liver occurrence after endocrine therapy in postoperative breast cancer patients. The nomogram model based on risk factors has a good effect on predicting the occurrence of fatty liver.