ObjectiveTo summarize the application experience of the "Three-Separation" model in the pre-service training of laboratory animal practitioners and scientifically evaluate its training effectiveness. MethodsA retrospective analysis was conducted on 8 009 practitioners who participated in pre-service training at the Fujian Provincial Laboratory Animal Training Base from January 2010 to December 2025. Among them, 1 250 trainees from January 2010 to December 2017 were included in the control group (traditional training model), and 6 759 trainees from January 2018 to December 2025 were included in the evaluation group ("Three-Separation" model, based on separating job requirements, differentiating exam content, and categorizing exam methods). Meanwhile, a questionnaire survey was conducted to evaluate training effectiveness, yielding 1 624 valid questionnaires (585 from the control group and 1 039 from the evaluation group). The "Three-Separation" model simultaneously optimized fee structures to improve affordability and inclusiveness. The comprehensive competencies and training satisfaction of the two groups after training were compared and analyzed using the independent sample t-test, χ² test, and Wilcoxon rank-sum test. ResultsThe overall satisfaction rate of the evaluation group was 92.49% (961/1 039), significantly higher than that of the control group (77.78%, 455/585), with a highly significant difference (χ²=72.50, P<0.001). The evaluation group showed a more favorable distribution of positive, neutral, and negative ratings in dimensions such as course pertinence, time flexibility, assessment rationality, and reasonableness of fees, and the positive rating rates were highly significantly higher than those in the control group (P<0.001). The comprehensive pass rate of the evaluation group (95.53%) was significantly higher than that of the control group (86.24%), with a highly significant difference (P<0.001). ConclusionThe scientific application of the "Three-Separation" model can accurately match the competency requirements of different positions, optimize the training and assessment system, reduce participation barriers by making training fees more affordable and accessible, effectively improve the comprehensive quality and training satisfaction of laboratory animal practitioners, achieve better teaching outcomes, and provide replicable and promotable practical references for the standardized upgrading of the training system in the laboratory animal industry and other industries.

Objective:To construct and validate a predictive model and application program for stone recurrence after endoscopic retrograde cholangiopancreatography (ERCP) based on machine learning algorithms in patients with common bile duct stones (CBDS).Methods:A multicenter retrospective cohort study was conducted, 862 CBDS patients underwent ERCP from June 2020 to September 2023 in Changshu First People′s Hospital (data set 1, 759 cases, including a training set of 588 cases and a validation set of 171 cases) and Changshu Hospital of Traditional Chinese Medicine (data set 2, 103 cases, used as a test set). The demographics, medical history, ERCP procedural records and laboratory indices were collected. All patients were followed up for 1 year, and the stone recurrence was recorded. In training set, the feature selection was conducted by the least absolute shrinkage and selection operator (LASSO) algorithm, and a conventional Logistic regression model was constructed based on selected features. The 3 machine learning algorithms (gradient boosting machine model, extreme gradient boosting model and random forest model) and a conventional Logistic regression model (LASSO model) were trained to fit predictive models. The model performance was assessed by area under curve (AUC) of receiver operating characteristic curve. The model interpretability was analyzed by feature importance evaluation, Shapley additive explanations (SHAP) and force plots. The best-performing model was deployed as an online application by Streamlit framework (V1.36.0).Results:Among the 862 patients, 158 patients (18.33%) developed stone recurrence after ERCP. There were no statistical difference in demographics, medical history, ERCP procedural records and laboratory indices between training set and a validation set ( P>0.05). LASSO regression analysis result showed that 6 key variables (in descending order of significance: endoscopic sphincterotomy, common bile duct angulation, stone diameter, stone count, common bile duct diameter, and periampullary diverticulum) influencing stone recurrence. ROC curve analysis result showed that the random forest model exhibited the highest predictive performance (it had the largest AUC of 0.900). SHAP analysis result showed that common bile duct angulation, common bile duct diameter, stone diameter, endoscopic sphincterotomy and stone count were the top 5 contributing factors in the random forest model. Using Python, the random forest model was implemented into a Streamlit-based application with a user-friendly visual interface, providing predictive outcomes, confidence levels, SHAP force diagram and health recommendations. In the test set, the application program achieved an accuracy of 84.5% (87/103), sensitivity of 82.6% (19/23), and specificity of 85.0% (68/80). SHAP plots and force diagram intuitively illustrated the impact of key features on stone recurrence prediction, offering a clear visualization of each variable′s role within the model. Conclusions:The predictive model and application program based on the random forest machine learning algorithms demonstrate excellent predictive performance and practical usability in predicting stone recurrence after ERCP in patients with CBDS.

ObjectiveTo develop a predictive model and application for spontaneous passage of common bile duct stones using automated machine learning algorithms given the complexity of treatment decision-making for patients with common bile duct stones， and to reduce unnecessary endoscopic retrograde cholangiopancreatography （ERCP） procedures. MethodsA retrospective analysis was performed for the data of 835 patients who were scheduled for ERCP after a confirmed diagnosis of common bile duct stones based on imaging techniques in Changshu First People’s Hospital （dataset 1） and Changshu Traditional Chinese Medicine Hospital （dataset 2）. The dataset 1 was used for the training and internal validation of the machine learning model and the development of an application， and the dataset 2 was used for external testing. A total of 22 potential predictive variables were included for the establishment and internal validation of the LASSO regression model and various automated machine learning models. The area under the receiver operating characteristic curve （AUC）， sensitivity， specificity， and accuracy were used to assess the performance of models and identify the best model. Feature importance plots， force plots， and SHAP plots were used to interpret the model. The Python Dash library and the best model were used to develop a web application， and external testing was conducted using the dataset 2. The Kolmogorov-Smirnov test was used to examine whether the data were normally distributed， and the Mann-Whitney U test was used for comparison between two groups， while the chi-square test or the Fisher’s exact test was used for comparison of categorical data between groups. ResultsAmong the 835 patients included in the study， 152 （18.20%） experienced spontaneous stone passage. The LASSO model achieved an AUC of 0.875 in the training set （n=588） and 0.864 in the validation set （n=171）， and the top five predictive factors in terms of importance were solitary common bile duct stones， non-dilated common bile duct， diameter of common bile duct stones， a reduction in serum alkaline phosphatase （ALP）， and a reduction in gamma-glutamyl transpeptidase （GGT）. A total of 55 models were established using automated machine learning， among which the gradient boosting machine （GBM） model had the best performance， with an AUC of 0.891 （95% confidence interval： 0.859‍ ‍—‍ ‍0.927）， outperforming the extreme randomized tree mode， the deep learning model， the generalized linear model， and the distributed random forest model. The GBM model had an accuracy of 0.855， a sensitivity of 0.846， and a specificity of 0.857 in the test set （n=76）. The variable importance analysis showed that five factors had important influence on the prediction of spontaneous stone passage， i.e.， were solitary common bile duct stones， non-dilated common bile duct， a stone diameter of <8 mm， a reduction in serum ALP， and a reduction in GGT. The SHAP analysis of the GBM model showed a significant increase in the probability of spontaneous stone passage in patients with solitary common bile duct stones， non-dilated common bile duct， a stone diameter of <8 mm， and a reduction in serum ALP or GGT. ConclusionThe GBM model and application developed using automated machine learning algorithms exhibit excellent predictive performance and user-friendliness in predicting spontaneous stone passage in patients with common bile duct stones. This application can help avoid unnecessary ERCP procedures， thereby reducing surgical risks and healthcare costs.

Objective:To construct and validate a predictive model and application program for stone recurrence after endoscopic retrograde cholangiopancreatography (ERCP) based on machine learning algorithms in patients with common bile duct stones (CBDS).Methods:A multicenter retrospective cohort study was conducted, 862 CBDS patients underwent ERCP from June 2020 to September 2023 in Changshu First People′s Hospital (data set 1, 759 cases, including a training set of 588 cases and a validation set of 171 cases) and Changshu Hospital of Traditional Chinese Medicine (data set 2, 103 cases, used as a test set). The demographics, medical history, ERCP procedural records and laboratory indices were collected. All patients were followed up for 1 year, and the stone recurrence was recorded. In training set, the feature selection was conducted by the least absolute shrinkage and selection operator (LASSO) algorithm, and a conventional Logistic regression model was constructed based on selected features. The 3 machine learning algorithms (gradient boosting machine model, extreme gradient boosting model and random forest model) and a conventional Logistic regression model (LASSO model) were trained to fit predictive models. The model performance was assessed by area under curve (AUC) of receiver operating characteristic curve. The model interpretability was analyzed by feature importance evaluation, Shapley additive explanations (SHAP) and force plots. The best-performing model was deployed as an online application by Streamlit framework (V1.36.0).Results:Among the 862 patients, 158 patients (18.33%) developed stone recurrence after ERCP. There were no statistical difference in demographics, medical history, ERCP procedural records and laboratory indices between training set and a validation set ( P>0.05). LASSO regression analysis result showed that 6 key variables (in descending order of significance: endoscopic sphincterotomy, common bile duct angulation, stone diameter, stone count, common bile duct diameter, and periampullary diverticulum) influencing stone recurrence. ROC curve analysis result showed that the random forest model exhibited the highest predictive performance (it had the largest AUC of 0.900). SHAP analysis result showed that common bile duct angulation, common bile duct diameter, stone diameter, endoscopic sphincterotomy and stone count were the top 5 contributing factors in the random forest model. Using Python, the random forest model was implemented into a Streamlit-based application with a user-friendly visual interface, providing predictive outcomes, confidence levels, SHAP force diagram and health recommendations. In the test set, the application program achieved an accuracy of 84.5% (87/103), sensitivity of 82.6% (19/23), and specificity of 85.0% (68/80). SHAP plots and force diagram intuitively illustrated the impact of key features on stone recurrence prediction, offering a clear visualization of each variable′s role within the model. Conclusions:The predictive model and application program based on the random forest machine learning algorithms demonstrate excellent predictive performance and practical usability in predicting stone recurrence after ERCP in patients with CBDS.

Objective To explore the risk factors for postoperative bleeding after endoscopic resection of gastric gastrointestinal stromal tumor(gGIST)and to construct prediction models using 4 different machine learning algorithms for accurately predicting postoperative bleeding.Methods The clinical data of gGIST patients were collected,and the patients were randomly divided into a training cohort(n=502)and a validation cohort(n=130)at an 8:2 ratio.Synthetic minority over-sampling technique-nominal continuous was used for oversampling in the training cohort.Four prediction models were constructed using gradient boost machine(GBM),deep learning,generalized linear model and distributed random forest,separately;and in addition,the least absolute shrinkage and selection operator was used to screen variables and construct a traditional Logistic regression model.Model performance was evaluated by calculating the area under the receiver operating characteristic curve(AUC),sensitivity,specificity,accuracy,positive predictive value and negative predictive value.Interpretability analyses,including feature importance,SHapley additive exPlanation and force plot,were performed on the optimal model,and a practically applicable web application was developed.Results Among 632 patients,78(12.3%)experienced postoperative bleeding.In the validation cohort,GBM model performed best among 5 prediction models,with an AUC value of 0.889 and a 95%CI of 0.829-0.948,superior to the other 4 models.Variable importance analysis identified surgeon experience,operation time,intraoperative hemorrhage,tumor size as the factors affecting postoperative bleeding prediction.The SHapley additive exPlanation plot and force plot showed the distribution characteristics of variables in the binary classification prediction results and the effect of each variable on the prediction results.Conclusion GBM model has high predictive value for postoperative bleeding following endoscopic resection of gGIST,and the construction of the web application facilitates its clinical use.

Objective To explore the risk factors for postoperative bleeding after endoscopic resection of gastric gastrointestinal stromal tumor(gGIST)and to construct prediction models using 4 different machine learning algorithms for accurately predicting postoperative bleeding.Methods The clinical data of gGIST patients were collected,and the patients were randomly divided into a training cohort(n=502)and a validation cohort(n=130)at an 8:2 ratio.Synthetic minority over-sampling technique-nominal continuous was used for oversampling in the training cohort.Four prediction models were constructed using gradient boost machine(GBM),deep learning,generalized linear model and distributed random forest,separately;and in addition,the least absolute shrinkage and selection operator was used to screen variables and construct a traditional Logistic regression model.Model performance was evaluated by calculating the area under the receiver operating characteristic curve(AUC),sensitivity,specificity,accuracy,positive predictive value and negative predictive value.Interpretability analyses,including feature importance,SHapley additive exPlanation and force plot,were performed on the optimal model,and a practically applicable web application was developed.Results Among 632 patients,78(12.3%)experienced postoperative bleeding.In the validation cohort,GBM model performed best among 5 prediction models,with an AUC value of 0.889 and a 95%CI of 0.829-0.948,superior to the other 4 models.Variable importance analysis identified surgeon experience,operation time,intraoperative hemorrhage,tumor size as the factors affecting postoperative bleeding prediction.The SHapley additive exPlanation plot and force plot showed the distribution characteristics of variables in the binary classification prediction results and the effect of each variable on the prediction results.Conclusion GBM model has high predictive value for postoperative bleeding following endoscopic resection of gGIST,and the construction of the web application facilitates its clinical use.

【Objective】 To explore the clinicopathological characteristics and comprehensive treatment strategies of prostate mucosa adenocarcinoma under multidisciplinary diagnosis and treatment (MDT) mode. 【Methods】 Data of two patients with typical prostate mucosa adenocarcinoma treated in our hospital during Sep.2020 and Apr.2023 were retrospectively analyzed. 【Results】 In case 1, the clinical manifestation was macroscopic hematuria; multiparametric magnetic resonance imaging (mpMRI) indicated solid prostatic nodules, clinical stage T4N1Mx; initial prostate specific antigen (PSA) was 1.2 ng/mL; ⁶⁸68Ga-prostate specific membrane antigen PET/CT (⁶⁸Ga-PSMA PET/CT) suggested abnormal uptake of nuclear lesions in the prostate (SUV4.2-5.3); biopsy results indicated invasive mucinous adenocarcinoma.After prostate and pelvic field radiotherapy + androgen deprivation therapy (ADT) + antihypertensive treatment, lesions were significantly reduced, and hematuria symptoms were relieved.In case 2, the clinical manifestation was dysuria; initial PSA was 91.78 ng/mL; mpMRI suggested invasion of prostate mass into the bladder and clinical stage of T4N1M1b; ⁶⁸Ga-PSMA PET/CT indicated prostate and pelvic lymph nodes, and multiple bone lesions showed increased nuclide uptake; biopsy results indicated prostate adenocarcinoma with mucinous adenocarcinoma.Initial endocrine treatment was performed.After 3 months, PSA was reduced to 0.083 ng/mL, and imaging showed the tumor was significantly reduced.Robotic-assisted laparoscopic tumor prostatectomy with extended pelvic lymph node dissection was performed, and endocrine adjuvant therapy was continued after surgery. 【Conclusion】 Prostate mucosa adenocarcinoma has different clinicopathological characteristics and prognosis from conventional acinar adenocarcinoma, and the whole-process management under MDT mode is of great value in the diagnosis and treatment of this disease.

Objective:To evaluate the changes of fetal renal artery blood flow parameters in fetuses with isolated borderline oligohydramnios (IBO) in the middle and third trimesters by Doppler ultrasound, and to assess its correlations with maternal and infant pregnancy outcomes.Methods:Twenty-seven IBO fetuses (IBO group) and 27 gestational age-matched normal fetuses (control group) from April to October 2019 in the Second Xiangya Hospital of Central South University underwent prenatal ultrasound examination during the middle and third trimesters. Renal artery blood flow parameters, including renal artery pulsatility index (RAPI), volume corrected renal artery pulsatility index (vcRAPI) and pregnancy outcomes were measured and compared between the two groups. Once diagnosed IBO, patients were recommended to the obstetric clinic for consultation and intervention. The correlation between RAPI, vcRAPI measured before intervention and prepartum amniotic fluid volume and pregnancy outcomes was analyzed, the ROC curve was plotted to find the better predictor.Results:The vcRAPI of the IBO group was higher than that of the control group ( P=0.015). In the IBO group, the vcRAPI measured before intervention was higer in those fetuses who were still IBO before delivery( P=0.048). In the IBO group, the correlation of the vcRAPI measured before intervention and IBO before delivery was statistically significant ( OR=2.41, 95% CI=1.06-5.43, P=0.035). The ROC curve showed that the sensitivity of vcRAPI to IBO was 0.67, the specificity was 0.75( P=0.002). Conclusions:Compared with RAPI, The vcRAPI may reflect the increase in fetal renal artery perfusion resistance of IBO group more timely. The higher vcRAPI before intervention in the IBO group have difficulty in recovering amniotic fluid volume before delivery.Increased vcRAPI is a better predictor of IBO before delivery.

Objective:To evaluate the clinical efficacy of Jianpi-Huazhuo Decoction in the treatment of complication patients with phlegm-dampness in type 2 diabetes mellitus (T2DM) and Diabetic nephropathies (DN). Methods:A total of 72 patients with with phlegm dampness T2DM and DN in Huaibei Traditional Chinese Medicine Hospital of Anhui Province from June 2018 to June 2020 were randomly divided into two groups with 36 in each group. The control group were treated with oral metformin sustained-release tablets on the basis of diabetes propaganda. The observation group was treated with Jianpi-Huazhuo Decoction on the basis of the control group. Both groups were treated for 4 weeks. The blood glucose (plasma, enzyme method), HbA1c (whole blood, high performance liquid chromatography) and fasting insulin (serum, chemiluminescence method) were measured, and homeostasis model assessment of insulin resistance (HOMA-IR) was calculated. Plasma BUN, SCR and urinary albumin excretion rate (UAER) were measured by automatic biochemical analyzer. The plasma laminin (LN), procollagen Ⅲ (PC Ⅲ) and collagen type Ⅳ (Ⅳ-c) were detected by ELISA. The adverse events during treatment were observed and the clinical efficacy was evaluated. Results:The total effective rate was 86.1% (31/36) in the observation group and 58.3% (21/36) in the control group ( χ2 =6.923, P=0.009). After treatment, the levels of FBG, 2 hPG, HbA1c, FINS and HOMA-IR in the observation group were significanlty lower than those in the control group ( t values were 4.242, 2.751, 3.565, 3.613 and 4.512, respectively, all Ps<0.05). After treatment, the plasma levels of LN, PC Ⅲ and Ⅳ-c were significanlty lower than those in the control group ( t values were 3.612, 1.864 and 2.046, respectively, all Ps<0.05). After treatment, the levels of serum creatinine and urinary albumin excretion rate in the control group were significanlty lower than those in the control group ( t values were 5.864 and 3.286, respectively, all Ps<0.05). Conclusion:The Jianpi-Huazhuo Decoction can reduce the blood glucose level and renal fibrosis related factors in patients with phlegm dampness T2DM complicated with DN, improve the clinical symptoms and improve the clinical curative effect.

Objective To discuss the dosimetric differences in the planning methods between physical and biological optimization during thehypofractionated radiotherapy for lung cancer.MethodsTen cases of non-small cell lung cancer (NSCLC) receiving radiotherapy were selected in this retrospective study.The VMAT plans for all patients were re-designed by physical functions (DV group),biological combined with physical functions (DV+EUD group and EUD+DV group) and biological functions (EUD group).The constrained functions were different,whereas the constrained conditions and optimized parameters were identical among four groups.The dosimetric differences among four optimization methods during thehypofractionated radiotherapy for lung cancer were evaluated through calculating and analyzing each dosimetry parameter.Results For the target area,the equivalent uniform dose was approximate between the EUD and EUD+DV groups.The EUD in these two groups was approximately 2.8％-3.6％ and 3.2％-3.7％ higher than those in the DV and DV+EUD groups.The average tumor control probability (TCP) in the EUD and EUD +DV groupswas considerably higher than those in the other two groups (both P＜0.05).The homogeneity index (HI) significantly differed (all P＜0.05),whereas the conformity index (CI) did not differ (all P＞0.05).For the organ at risk (OAR) area,the differences of EUD,V5,V1o,V20,V30 of normal lung tissues and the difference of dosimetry parameters in heart and spinal cord were not statistically significant (all P＞0.05).The mean dose of all lungs in the EUD and EUD+DV groupswas slightly lower than those in the other two groups.ConclusionsBiological optimization method has certain advantages in increasing EUD and TCP in the target area and decreasing the irradiation dose of normal lung tissues,which provides references for selecting the optimization method with biological functions in clinical practice.