ABSTRACT
OBJECTIVE@#To establish a machine learning model based on extreme gradient boosting (XGBoost) algorithm for early prediction of severe acute pancreatitis (SAP), and explore its predictive efficiency.@*METHODS@#A retrospective cohort study was conducted. The patients with acute pancreatitis (AP) who admitted to the First Affiliated Hospital of Soochow University, the Second Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University from January 1, 2020 to December 31, 2021 were enrolled. Demography information, etiology, past history, and clinical indicators and imaging data within 48 hours of admission were collected according to the medical record system and image system, and the modified CT severity index (MCTSI), Ranson score, bedside index for severity in acute pancreatitis (BISAP) and acute pancreatitis risk score (SABP) were calculated. The data sets of the First Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University were randomly divided into training set and validation set according to 8 : 2. Based on XGBoost algorithm, the SAP prediction model was constructed on the basis of hyperparameter adjustment by 5-fold cross validation and loss function. The data set of the Second Affiliated Hospital of Soochow University was served as independent test set. The predictive efficacy of the XGBoost model was evaluated by drawing the receiver operator characteristic curve (ROC curve), and compared it with the traditional AP related severity score; variable importance ranking diagram and Shapley additive explanation (SHAP) diagram were drawn to visually explain the model.@*RESULTS@#A total of 1 183 AP patients were enrolled finally, of which 129 (10.9%) developed SAP. Among the patients from the First Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University, there were 786 patients in the training set and 197 in the validation set; 200 patients from the Second Affiliated Hospital of Soochow University were used as the test set. Analysis of all three datasets showed that patients who advanced to SAP exhibited pathological manifestation such as abnormal respiratory function, coagulation function, liver and kidney function, and lipid metabolism. Based on the XGBoost algorithm, an SAP prediction model was constructed, and ROC curve analysis showed that the accuracy for prediction of SAP reached 0.830, the area under the ROC curve (AUC) was 0.927, which was significantly improved compared with the traditional scoring systems including MCTSI, Ranson, BISAP and SABP, the accuracy was 0.610, 0.690, 0.763, 0.625, and the AUC was 0.689, 0.631, 0.875, and 0.770, respectively. The feature importance analysis based on the XGBoost model showed that the top ten items ranked by the importance of model features were admission pleural effusion (0.119), albumin (Alb, 0.049), triglycerides (TG, 0.036), Ca2+ (0.034), prothrombin time (PT, 0.031), systemic inflammatory response syndrome (SIRS, 0.031), C-reactive protein (CRP, 0.031), platelet count (PLT, 0.030), lactate dehydrogenase (LDH, 0.029), and alkaline phosphatase (ALP, 0.028). The above indicators were of great significance for the XGBoost model to predict SAP. The SHAP contribution analysis based on the XGBoost model showed that the risk of SAP increased significantly when patients had pleural effusion and decreased Alb.@*CONCLUSIONS@#A SAP prediction scoring system was established based on the machine automatic learning XGBoost algorithm, which can predict the SAP risk of patients within 48 hours of admission with good accuracy.
Subject(s)
Humans , Pancreatitis , Acute Disease , Retrospective Studies , Hospitalization , AlgorithmsABSTRACT
Acute pancreatitis （AP） is a gastrointestinal disease that requires early intervention， and when it progresses to moderate-severe AP （MSAP） or severe AP （SAP）， there will be a significant increase in the mortality rate of patients. Machine learning （ML） has achieved great success in the early prediction of AP using clinical data with the help of its powerful computational and learning capabilities. This article reviews the research advances in ML in predicting the severity， complications， and death of AP， so as to provide a theoretical basis and new insights for clinical diagnosis and treatment of AP through artificial intelligence.
ABSTRACT
Objective:To evaluate narrow band imaging-magnifying endoscopy (NBI-ME) for the further assessment of lesions of low-grade intraepithelial neoplasia (LGIN) in the gastric biopsy.Methods:Data of 180 patients who underwent NBI-ME before endoscopic submucosal dissection (ESD) for biopsy of gastric LGIN at the First Affiliated Hospital of Soochow University from January 2017 to October 2020 were analyzed retrospectively. Taking the pathological results after ESD as the gold standard, the sensitivity, the specificity, the positive predictive value, the negative predictive value, and the accuracy of NBI-ME in predicting the pathological upgrading of gastric LGIN lesions after ESD were calculated, and the receiver operator characteristic (ROC) curve was drawn.Results:Among 180 gastric LGIN lesions, 115 (63.89%) were pathological upgraded and 65 (36.11%) were not after ESD. There were 10 missed diagnoses, 19 misdiagnoses, and 151 correct diagnoses in NBI-ME examination before ESD. The sensitivity, the specificity, the positive predictive value, the negative predictive value, and the accuracy of NBI-ME in predicting the pathological upgrading of gastric LGIN lesions after ESD were 91.3% (105/115), 70.8% (46/65), 84.7% (105/124), 82.1%(46/56) and 83.9% (151/180), respectively. The area under the ROC curve was 0.810 (95% CI: 0.737-0.883). Conclusion:Further NBI-ME examination of gastric LGIN lesions diagnosed by biopsy pathology can accurately predict whether the lesions have pathological upgrading after ESD, which is of important guiding significance for the patients to choose the treatment strategy of further follow-up or endoscopic resection.
ABSTRACT
Objective By using optical coherence tomography (OCT) to evaluate the vascular neointimal hyperplasia and the stent strut coverage degree in patients with acute myocardial infarction (AMI) and in patients with stable angina (SA) one year after receiving drug-eluting stent (DES) implantation, and to compare the clinical results between the two groups. Methods A total of 39 patients, who received DES implantation due to coronary heart disease, including AMI (n=16, AMI group) and SA (n=23, SA group), during the period from March 2011 to July 2012, were enrolled in this study. One year after DES implantation, coronary angiography and OCT reexaminations were performed in all patients. The neointimal hyperplasia (NIH) thickness, NIH area, NIH volume, strut coverage and apposition rate were determined with OCT. The results were compared between the two groups. Results OCT measuring results showed that the mean NIH thickness of AMI group and SA group was ( 66 . 8 ± 20 . 7 ) mm and ( 121 . 6 ± 135 . 7 ) mm respectively (P=0.022); the NIH volume ratio were 5.66%±3.18% and 11.88%±8.22% respectively (P=0.005); the percentage of cross-section with NIH thickness over 100 μm was 22.56%±23.99% and 40.14%± 30.01% respectively (P=0.034); and the percentage of overall stent strut coverage was 89.27%±6.40% and 93.42%±7.03% respectively (P=0.007). All the above mentioned data of AMI group were obviously lower than those of SA group. Conclusion After DES implantation, the intimal repair, intimal hyperplasia and stent strut coverage in AMI patients are poorer.
ABSTRACT
Objective To investigate the possible inhibition effect of a domestic novel drug-eluting balloon(DEB), such as controlling the vascular stenosis and reducing the intimal hyperplasia, on obstructive peripheral arterial disease (PAD) of lower extremity. Methods Male New Zealand rabbits (weighted 2.5-3.0 kg) were used for this study. Mustang bare metal stents (BMS) were separately implanted at both the proximal and distal site of abdominal aorta to establish the PAD models in rabbits. Bare balloon (PTA group) or drug-eluting balloon (DEB group) was separately used to dilate the stent segment of the aorta. The animals were sacrificed 28 days after the treatment. The aorta stent specimens were collected and sent for pathologic examination. Using Masson staining method the degree of intimal hyperplasia was estimated and the expression level of paclitaxel-targeted β-tubulin was determined with immunohistochemical method. The results were compared between the two groups. Results Morphologically, both the degree of intimal hyperplasia and the vascular stenosis rate in DEB group were significantly lower than those in PTA group. Meanwhile, immunohistochemical examination of paclitaxel-targeted β-tubulin indicated that its expression level in DEB group was obviously higher than that in PTA group. Conclusion This experimental study indicates that the domestic novel DEB has potential inhibition effect in respect of controlling the loss of lumen as well as reducing the intimal hyperplasia. This novel DEB carries excellent property, suggesting the possibility to use DEB in clinical treatment of PAD.
ABSTRACT
Objective To study the expression of receptor for advanced glycation end products ( RAGE ) in chronic adriamycin ( ADR )-induced cardiotoxicity mice . Method A mouse model of chronic cardiotoxicity was established by intraperitoneal injection of 5 mg/kg ADR once a week for continuous 3 weeks.Mice treated with the same volume of saline were used as control group (6 mice in each group).Echocardiography and histopathological examination were used to evaluate the myocardial injury , cardiac function and ventricular remodeling .The expression of RAGE in cardiac tissues was detected by immunohistochemical staining and Western blotting .Results Echocardiography indicated that ADR-treated mice had reduced left ventricular ejection fraction , while histopathological analysis of cardiac tissues showed obvious cardiac damage . RAGE protein in cardiac tissues was highly expressed in chronic ADR-induced cardiotoxicity mice .Conclusion RAGE is highly induced and may be involved in the cardiotoxicity after ADR treatment .
ABSTRACT
OBJECTIVE To explore the clinical efficacy of adenoidectomy in treating pediatric secretory otitis media,sinusitis and snoring. METHODS A retrospective study was conducted to analyze the clinical features,therapeutic methods and prognosis of 68 children in hospital who underwent adenoidectomy to treat secretory otitis media,sinusitis or snoring. RESULTS The conditions of the 68 children were marked improved after the removal of hypertrophic adenoids. The total clinical effectiveness rate was 100 % and the cure rate was 86.8 %. CONCLUSION Hypertrophic adenoids is a fundamental cause of pediatric secretory otitis media,sinusitis and snoring. The removal of hypertrophic adenoids is a safe and effective method for treating pediatric secretory otitis media,sinusitis and snoring.