Establishment and validation of risk prediction model for colorectal adenomatous polyps
Objective To analyze the risk factors of colorectal adenomatous polyps,and build a Nomogram model to predict the risk of colorectal adenomatous polyps individually,and verify the model.Methods The clinical data of pa-tients undergoing colonoscopy from Jan.2020 to Sep.2022 in the Third Affiliated Hospital of Xinxiang Medical Univer-sity were collected retrospectively.According to the ratio of 7∶3,the subjects were randomly divided into training set and validation set.The risk factors of colorectal adenomatous polyps were screened by univariate and multivariate Logistic re-gression analysis,and the nomograph model of colorectal adenomatous polyps risk was constructed using R software.The validation set was used for internal validation of the model.The discrimination,calibration and clinical practicability of the model were evaluated by the ROC curve,Calibration curve and decision curve analysis.Results 513 patients were included in the analysis,including 347 patients in the training set(225 males and 122 females)and 166 patients in the validation set(115 males and 51 females),with the ages of(58.13±10.39)years old and(56.70±11.10)years old,respectively.Regression analysis showed that age,BMI,smoking,fatty liver,and low density lipoprotein(LDL)were associated with colorectal adenomatous polyps.The AUC of the training set and validation set of the model were 0.772(95%CI:0.722-0.822)and 0.745(95%CI:0.667-0.823),respectively.Calibration curve had good fit.Deci-sion curve analysis showed that the nomogram model had good clinical applicability.Conclusion The nomograph model based on age,BMI,smoking,LDL and fatty liver has a good predictive value for the risk of adenomatous polyps in colo-rectal polyps.
NETL
NSTL
万方数据
