Construction of a risk prediction model for gastric cancer based on random forest
WANG Qingqing
WU Wenbo
WAN Shaoping
ZHANG Linglin
LI Yuting
RONG Lilou
Abstract：Objective:To construct a risk prediction model for gastric cancer in Sichuan province and to identify its predictive factors.Methods:A total of 378 patients diagnosed with gastric cancer in four medical institutions in Sichuan province from July 2021 to August 2022 were selected as the case group.A total of 378 healthy individuals undergoing physical examinations during the same period were selected as the control group.The random forest method was used to construct a risk prediction model for gastric cancer.The performance of the model was evaluated by using accuracy,sensitivity,specificity,area under the ROC curve(AUC),calibration curve,and decision curve analysis(DCA).Results:Educational level,occupation,white meat intake,processed meat intake,coarse grain intake,pickled and sun dried food intake,dietary taste,dietary coldness and heat,eating speed,breakfast status,smoking status,alcohol consumption,tea drinking status,history of superficial gastritis,atrophic gastritis,gastric ulcer,hypertension,hyperlipidemia,family history of cancer,and personality were independent influencing factors for gastric cancer.The top three influencing factors in terms of importance score were occupation,education level,and smoking status.The accuracy of the prediction model was 99.8%.The sensitivity was 100.0%.The specificity was 99.6%.The AUC was 0.999(95%CI 0.998-1.000).The calibration curve of the prediction model indicated good consistency between the model and the actual observation results.The DCA curve suggested that the model has good clinical efficacy in predicting gastric cancer.Conclusions:The risk prediction model for gastric cancer of Sichuan province constructed based on the random forest method in this study has good predictive performance and is helpful for early identification of high-risk population for gastric cancer.
Keywords：gastric cancerrandom forestrisk prediction modelinfluencing factors
Publication Date：2026-04-10
Online Publishing Date：2026-03-31(First online date of this platform, not the publication date of the document)
Pages：11( 1070-1080 )