Abstract : Objective To construct a model for puredicting the benign and malignant risk of the patients with gastric diseases using con-ventional laloratory indicators such as tumor markers，blood routine，and coagulation indicators，and validate its predictive value.Methods The medival reconls of patients with gastric diseass who visited the First Affiliated Hospital of Guangzhou Medlical Univer-sity from January 2018 to January 2023 were analyzed retrospectively. According to the pathological results,the paients were dividedinto the gastric cancer (CC) group (n= 134) and chronic non-atrophic gastritis (CNAG)group ( n=298). Their routine test data suchas serum and whole blood tests were collected. The statistical analysis was conducted using the R 4.2.3 software， and a model for pre-dicting the risk of CC was constructed and validated. Results A model for predlicting the risk of CC was constructed successfiully usingthe Logistic regression analysis , which included D-dlimer , carcinoembryonic antigen (CEA) , carbohydrate antigen 72-4(CA72-4) andhemoglobin ( Hh). A visual nomogram was ploted as the final predliction model The areas under the receiver operating characteristic(ROC) curve (AUc.%) of the modlel in the taining and testing sets were 0.809 (95%CI;0.754-0.864) and 0.808 (95%CI:0.724-0.892) , respectively.The sensitivity and specificity of the model were 58.5% and 93.3%6 , respectively , indicating that it had good pre-dictive abilit. ConclusionThe model for pmredicting the malignant risk of patients with gastric diseases constructedl using routine te-ting indlicators has good accuracy and can effectively predict the risk of gastric disease transforning into gastric cancer. lt helps to findearly gastric cancer patients in clinical practice and take targeted prevention and intervention measures.