Purpose Gastric cancer is often detected in the late stages, due to which its mortality rate remains high. Early detection of gastric cancer could significantly improve the prognosis of patients since the survival rate of early gastric cancer after treatment exceeds 96%. This study aimed to analyze early gastric cancer (EGC) risk factors and construct a nomogram model to predict EGC patients. Methods A retrospective study was conducted on 589 patients, including 325 patients with EGC and 264 patients with benign gastric disease. Age, sex, neutrophil to lymphocyte ratio (NLR), creatinine, hypertension, diabetes and other clinical data were collected accordingly. A nomogram was then constructed using univariate analysis and multivariate analysis. Moreover, a correction curve and AUCs were utilized to determine the accuracy of our model. Results Our findings revealed that sex, age, NLR, creatinine, basophil, hypertension and diabetes were risk factors for EGC. A predictive nomogram model was constructed based on the above risk factors showing good consistency and accuracy (AUC = 0.77), the validation cohort showed good consistency(AUC = 0.776). Conclusion The nomogram model presented good reliability, and it will help clinicians to predict and diagnose EGC patients timely while avoiding unnecessary gastrectomy.