This study developed a nomogram model utilizing dairy cow-level risk factors to predict the risk of Mycobacterium avium subspecies paratuberculosis (MAP) infection. MAP antibody status was detected by ELISA in 1,589 dairy cows on commercial farms in Henan Province, China. Dairy Herd Improvement (DHI) data was also collected for each cow. Univariate analysis was used to identify MAP risk factors and multivariate logistic regression with backward bootstrap screening was used to determine the independent predictor for inclusion in the nomogram model. Model performance was evaluated by area under the receiver operating characteristic curve (AUC), calibration plots, and decision curve analysis. Finally, 1,481 cows with complete data were included, with a 24.9% MAP positive rate (n=369). The nomogram model demonstrated good discrimination (AUC 0.71) and accuracy (70.2%). Calibration was excellent (Hosmer-Lemeshow χ2=3.26, P=0.92), and decision curve analysis indicated this predictive model has clinical utility for diagnostic testing. The nomogram predicted individual MAP risk based on routinely available DHI data including age, milk production, mammary health status, milk losses, and milk fat. Our study provides a method for screening high-risk dairy cows and developing intervention strategies based on DHI reports.