This study aims to leverage machine learning (ML) approaches to screen high-risk groups for ovarian cancer (OC) and enhance preventive strategies. Using data from 1516 suspicious OC cases across six clinical settings in Sari City from 2015 to 2019, six ML algorithms—XG-Boost, Random Forest (RF), J-48, support vector machine (SVM), K-nearest neighbor (KNN), and artificial neural network (ANN)—were employed to construct prediction models. The XG-Boost model, with an area under the receiver characteristic operator curve (AU-ROC) of 0.93 (0.95 CI [0.91–0.95]), was identified as the most effective model for predicting OC. Key influencing factors included family history of cancer, menopausal age, history of chest X-ray, personal history of breast cancer, and postmenopausal hormone therapy. The model's generalizability was validated using external data from two clinical settings, achieving AU-ROC values of 0.85 and 0.89. The study concludes that ML approaches can significantly improve the efficiency and interoperability of OC screening and preventive strategies.This study aims to leverage machine learning (ML) approaches to screen high-risk groups for ovarian cancer (OC) and enhance preventive strategies. Using data from 1516 suspicious OC cases across six clinical settings in Sari City from 2015 to 2019, six ML algorithms—XG-Boost, Random Forest (RF), J-48, support vector machine (SVM), K-nearest neighbor (KNN), and artificial neural network (ANN)—were employed to construct prediction models. The XG-Boost model, with an area under the receiver characteristic operator curve (AU-ROC) of 0.93 (0.95 CI [0.91–0.95]), was identified as the most effective model for predicting OC. Key influencing factors included family history of cancer, menopausal age, history of chest X-ray, personal history of breast cancer, and postmenopausal hormone therapy. The model's generalizability was validated using external data from two clinical settings, achieving AU-ROC values of 0.85 and 0.89. The study concludes that ML approaches can significantly improve the efficiency and interoperability of OC screening and preventive strategies.