Objective:To compare and validate the diagnostic performance of the Ovarian-Adnexal Reporting and Data System (O-RADS ) and the ADNEX model in the diagnosis of malignant ovarian-adnexal lesions.Methods:A total of 275 patients who underwent surgery for ovarian-adnexal lesions at Tianjin Medical University Cancer Institute and Hospital from December 2020 to December 2022 were retrospectively collected. The clinical, pathological aud ultrasound dates of the patients were collected.Statistical methods, including chi-square tests and ROC curve analysis, were employed to assess the diagnostic performance of O-RADS and the ADNEX model for ovarian-adnexal lesions.Results:Among the 275 patients included in this study, 127 (46.2%) had benign lesions, and 148 (53.8%) had malignant lesions.Based on the O-RADS classification, 46 cases (16.7%) were O-RADS 2, 50 cases (18.2%) were O-RADS 3, 66 cases (24.0%) were O-RADS 4, and 113 cases (41.1%) were O-RADS 5. The malignancy rates for O-RADS 2, O-RADS 3, O-RADS 4, and O-RADS 5 were 0%, 0.08%, 56.06%, and 94.7%, respectively. ROC curve analysis for malignant ovarian-adnexal lesions yielded an area under ROC curve of 0.93(95% CI=0.90-0.96) for O-RADS and 0.94(95% CI=0.91-0.97) for the ADNEX model. Using O-RADS ≥4 and ADNEX model ≥10% as cutoff values, there was no significant difference in sensitivity between the two methods( P=0.740), but O-RADS exhibited higher specificity compared to the ADNEX model (72.4% vs 56.7%, P=0.044). Conclusions:When O-RADS ≥4 and the ADNEX model ≥10% are used as cutoff values, both methods demonstrate excellent diagnostic performance for malignant ovarian-adnexal lesions, with O-RADS exhibiting higher specificity.