Objective: To evaluate the accuracy of the 5 sonographic morphology scoring (SMS) systems (Sassone, DePriest, Lerner, Vera and Kawai and Valentin) for prediction of malignant ovarian tumors. Methods: A diagnostic study was conducted at Songklanagarind Hospital during November 2008 to June 2009. All of the patients scheduled for elective surgery due to ovarian tumors underwent transabdominal or transvaginal sonography within 72 h before the operation. The pictures were recorded. Attention was given to volume, wall and septal thickness, locularity, echogenicity, and papillary and internal surface of the tumor. The 5 SMS systems were applied later by the first author, who was not aware of the clinical data. The final diagnosis was determined by a histopathological report and was categorized into benign or malignant ovarian tumor. Borderline tumors were included in the malignant group. Results: One hundred and forty-six patients were recruited; 82 benign (56.2%), 14 borderline (9.6%), and 50 malignant tumors (34.2%). The sensitivities of the SMS by Sassone, DePriest, Lerner, Vera and Kawai and Valentin were 75, 89.1, 82.8, 79.7 and 82.8% and the specificities were 79.3, 73.2, 68.3, 82.9 and 85.4%, respectively. Conclusions: Among the 5 systems, the DePriest system is the most sensitive SMS for the prediction of ovarian cancer.