Aim: The study aimed to compare the risk of ovarian malignancy algorithm (ROMA) with risk of malignancy index (RMI), cancer antigen 125 (CA125), human epididymis protein 4 (HE4) and Sassone ultrasonography (US) score in predicting ovarian cancer (OC) in women who present with pelvic or adnexal masses. Methods: Pelvic US, serum CA125 and HE4 levels were investigated preoperatively in consecutively enrolled Thai women over 18 years with clinically diagnosed pelvic or adnexal masses who were undergoing elective surgery at a super tertiary hospital in Thailand in 2012. ROMA, RMI and Sassone US score were calculated. Results: Of 260 women who were evaluated, 74 had OC. Areas under the receiver operating characteristic curve (AUC) in predicting OC were ROMA, 86.2%; RMI, 87.6%; CA125, 80.6%; HE4, 82.4%; and Sassone score, 77.1%. ROMA and RMI performed similarly well (AUCs 84.4 and 85.6%) in premenopausal women but RMI outperformed ROMA in postmenopausal women (AUCs 87.9 and 84.0%, respectively). At the conventional cutoff value, ROMA and CA125 showed the highest sensitivity (83.8%) but HE4 had the highest specificity (86.0%). Conclusions: ROMA was comparable to RMI in predicting OC. Either ROMA or RMI can be applied to women with pelvic masses, stratified into low- and high-risk groups for OCs.