OBJECTIVE To compare the diagnostic accuracy of positron emission tomography/computed tomography (PET/CT), pelvic Doppler ultrasonography (US), abdomino-pelvic computed tomography (CT), and pelvic magnetic resonance imaging (MRI) for detection of ovarian cancer and to assess the role of PET/CT in evaluating the dissemination of ovarian cancer. METHODS One hundred thirty-three women suspected to have ovarian cancer were enrolled in a prospective study before surgery between March 2005 and August 2007. The accuracy of each modality in detection of malignancy was estimated by computing the relevant areas under a receiver operating characteristics curve. Histopathologic results served as the reference standard. RESULTS Histopathology showed benign tumors in 25 patients, borderline tumors in 13 patients, and malignant tumors in 95 patients. In distinguishing malignant/borderline from benign ovarian tumors, the accuracy of PET/CT (0.921) was higher than that of pelvis US (0.830) and abdomino-pelvic CT or pelvis MRI (0.749; P=0.013). Radiologic staging by PET/CT was concordant with surgical staging in 78% of patient and PET/CT revealed 15 (15.8%) unpredicted extra-abdominal lymph node metastasis in 95 patients with ovarian cancer. In addition, PET/CT detected new, unexpected co-existing malignant tumors in five (3.8%) cases including two thyroid tumors, two breast tumors, and one pancreatic neuroendocrine cancer. CONCLUSIONS PET/CT is superior to pelvis US, abdomino-pelvic CT, and pelvic MRI for diagnosis of malignant ovarian tumors and is useful in revealing metastatic ovarian cancer and co-existing malignant tumors. Therefore, we suggest that PET/CT could be used during pre-operative evaluation of patients suspected to have ovarian cancer.