The development of new biomarkers for ovarian cancer is clearly necessary for the improved detection and monitoring of the disease. Surface enhanced laser desorption and ionization time-of-flight mass spectrometry (SELDI-TOF-MS) can be employed in the identification of differentially expressed proteins in cancer cells. The objective of this study was, then, to identify potential diagnostic serological biomarkers for ovarian cancer. We performed protein expression difference analyses of 45 serum samples using SELDI protein chip array. Forty-five sera obtained from ovarian cancer patients (n=35) and normal healthy females (n=10), were profiled on the surface of SELDI protein chip. The candidate biomarkers were purified by CM-Sepharose, and their N-terminal amino sequence was determined. The amounts of hemoglobin (Hb) in cancer patient's sera versus that of normal sera were measured by ELISA. Nine sera proteins that were found to be significantly differentially expressed (P<0.05) between the sera of ovarian cancer patients and that of normal healthy females were selected using the WCX2 array. The most distinctive polypeptide peaks detected in the ovarian cancer samples were at 15.1 and 15.8 kDa and these two peaks were identified as the hemoglobin-alpha (Hb-alpha) and -beta (Hb-beta) chain, respectively. ELISA indicated that the sensitivity for intact Hb level was 77% in sera obtained from ovarian cancer patients, as compared with normal healthy female sera. In conclusion, two ovarian cancer biomarker proteins were discovered and identified as Hb-alpha and Hb-beta. Hb levels were significantly different in ovarian cancer serum samples and those obtained from normal healthy females, as determined by ELISA. Additional studies are required to further validate Hb-alpha and Hb-beta biomarkers.