[Field trial for the early detection of patients with ovarian cancer--discrimination of ovarian cancer patients by the statistical analysis using Mahalanobis' generalized distance]
Related Experiment Video
Updated: Aug 19, 2026

Murine Model for Non-invasive Imaging to Detect and Monitor Ovarian Cancer Recurrence
Published on: November 2, 2014
H Kobayashi1, K Sumimoto, T Terao
1Department of Obstetrics and Gynecology, Hamamatsu University School of Medicine.
Abstract:
Six tumor-associated antigens, cancer antigen 125 (CA125), tissue polypeptide antigen (TPA), ferritin (Fr), carcinoembryonic antigen (CEA), alpha fetoprotein (AFP), and sialyl Lex-i (SLX) were measured simultaneously for the early detection of ovarian cancer. To decrease the number of both false positive and false negative cases in the combination assay, statistical discrimination analysis employing the serum values for appropriate tumor markers has been studied with respect to ovarian cancer by the method of Mahalanobis' generalized distance. The new "ovarian cancer screening test" designed by us has been used in Shizuoka Prefecture since 1988, and 23,307 serum samples have been analyzed. One hundred twenty-seven of 165 ovarian cancer patients were suspected as having cancer by such clinical procedures as pelvic examination and/or ultrasonography, while in 150 patients cancer was detected by the statistical discrimination method. Thirty-one of 38 patients with ovarian cancer overlooked by the clinical procedures could be found by the statistical method. We conclude that clinical procedures and the statistical method can be complementary in detecting patients with this malignancy.

