Improving performance for classifying ovarian cancer with menopause information

Tumor biomarker testing is a relatively simple test using blood, which allows low-cost cancer screening. In previous studies, a diagnostic model of ovarian cancer was made with a combination of two or three optimal biomarkers. This study proposes a classification model of ovarian cancer screening by adding menopausal information, which is important clinical information for diagnosis of ovarian cancer, to biomarker combination. The classification model was evaluated by the area under the ROC curve (AUC). The performance of the classification model including menopausal clinical information showed better performance than the classification model which did not include clinical information.