Using artificial intelligence to select patients for primary surgery in advanced stage ovarian cancer

Abstract

Objective: The mainstay of therapy for high-grade serous ovarian cancer (HGSC) is primary cytoreductive surgery with postoperative chemotherapy. However, optimal surgical outcomes with <1 cm of residual disease are achieved in only a portion of surgical patients. Thus, methods that would objectively select patients for optimal surgical outcomes are needed. Our objective was to create models that will identify HGSC patients who will have optimal surgical outcomes using radiologic images analyzed with artificial intelligence (AI).