Prognosis Prediction of Breast Cancer Based on CGAN

Breast cancer is the malignancy with the highest morbidity and mortality rate in women worldwide, and prognosis prediction of breast cancer is of great practical importance for both patients and clinical practitioners. In this paper, we use a modified Conditional Generative Adversarial Networks (CGAN) to train the generators in a GAN into a predictive model that can perform prognosis of breast cancer on clinical data from patients and compare it with a multi-factor Cox proportional hazards model. In this paper, the accuracy of the prognostic model using CGAN was 0.950 with an AUC of 0.915; the AUC value of prognosis obtained using the multi-factor Cox proportional hazards model was 0.837. The experimental results demonstrated that the breast prognostic model based on CGAN can more accurately quantify and assess the prognosis of patients.