Deep Learning Based Cancer Survival Analysis Algorithm

Abstract Histopathological image examination is the gold-standard tool of cancer diagnosis. However, it subjective to only rely on doctor qualitative analysis pathological image. What’s more, easy miss information. The texture images complex and highly heterogeneous, so difficult extract deep features histopathological images. In this paper, a learning-based survival algorithm proposed process data, which can effectively improve accuracy model. average C-Index method 0.7233, improved by 28.0% 19.9% compared with two state-of-the-art methods.