Quantitative assessment of colorectal cancer via conditional generative adversarial networks

Quoc Dang Vu, Kyungeun Kim, Jin Tae Kwak

Abstract: Grading for cancer, based upon the degree of cancer differentiation, plays a major role in describing the characteristics and behavior of the cancer and determining treatment plan for patients. The grade is determined by a sub-jective and qualitative assessment of tissues under microscope, which suffers from high inter- and intra-observer variability among pathologists. Digital pathology offers an alternative means to automate the procedure as well as to improve the accuracy and robustness of cancer grading. However, most of such methods tend to mimic or reproduce cancer grade determined by hu-man experts. Herein, we propose a quantitative means of assessing and char-acterizing cancer via conditional generative adversarial networks. The pro-posed method is evaluated using tissue microarrays (TMA) of colorectal can-cer. The results suggest that the proposed method holds a potential for quan-tifying cancer characteristics and improving cancer pathology.

TL;DR: We propose a cGAN-based method to learn and quantify the characteristics of the tissue that are relevant to tumor differentiation.

Keywords: Colorectal cancer, Tumor grading, differentication, GAN