A Coarse-to-fine Framework for The 2021 Kidney and Kidney Tumor Segmentation Challenge

Zhao Zhong Chen

Keywords: Automatic kidney segmentation \and Kidney cancer \and Coarse-to-fine framework.

Abstract: Kidney cancer is one of the most common malignant tumors in the world. Automatic segmentation of kidney, kidney tumor, and kidney cyst is a promising tool for kidney cancer surgery. In this paper, we use a coarse-to-fine framework which is based on the nnU-Net, and realize accurate and fast segmentation of the kidney and its masses. The average Dice of segmentation predicted by our algorithm is 0.9099, and the Surface Dice is 0.8348.