Kidney and Tumor Segmentation

Run the codes from nnU-Net [1].

3d lowres (Part 1)

dataset_split_1: dice similarity coefficients of kidney and kidney tumor segmentation
Dice of Kidney Segmentation: 0.9570647169854021
Dice of Tumor Segmentation: 0.7809852033199037

dataset_split_1: examples from validation set

3d_cascade_fullres (Part 2)

3d_fullres (Single part)

References：

[1] nnU-Net Breaking the Spell on Successful Medical Image Segmentation 2019 [paper] [code]
[2] nnU-Net Self-adapting Framework for U-Net-Based Medical Image Segmentation 2018 [paper]
[3] No New-Net 2018 [paper]