Multi Scale Supervised 3D U-Net for Kidney and Tumor Segmentation

U-Net has achieved huge success in various medical image segmentation challenges. Kinds of new architectures with bells and whistles might succeed in certain dataset when employed with optimal hyper-parameter, but their generalization always can't be guaranteed. Here, we focused on the basic U-Net architecture and proposed a multi scale supervised 3D U-Net for the segmentation task in KiTS19 challenge. To enhance the performance, our work can be summarized as three folds: first, we used multi scale supervision in the decoder pathway, which could encourage the network to predict right results from the deep layers; second, with the aim to alleviate the bad effect from the sample imbalance of kidney and tumor, we adopted exponential logarithmic loss; third, a connected-component based post processing method was designed to remove the obviously wrong voxels. In the published KiTS19 training dataset (totally 210 patients), we divided 42 patients to be test dataset and finally obtained DICE scores of 0.969 and 0.805 for the kidney and tumor respectively. In the challenge, we finally achieved the 7th place among 106 teams with the Composite Dice of 0.8961, namely 0.9741 for kidney and 0.8181 for tumor.

[1]  Kaiming He,et al.  Feature Pyramid Networks for Object Detection , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Robert G Uzzo,et al.  The R.E.N.A.L. nephrometry score: a comprehensive standardized system for quantitating renal tumor size, location and depth. , 2009, The Journal of urology.

[3]  Nassir Navab,et al.  Concurrent Spatial and Channel Squeeze & Excitation in Fully Convolutional Networks , 2018, MICCAI.

[4]  Alistair A. Young,et al.  Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) , 2017, MICCAI 2017.

[5]  Loïc Le Folgoc,et al.  Attention U-Net: Learning Where to Look for the Pancreas , 2018, ArXiv.

[6]  Paul Russo,et al.  Commentary on The R.E.N.A.L. nephrometry score: A comprehensive standardized system for quantitating renal tumor size, location and depth , 2010 .

[7]  Tanveer F. Syeda-Mahmood,et al.  3D Segmentation with Exponential Logarithmic Loss for Highly Unbalanced Object Sizes , 2018, MICCAI.

[8]  Kaiming He,et al.  Focal Loss for Dense Object Detection , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[9]  Chi-Wing Fu,et al.  H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation From CT Volumes , 2018, IEEE Transactions on Medical Imaging.

[10]  Klaus H. Maier-Hein,et al.  nnU-Net: Breaking the Spell on Successful Medical Image Segmentation , 2019, ArXiv.

[11]  Tao Zhao,et al.  Kid-Net: Convolution Networks for Kidney Vessels Segmentation from CT-Volumes , 2018, MICCAI.

[12]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[13]  Thomas Brox,et al.  3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation , 2016, MICCAI.

[14]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[15]  Walter Artibani,et al.  Preoperative aspects and dimensions used for an anatomical (PADUA) classification of renal tumours in patients who are candidates for nephron-sparing surgery. , 2009, European urology.

[16]  Nikolaos Papanikolopoulos,et al.  The KiTS19 Challenge Data: 300 Kidney Tumor Cases with Clinical Context, CT Semantic Segmentations, and Surgical Outcomes , 2019, ArXiv.