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

Abstract

Accurate segmentation of kidneys and kidney tumors is an essential step for radiomic analysis as well as developing advanced surgical planning techniques. In clinical analysis, the segmentation is currently performed by clinicians from the visual inspection images gathered through a computed tomography (CT) scan. This process is laborious and its success significantly depends on previous experience. Moreover, the uncertainty in the tumor location and heterogeneity of scans across patients increases the error rate. To tackle this issue, computer-aided segmentation based on deep learning techniques have become increasingly popular. We present a multi-scale supervised 3D U-Net, MSS U-Net, to automatically segment kidneys and kidney tumors from CT images. Our architecture combines deep supervision with exponential logarithmic loss to increase the 3D U-Net training efficiency. Furthermore, we introduce a connected-component based post processing method to enhance the performance of the overall process. This architecture shows superior performance compared to state-of-the-art works using data from KiTS19 public dataset, with the Dice coefficient of kidney and tumor up to 0.969 and 0.805 respectively. The segmentation techniques introduced in this paper have been tested in the KiTS19 challenge with its corresponding dataset.

References in corpus (12)

• Attention U-Net: Learning Where to Look for the Pancreas
• Brain Tumor Segmentation with Deep Neural Networks
• Fully Convolutional Networks for Semantic Segmentation
• 3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation
• Modified U-Net (mU-Net) with Incorporation of Object-Dependent High Level Features for Improved Liver and Liver-Tumor Segmentation in CT Images
• Deformable Convolutional Networks
• The KiTS19 Challenge Data: 300 Kidney Tumor Cases with Clinical Context, CT Semantic Segmentations, and Surgical Outcomes
• nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation
• 3D Segmentation with Exponential Logarithmic Loss for Highly Unbalanced Object Sizes
• Automatic Liver and Tumor Segmentation of CT and MRI Volumes using Cascaded Fully Convolutional Neural Networks
• Crossbar-Net: A Novel Convolutional Network for Kidney Tumor Segmentation in CT Images
• 3D Kidneys and Kidney Tumor Semantic Segmentation using Boundary-Aware Networks

Cited by in corpus (5)

• U-Net and its variants for medical image segmentation: theory and applications
• Multi-organ segmentation: a progressive exploration of learning paradigms under scarce annotation
• Improve Cross-Modality Segmentation by Treating T1-Weighted MRI Images as Inverted CT Scans
• Crosslink-Net: Double-branch Encoder Segmentation Network via Fusing Vertical and Horizontal Convolutions
• Comparative Evaluation of 3D and 2D Deep Learning Techniques for Semantic Segmentation in CT Scans