Self-supervised Tumor Segmentation through Layer Decomposition

Abstract

In this paper, we target self-supervised representation learning for zero-shot tumor segmentation. We make the following contributions: First, we advocate a zero-shot setting, where models from pre-training should be directly applicable for the downstream task, without using any manual annotations. Second, we take inspiration from "layer-decomposition", and innovate on the training regime with simulated tumor data. Third, we conduct extensive ablation studies to analyse the critical components in data simulation, and validate the necessity of different proxy tasks. We demonstrate that, with sufficient texture randomization in simulation, model trained on synthetic data can effortlessly generalise to segment real tumor data. Forth, our approach achieves superior results for zero-shot tumor segmentation on different downstream datasets, BraTS2018 for brain tumor segmentation and LiTS2017 for liver tumor segmentation. While evaluating the model transferability for tumor segmentation under a low-annotation regime, the proposed approach also outperforms all existing self-supervised approaches, opening up the usage of self-supervised learning in practical scenarios.

Project webpage: https://xiaoman-zhang.github.io/Layer-Decomposition/

References in corpus (13)

• A Simple Framework for Contrastive Learning of Visual Representations
• Neural Discrete Representation Learning
• Learning deep representations by mutual information estimation and maximization
• The Liver Tumor Segmentation Benchmark (LiTS)
• CHAOS Challenge -- Combined (CT-MR) Healthy Abdominal Organ Segmentation
• Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition
• A large annotated medical image dataset for the development and evaluation of segmentation algorithms
• Modified U-Net (mU-Net) with Incorporation of Object-Dependent High Level Features for Improved Liver and Liver-Tumor Segmentation in CT Images
• Sim2real transfer learning for 3D human pose estimation: motion to the rescue
• PGL: Prior-Guided Local Self-supervised Learning for 3D Medical Image Segmentation
• Self-Supervised Similarity Learning for Digital Pathology
• A self-supervised learning strategy for postoperative brain cavity segmentation simulating resections
• Self-supervised Depth Estimation to Regularise Semantic Segmentation in Knee Arthroscopy

Cited by in corpus (1)

• Self-Supervised Anomaly Detection in Computer Vision and Beyond: A Survey and Outlook