Momentum Contrastive Voxel-wise Representation Learning for Semi-supervised Volumetric Medical Image Segmentation

Contrastive learning (CL) aims to learn useful representation without relying on expert annotations in the context of medical image segmentation. Existing approaches mainly contrast a single positive vector ( , an augmentation of the same image) against a set of negatives within the entire remainder...

Full description

Saved in:
Bibliographic Details
Published in:Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention Vol. 13434; p. 639
Main Authors: You, Chenyu, Zhao, Ruihan, Staib, Lawrence, Duncan, James S
Format: Journal Article
Language:English
Published: Germany 01.01.2022
Subjects:
Contrastive Learning
Medical Image Segmentation
Semi-Supervised Learning
Online Access:Get more information
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Contrastive learning (CL) aims to learn useful representation without relying on expert annotations in the context of medical image segmentation. Existing approaches mainly contrast a single positive vector ( , an augmentation of the same image) against a set of negatives within the entire remainder of the batch by simply mapping all input features into the same constant vector. Despite the impressive empirical performance, those methods have the following shortcomings: (1) it remains a formidable challenge to prevent the collapsing problems to trivial solutions; and (2) we argue that not all voxels within the same image are equally positive since there exist the dissimilar anatomical structures with the same image. In this work, we present a novel ontrastive oxel-wise epresentation earning (CVRL) method to effectively learn low-level and high-level features by capturing 3D spatial context and rich anatomical information along both the feature and the batch dimensions. Specifically, we first introduce a novel CL strategy to ensure feature diversity promotion among the 3D representation dimensions. We train the framework through bi-level contrastive optimization ( , low-level and high-level) on 3D images. Experiments on two benchmark datasets and different labeled settings demonstrate the superiority of our proposed framework. More importantly, we also prove that our method inherits the benefit of hardness-aware property from the standard CL approaches. Codes will be available soon.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
DOI:10.1007/978-3-031-16440-8_61