Dcl-Net: Dual Contrastive Learning Network for Semi-Supervised Multi-Organ Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wen, Lu, Feng, Zhenghao, Hou, Yun, Wang, Peng, Wu, Xi, Zhou, Jiliu, Wang, Yan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917660802940928
author Wen, Lu
Feng, Zhenghao
Hou, Yun
Wang, Peng
Wu, Xi
Zhou, Jiliu
Wang, Yan
author_facet Wen, Lu
Feng, Zhenghao
Hou, Yun
Wang, Peng
Wu, Xi
Zhou, Jiliu
Wang, Yan
contents Semi-supervised learning is a sound measure to relieve the strict demand of abundant annotated datasets, especially for challenging multi-organ segmentation . However, most existing SSL methods predict pixels in a single image independently, ignoring the relations among images and categories. In this paper, we propose a two-stage Dual Contrastive Learning Network for semi-supervised MoS, which utilizes global and local contrastive learning to strengthen the relations among images and classes. Concretely, in Stage 1, we develop a similarity-guided global contrastive learning to explore the implicit continuity and similarity among images and learn global context. Then, in Stage 2, we present an organ-aware local contrastive learning to further attract the class representations. To ease the computation burden, we introduce a mask center computation algorithm to compress the category representations for local contrastive learning. Experiments conducted on the public 2017 ACDC dataset and an in-house RC-OARs dataset has demonstrated the superior performance of our method.
format Preprint
id arxiv_https___arxiv_org_abs_2403_03512
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Dcl-Net: Dual Contrastive Learning Network for Semi-Supervised Multi-Organ Segmentation
Wen, Lu
Feng, Zhenghao
Hou, Yun
Wang, Peng
Wu, Xi
Zhou, Jiliu
Wang, Yan
Computer Vision and Pattern Recognition
Semi-supervised learning is a sound measure to relieve the strict demand of abundant annotated datasets, especially for challenging multi-organ segmentation . However, most existing SSL methods predict pixels in a single image independently, ignoring the relations among images and categories. In this paper, we propose a two-stage Dual Contrastive Learning Network for semi-supervised MoS, which utilizes global and local contrastive learning to strengthen the relations among images and classes. Concretely, in Stage 1, we develop a similarity-guided global contrastive learning to explore the implicit continuity and similarity among images and learn global context. Then, in Stage 2, we present an organ-aware local contrastive learning to further attract the class representations. To ease the computation burden, we introduce a mask center computation algorithm to compress the category representations for local contrastive learning. Experiments conducted on the public 2017 ACDC dataset and an in-house RC-OARs dataset has demonstrated the superior performance of our method.
title Dcl-Net: Dual Contrastive Learning Network for Semi-Supervised Multi-Organ Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.03512