Paper Info
Title
All-Around Real Label Supervision: Cyclic Prototype Consistency Learning for Semi-Supervised Medical Image Segmentation
Abstract
Semi-supervised learning has substantially advanced medical image segmentation since it alleviates the heavy burden of acquiring the costly expert-examined annotations. Especially, the consistency-based approaches have attracted more attention for their superior performance, wherein the real labels are only utilized to supervise their paired images via supervised loss while the unlabeled images are exploited by enforcing the perturbation-based <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">“unsupervised”</i> consistency without explicit guidance from those real labels. However, intuitively, the expert-examined real labels contain more reliable supervision signals. Observing this, we ask an unexplored but interesting question: can we exploit the unlabeled data via explicit real label supervision for semi-supervised training? To this end, we discard the previous perturbation-based consistency but absorb the essence of non-parametric prototype learning. Based on the prototypical networks, we then propose a novel cyclic prototype consistency learning (CPCL) framework, which is constructed by a labeled-to-unlabeled (L2U) prototypical forward process and an unlabeled-to-labeled (U2L) backward process. Such two processes synergistically enhance the segmentation network by encouraging morediscriminative and compact features. In this way, our framework turns previous <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">“unsupervised”</i> consistency into new <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">“supervised”</i> consistency, obtaining the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">“all-around real label supervision”</i> property of our method. Extensive experiments on brain tumor segmentation from MRI and kidney segmentation from CT images show that our CPCL can effectively exploit the unlabeled data and outperform other state-of-the-art semi-supervised medical image segmentation methods.
Year
DOI
Venue
2022
10.1109/JBHI.2022.3162043
IEEE Journal of Biomedical and Health Informatics
Keywords
DocType
Volume
Brain Neoplasms,Humans,Kidney,Magnetic Resonance Imaging,Supervised Machine Learning
Journal
26
Issue
ISSN
Citations 
7
2168-2194
1
PageRank 
References 
Authors
0.35
20
9
Name
Order
Citations
PageRank
Zhe Xu111.03
Yixin Wang21712.87
Donghuan Lu310.35
Lequan Yu470639.80
Yan Jiangpeng514.75
Jie Luo670673.44
Kai Ma74918.48
Yefeng Zheng81391114.67
Raymond Kai-yu Tong910.35