Contrastive Semi-Supervised Learning for 2D Medical Image Segmentation
This work addresses medical image segmentation, a domain-specific problem, with incremental improvements in semi-supervised learning techniques.
The paper tackles the problem of 2D medical image segmentation by proposing a semi-supervised method using contrastive learning on patches and a consistency regularization scheme, resulting in consistent improvements over state-of-the-art approaches across multiple datasets.
Contrastive Learning (CL) is a recent representation learning approach, which encourages inter-class separability and intra-class compactness in learned image representations. Since medical images often contain multiple semantic classes in an image, using CL to learn representations of local features (as opposed to global) is important. In this work, we present a novel semi-supervised 2D medical segmentation solution that applies CL on image patches, instead of full images. These patches are meaningfully constructed using the semantic information of different classes obtained via pseudo labeling. We also propose a novel consistency regularization (CR) scheme, which works in synergy with CL. It addresses the problem of confirmation bias, and encourages better clustering in the feature space. We evaluate our method on four public medical segmentation datasets and a novel histopathology dataset that we introduce. Our method obtains consistent improvements over state-of-the-art semi-supervised segmentation approaches for all datasets.