Semi-supervised Left Atrium Segmentation with Mutual Consistency Training

TL;DR

This paper introduces MC-Net, a semi-supervised learning approach for left atrium segmentation in 3D MR images, emphasizing challenging regions to improve accuracy and outperform existing methods.

Contribution

The paper presents a novel mutual consistency training scheme with dual decoders and a cycled pseudo label approach for semi-supervised medical image segmentation.

Findings

Outperforms six recent semi-supervised methods

Achieves state-of-the-art results on LA database

Effectively exploits unlabeled data in challenging regions

Abstract

Semi-supervised learning has attracted great attention in the field of machine learning, especially for medical image segmentation tasks, since it alleviates the heavy burden of collecting abundant densely annotated data for training. However, most of existing methods underestimate the importance of challenging regions (e.g. small branches or blurred edges) during training. We believe that these unlabeled regions may contain more crucial information to minimize the uncertainty prediction for the model and should be emphasized in the training process. Therefore, in this paper, we propose a novel Mutual Consistency Network (MC-Net) for semi-supervised left atrium segmentation from 3D MR images. Particularly, our MC-Net consists of one encoder and two slightly different decoders, and the prediction discrepancies of two decoders are transformed as an unsupervised loss by our designed cycled pseudo label scheme to encourage mutual consistency. Such mutual consistency encourages the two decoders to have consistent and low-entropy predictions and enables the model to gradually capture generalized features from these unlabeled challenging regions. We evaluate our MC-Net on the public Left Atrium (LA) database and it obtains impressive performance gains by exploiting the unlabeled data effectively. Our MC-Net outperforms six recent semi-supervised methods for left atrium segmentation, and sets the new state-of-the-art performance on the LA database.