Cross-View Mutual Learning for Semi-Supervised Medical Image Segmentation

Song Wu, Xiaoyu Wei, Xinyue Chen, Yazhou Ren, Jing He, Xiaorong Pu

Abstract: Semi-supervised medical image segmentation has gained increasing attention due to its potential to alleviate the manual annotation burden. Mainstream methods typically involve two subnets, and conduct a consistency objective to ensure them producing consistent predictions for unlabeled data. However, they often ignore that the complementarity of model predictions is equally crucial. To realize the potential of the multi-subnet architecture, we propose a novel cross-view mutual learning method with a two-branch co-training framework. Specifically, we first introduce a novel conflict-based feature learning (CFL) that encourages the two subnets to learn distinct features from the same input. These distinct features are then decoded into complementary model predictions, allowing both subnets to understand the input from different views. More importantly, we propose a cross-view mutual learning (CML) to maximize the effectiveness of CFL. This approach requires only modifications to the model inputs and supervisory signals, and implements a heterogeneous consistency objective to fully explore the complementarity of model predictions. Consequently, the aggregated predictions can effectively capture both consistency and complementarity across two subnets. Experimental results on three public datasets demonstrate the superiority of CML over previous SoTA methods. Code is available at https://github.com/SongwuJob/CML.

Primary Subject Area: [Content] Vision and Language

Relevance To Conference: In the realm of multimodal medical diagnosis, medical images are typically multimodal and the amount of data is huge, while the acquisition of high-quality labeled data entails significant expense. Semi-supervised learning can effectively utilize unlabeled data and reduce the reliance on large amounts of labeled data, thereby improving the efficiency of data processing and reducing costs. In light of this, we aim to design a simple yet effective semi-supervised medical image segmentation paradigm that greatly reduces label requirement while achieving superior segmentation performance.

Submission Number: 1552