Efficient Prototype Consistency Learning in Medical Image Segmentation via Joint Uncertainty and Data Augmentation

TL;DR

This paper introduces EPCL-JUDA, a novel semi-supervised medical image segmentation method that enhances prototype learning through joint uncertainty quantification and data augmentation, leading to improved segmentation performance.

Contribution

The paper proposes a new framework combining uncertainty quantification and data augmentation within prototype learning to better represent classes with limited labeled data in medical segmentation.

Findings

EPCL-JUDA outperforms previous methods on multiple datasets.

The approach effectively fuses labeled and unlabeled prototypes.

Prototype network reduces memory requirements.

Abstract

Recently, prototype learning has emerged in semi-supervised medical image segmentation and achieved remarkable performance. However, the scarcity of labeled data limits the expressiveness of prototypes in previous methods, potentially hindering the complete representation of prototypes for class embedding. To overcome this issue, we propose an efficient prototype consistency learning via joint uncertainty quantification and data augmentation (EPCL-JUDA) to enhance the semantic expression of prototypes based on the framework of Mean-Teacher. The concatenation of original and augmented labeled data is fed into student network to generate expressive prototypes. Then, a joint uncertainty quantification method is devised to optimize pseudo-labels and generate reliable prototypes for original and augmented unlabeled data separately. High-quality global prototypes for each class are formed by fusing labeled and unlabeled prototypes, which are utilized to generate prototype-to-features to conduct consistency learning. Notably, a prototype network is proposed to reduce high memory requirements brought by the introduction of augmented data. Extensive experiments on Left Atrium, Pancreas-NIH, Type B Aortic Dissection datasets demonstrate EPCL-JUDA's superiority over previous state-of-the-art approaches, confirming the effectiveness of our framework. The code will be released soon.