Segmentation Consistency Training: Out-of-Distribution Generalization for Medical Image Segmentation

TL;DR

This paper introduces Consistency Training with a novel Segmentation Inconsistency Loss to improve out-of-distribution generalization in medical image segmentation, outperforming traditional data augmentation methods.

Contribution

The paper proposes a new training method and loss function that enhance model robustness to distribution shifts in medical imaging tasks.

Findings

Outperforms data augmentation on multiple OOD datasets

Effective in polyp segmentation tasks

Improves model robustness to hospital and imaging routine changes

Abstract

Generalizability is seen as one of the major challenges in deep learning, in particular in the domain of medical imaging, where a change of hospital or in imaging routines can lead to a complete failure of a model. To tackle this, we introduce Consistency Training, a training procedure and alternative to data augmentation based on maximizing models' prediction consistency across augmented and unaugmented data in order to facilitate better out-of-distribution generalization. To this end, we develop a novel region-based segmentation loss function called Segmentation Inconsistency Loss (SIL), which considers the differences between pairs of augmented and unaugmented predictions and labels. We demonstrate that Consistency Training outperforms conventional data augmentation on several out-of-distribution datasets on polyp segmentation, a popular medical task.