Segmentation Consistency Training: Out-of-Distribution Generalization for Medical Image Segmentation
This addresses the challenge of model failure due to hospital or imaging routine changes in medical imaging, offering a method to improve robustness, though it appears incremental as an alternative to data augmentation.
The paper tackles the problem of poor out-of-distribution generalization in medical image segmentation by introducing Consistency Training, which uses a novel Segmentation Inconsistency Loss to maximize prediction consistency across augmented and unaugmented data, and demonstrates that it outperforms conventional data augmentation on several out-of-distribution datasets for polyp segmentation.
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.