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.
@article{arxiv.2205.15428,
title = {Segmentation Consistency Training: Out-of-Distribution Generalization for Medical Image Segmentation},
author = {Birk Torpmann-Hagen and Vajira Thambawita and Kyrre Glette and Pål Halvorsen and Michael A. Riegler},
journal= {arXiv preprint arXiv:2205.15428},
year = {2022}
}