Comparing to Learn: Surpassing ImageNet Pretraining on Radiographs By Comparing Image Representations
This addresses the problem of domain mismatch in medical image analysis for researchers and practitioners, offering a more effective pretraining approach, though it appears incremental as it builds on existing self-supervised learning ideas.
The paper tackles the domain gap between natural and medical images by proposing a new pretraining method called Comparing to Learn (C2L), which learns from 700k unlabeled radiographs and significantly outperforms ImageNet pretraining and previous state-of-the-art approaches on radiograph tasks.
In deep learning era, pretrained models play an important role in medical image analysis, in which ImageNet pretraining has been widely adopted as the best way. However, it is undeniable that there exists an obvious domain gap between natural images and medical images. To bridge this gap, we propose a new pretraining method which learns from 700k radiographs given no manual annotations. We call our method as Comparing to Learn (C2L) because it learns robust features by comparing different image representations. To verify the effectiveness of C2L, we conduct comprehensive ablation studies and evaluate it on different tasks and datasets. The experimental results on radiographs show that C2L can outperform ImageNet pretraining and previous state-of-the-art approaches significantly. Code and models are available.