- Shekoofeh Azizi
- Laura Anne Culp
- Jan Freyberg
- Basil Mustafa
- Sebastien Baur
- Simon Kornblith
- Ting Chen
- Patricia MacWilliams
- Sara Mahdavi
- Ellery Wulczyn
- Boris Babenko
- Megan Zoë Walker
- Aaron Loh
- Cameron Chen
- Yuan Liu
- Pinal Bavishi
- Scott Mayer McKinney
- Jim Winkens
- Abhijit Guha Roy
- Zach William Beaver
- Fiona Keleher Ryan
- Justin David Krogue
- Mozziyar Etemadi
- Umesh Telang
- Yun Liu
- Lily Hao Yi Peng
- Greg Corrado
- Dale Richard Webster
- David James Fleet
- Geoffrey Everest Hinton
- Neil Houlsby
- Alan Karthikesalingam
- Mohammad Norouzi
- Vivek Natarajan
Abstract
Recent progress in Medical Artificial Intelligence (AI) has delivered systems that can reach clinical expert level performance. However, such systems tend to demonstrate sub-optimal out-of-distribution'' performance when evaluated in clinical settings different from the training environment. A common mitigation strategy is to develop separate systems for each clinical setting by retraining using site-specific data. However, this quickly becomes impractical as medical data is time-consuming to acquire and expensive to annotate. Thus, the problem of
data-efficient generalization'' presents an ongoing difficulty for Medical AI development. Although progress in representation learning shows promise, their benefits have not been rigorously studied, specifically for out-of-distribution settings. To meet these challenges, we present REMEDIS, a unified representation learning strategy to improve robustness and data-efficiency of medical imaging AI. REMEDIS uses a generic combination of large-scale supervised transfer learning on natural images with intermediate contrastive self-supervised learning on medical images and requires little task-specific customization. We study a diverse range of medical imaging tasks and simulate three realistic application scenarios using retrospective data. REMEDIS exhibits significantly improved in-distribution performance with up to 11.5% relative improvement in diagnostic accuracy over a strong supervised baseline. More importantly, our strategy leads to strong data-efficient generalization of medical imaging AI, matching strong supervised baselines using between 1% to 33% of retraining data across tasks. These results suggest that REMEDIS can significantly accelerate the life-cycle of medical imaging AI development thereby presenting an important step forward for medical imaging AI to deliver broad impact.
Research Areas
Learn more about how we do research
We maintain a portfolio of research projects, providing individuals and teams the freedom to emphasize specific types of work