EVA-X: a foundation model for general chest x-ray analysis with self-supervised learning.
Authors
Affiliations (11)
Affiliations (11)
- School of Electronic Information and Communications, Huazhong University of Science and Technology, Wuhan, Hubei, China.
- School of Electronic Information and Communications, Huazhong University of Science and Technology, Wuhan, Hubei, China. [email protected].
- Department of Radiology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei, China.
- Peter Munk Cardiac Centre, University Health Network, Toronto, ON, Canada.
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON, Canada.
- Vector Institute for Artificial Intelligence, Toronto, ON, Canada.
- Peter Munk Cardiac Centre, University Health Network, Toronto, ON, Canada. [email protected].
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON, Canada. [email protected].
- Vector Institute for Artificial Intelligence, Toronto, ON, Canada. [email protected].
- AI Hub, University Health Network, Toronto, ON, Canada. [email protected].
- Department of Computer Science, University of Toronto, Toronto, ON, Canada. [email protected].
Abstract
Artificial intelligence analysis methods for chest X-ray images are limited by insufficient annotation data and varying levels of annotation, resulting in weak generalization ability and difficulty in clinical dissemination. Here, we present EVA-X, an innovative foundational model based on X-ray images with broad applicability. EVA-X uses a self-supervised learning method capable of capturing both semantic and geometric information from unlabeled images for universal X-ray image representation. It has demonstrated exceptional performance in chest disease analysis and localization, becoming a model capable of spanning over 20 different chest pathologies and achieving leading results in over 11 different pathology detection tasks. Additionally, EVA-X significantly reduces the burden of data annotation in the medical AI field, showcasing strong potential in the domain of few-shot learning. The emergence of EVA-X will greatly propel the development and application of foundational medical models, leading to potential improvements in future medical research and clinical practice.