FairVision: Equitable Deep Learning for Eye Disease Screening via Fair Identity Scaling

TL;DR

This paper investigates fairness issues in 3D eye disease screening models, revealing biases across demographic groups, and introduces a novel method and dataset to improve equity in medical AI.

Contribution

It is the first comprehensive study on 3D medical imaging fairness, proposing a new fair identity scaling method and releasing a large-scale dataset for equitable AI research.

Findings

Significant biases found across race, gender, and ethnicity in 3D models

The proposed FIS method outperforms existing fairness techniques

Harvard-FairVision dataset enables large-scale fairness analysis

Abstract

Equity in AI for healthcare is crucial due to its direct impact on human well-being. Despite advancements in 2D medical imaging fairness, the fairness of 3D models remains underexplored, hindered by the small sizes of 3D fairness datasets. Since 3D imaging surpasses 2D imaging in SOTA clinical care, it is critical to understand the fairness of these 3D models. To address this research gap, we conduct the first comprehensive study on the fairness of 3D medical imaging models across multiple protected attributes. Our investigation spans both 2D and 3D models and evaluates fairness across five architectures on three common eye diseases, revealing significant biases across race, gender, and ethnicity. To alleviate these biases, we propose a novel fair identity scaling (FIS) method that improves both overall performance and fairness, outperforming various SOTA fairness methods. Moreover, we release Harvard-FairVision, the first large-scale medical fairness dataset with 30,000 subjects featuring both 2D and 3D imaging data and six demographic identity attributes. Harvard-FairVision provides labels for three major eye disorders affecting about 380 million people worldwide, serving as a valuable resource for both 2D and 3D fairness learning. Our code and dataset are publicly accessible at \url{https://ophai.hms.harvard.edu/datasets/harvard-fairvision30k}.