On Fairness of Medical Image Classification with Multiple Sensitive Attributes via Learning Orthogonal Representations

TL;DR

This paper introduces a novel method for fair medical image classification that ensures independence between target and multiple sensitive attributes by learning orthogonal representations, addressing fairness in multi-sensitive demographic scenarios.

Contribution

The paper proposes a new orthogonal representation learning approach to mitigate unfairness across multiple sensitive attributes in medical imaging, a first in this domain.

Findings

Effective fairness improvement demonstrated on CheXpert dataset

First approach to handle multiple sensitive attributes in medical imaging fairness

Orthogonal representations enhance fairness without compromising accuracy

Abstract

Mitigating the discrimination of machine learning models has gained increasing attention in medical image analysis. However, rare works focus on fair treatments for patients with multiple sensitive demographic ones, which is a crucial yet challenging problem for real-world clinical applications. In this paper, we propose a novel method for fair representation learning with respect to multi-sensitive attributes. We pursue the independence between target and multi-sensitive representations by achieving orthogonality in the representation space. Concretely, we enforce the column space orthogonality by keeping target information on the complement of a low-rank sensitive space. Furthermore, in the row space, we encourage feature dimensions between target and sensitive representations to be orthogonal. The effectiveness of the proposed method is demonstrated with extensive experiments on the CheXpert dataset. To our best knowledge, this is the first work to mitigate unfairness with respect to multiple sensitive attributes in the field of medical imaging.