FairDiff: Fair Segmentation with Point-Image Diffusion

TL;DR

FairDiff introduces a novel point-image diffusion method for generating synthetic medical images with precise boundary control, enhancing data balance and fairness in medical image segmentation.

Contribution

The paper presents a new Point-Image Diffusion architecture that improves synthetic image control and fairness in medical segmentation tasks, outperforming existing methods.

Findings

Outperforms existing techniques in synthetic SLO fundus image generation.

Achieves superior fairness in segmentation when combining synthetic and real data.

Demonstrates significant improvements over state-of-the-art fairness models.

Abstract

Fairness is an important topic for medical image analysis, driven by the challenge of unbalanced training data among diverse target groups and the societal demand for equitable medical quality. In response to this issue, our research adopts a data-driven strategy-enhancing data balance by integrating synthetic images. However, in terms of generating synthetic images, previous works either lack paired labels or fail to precisely control the boundaries of synthetic images to be aligned with those labels. To address this, we formulate the problem in a joint optimization manner, in which three networks are optimized towards the goal of empirical risk minimization and fairness maximization. On the implementation side, our solution features an innovative Point-Image Diffusion architecture, which leverages 3D point clouds for improved control over mask boundaries through a point-mask-image synthesis pipeline. This method outperforms significantly existing techniques in synthesizing scanning laser ophthalmoscopy (SLO) fundus images. By combining synthetic data with real data during the training phase using a proposed Equal Scale approach, our model achieves superior fairness segmentation performance compared to the state-of-the-art fairness learning models. Code is available at https://github.com/wenyi-li/FairDiff.