Fair Text to Medical Image Diffusion Model with Subgroup Distribution Aligned Tuning

TL;DR

This paper introduces a fair text-to-medical-image diffusion model that reduces subgroup bias, particularly gender bias, by aligning generated image distributions with target datasets using subgroup classifiers and regularization techniques.

Contribution

The work proposes a novel fine-tuning approach for T2MedI models that aligns subgroup distributions, improving fairness without sacrificing image quality.

Findings

Reduced gender bias in generated medical images.

Enhanced alignment of subgroup proportions with target datasets.

Maintained high image quality through CLIP-consistency regularization.

Abstract

The text to medical image (T2MedI) with latent diffusion model has great potential to alleviate the scarcity of medical imaging data and explore the underlying appearance distribution of lesions in a specific patient status description. However, as the text to nature image models, we show that the T2MedI model can also bias to some subgroups to overlook the minority ones in the training set. In this work, we first build a T2MedI model based on the pre-trained Imagen model, which has the fixed contrastive language-image pre-training (CLIP) text encoder, while its decoder has been fine-tuned on medical images from the Radiology Objects in COntext (ROCO) dataset. Its gender bias is analyzed qualitatively and quantitatively. Toward this issue, we propose to fine-tune the T2MedI toward the target application dataset to align their sensitive subgroups distribution probability. Specifically, the alignment loss for fine-tuning is guided by an off-the-shelf sensitivity-subgroup classifier to match the classification probability between the generated images and the expected target dataset. In addition, the image quality is maintained by a CLIP-consistency regularization term following a knowledge distillation scheme. For evaluation, we set the target dataset to be enhanced as the BraST18 dataset, and trained a brain magnetic resonance (MR) slice-based gender classifier from it. With our method, the generated MR image can markedly reduce the inconsistency with the gender proportion in the BraTS18 dataset.