X-ray Insights Unleashed: Pioneering the Enhancement of Multi-Label Long-Tail Data

TL;DR

This paper introduces a novel data synthesis pipeline that enhances the representation of rare lesions in chest X-ray datasets by using diffusion models and large language model guidance, improving diagnostic accuracy.

Contribution

It presents a new method combining diffusion-based image inpainting with LLM guidance and incremental learning to better augment tail classes in medical imaging datasets.

Findings

Achieves state-of-the-art performance on MIMIC and CheXpert datasets.

Effectively generates realistic tail lesion images for training.

Improves diagnostic accuracy for rare pulmonary anomalies.

Abstract

Long-tailed pulmonary anomalies in chest radiography present formidable diagnostic challenges. Despite the recent strides in diffusion-based methods for enhancing the representation of tailed lesions, the paucity of rare lesion exemplars curtails the generative capabilities of these approaches, thereby leaving the diagnostic precision less than optimal. In this paper, we propose a novel data synthesis pipeline designed to augment tail lesions utilizing a copious supply of conventional normal X-rays. Specifically, a sufficient quantity of normal samples is amassed to train a diffusion model capable of generating normal X-ray images. This pre-trained diffusion model is subsequently utilized to inpaint the head lesions present in the diseased X-rays, thereby preserving the tail classes as augmented training data. Additionally, we propose the integration of a Large Language Model Knowledge Guidance (LKG) module alongside a Progressive Incremental Learning (PIL) strategy to stabilize the inpainting fine-tuning process. Comprehensive evaluations conducted on the public lung datasets MIMIC and CheXpert demonstrate that the proposed method sets a new benchmark in performance.