Exploring Foundation Models for Synthetic Medical Imaging: A Study on Chest X-Rays and Fine-Tuning Techniques

TL;DR

This paper investigates the use of foundation models, specifically Latent Diffusion Models, for generating realistic synthetic chest X-ray images, and evaluates how fine-tuning enhances their performance and realism.

Contribution

It introduces a methodology for fine-tuning foundation models with medical data and professional input to improve synthetic medical image quality.

Findings

Fine-tuning improves image realism and quality.

Professional input enhances the authenticity of generated images.

Latent Diffusion Models effectively generate realistic chest X-rays.

Abstract

Machine learning has significantly advanced healthcare by aiding in disease prevention and treatment identification. However, accessing patient data can be challenging due to privacy concerns and strict regulations. Generating synthetic, realistic data offers a potential solution for overcoming these limitations, and recent studies suggest that fine-tuning foundation models can produce such data effectively. In this study, we explore the potential of foundation models for generating realistic medical images, particularly chest x-rays, and assess how their performance improves with fine-tuning. We propose using a Latent Diffusion Model, starting with a pre-trained foundation model and refining it through various configurations. Additionally, we performed experiments with input from a medical professional to assess the realism of the images produced by each trained model.