A Generative Foundation Model for Chest Radiography
This work addresses the need for reliable and equitable AI models in medical imaging, offering a domain-specific solution that is incremental in applying generative techniques to chest X-rays.
The authors tackled the problem of scarce annotated medical images for AI in healthcare by developing ChexGen, a generative vision-language foundation model for chest radiographs, which improved disease classification, detection, and segmentation tasks using a small fraction of training data, as validated by expert evaluations and quantitative metrics.
The scarcity of well-annotated diverse medical images is a major hurdle for developing reliable AI models in healthcare. Substantial technical advances have been made in generative foundation models for natural images. Here we develop `ChexGen', a generative vision-language foundation model that introduces a unified framework for text-, mask-, and bounding box-guided synthesis of chest radiographs. Built upon the latent diffusion transformer architecture, ChexGen was pretrained on the largest curated chest X-ray dataset to date, consisting of 960,000 radiograph-report pairs. ChexGen achieves accurate synthesis of radiographs through expert evaluations and quantitative metrics. We demonstrate the utility of ChexGen for training data augmentation and supervised pretraining, which led to performance improvements across disease classification, detection, and segmentation tasks using a small fraction of training data. Further, our model enables the creation of diverse patient cohorts that enhance model fairness by detecting and mitigating demographic biases. Our study supports the transformative role of generative foundation models in building more accurate, data-efficient, and equitable medical AI systems.