Can Medical Vision-Language Pre-training Succeed with Purely Synthetic Data?
This addresses data scarcity in medical AI for zero-shot tasks, offering a potential solution to improve model performance with synthetic data, though it is incremental as it builds on existing generative models.
The study tackled the problem of scarce paired image-text data for Medical Vision-Language Pre-training (MedVLP) by using synthetic data generated from off-the-shelf models, finding that models trained solely on synthetic data outperformed those on real data by 3.8% in averaged AUC for zero-shot classification, with a combination improving performance by 9.07%.
Medical Vision-Language Pre-training (MedVLP) has made significant progress in enabling zero-shot tasks for medical image understanding. However, training MedVLP models typically requires large-scale datasets with paired, high-quality image-text data, which are scarce in the medical domain. Recent advancements in Large Language Models (LLMs) and diffusion models have made it possible to generate large-scale synthetic image-text pairs. This raises the question: "Can MedVLP succeed using purely synthetic data?" To address this, we use off-the-shelf generative models to create synthetic radiology reports and paired Chest X-ray (CXR) images, and propose an automated pipeline to build a diverse, high-quality synthetic dataset, enabling a rigorous study that isolates model and training settings, focusing entirely from the data perspective. Our results show that MedVLP models trained exclusively on synthetic data outperform those trained on real data by 3.8% in averaged AUC on zero-shot classification. Moreover, using a combination of synthetic and real data leads to a further improvement of 9.07%. Additionally, MedVLP models trained on synthetic or mixed data consistently outperform those trained on real data in zero-shot grounding, as well as in fine-tuned classification and segmentation tasks. Our analysis suggests MedVLP trained on well-designed synthetic data can outperform models trained on real datasets, which may be limited by low-quality samples and long-tailed distributions.