Medical domain knowledge in domain-agnostic generative AI

Jakob Nikolas Kather, Md, MSc, Jakob Nikolas Kather, Narmin Ghaffari Laleh, Sebastian Foersch, Daniel TruhnNpj Digit. Medicine 2022 · DOI

Introduction

Most current medical AI systems are "narrow"—highly engineered for specific tasks, such as detecting a particular tumor type on a specific scan. While effective, these models are often brittle, failing when faced with data from different hospitals or populations. This study investigates whether large-scale, domain-agnostic generative models, such as GLIDE, can overcome these limitations. These models are trained on millions of uncurated image-text pairs from the internet, raising the question: does this massive, general training accidentally convey useful medical knowledge?

Methodology

The researchers evaluated GLIDE (Guided Language to Image Diffusion for Generation and Editing), a text-to-image diffusion model. They compared two versions: a standard model and a CLIP-conditioned version, which uses a secondary model to guide the diffusion process toward the text prompt.

To quantify performance, the team recruited three medical experts—a pathologist, a radiologist, and an internist. These experts performed a blinded evaluation of images generated from a benchmark of 274 medical prompts. The prompts covered various modalities, including microscopy, histology, X-ray, MRI, CT, photography, scientific posters, and illustrations. Images were rated on an ordinal scale from 0 (completely wrong) to 4 (perfectly correct) for both style (visual appearance) and content (anatomical accuracy).

Figure 3: A comprehensive grid showing GLIDE's generated outputs across various anatomical regions and imaging modalities, illustrating the breadth of the evaluation.

Key Results and Insights

Histology vs. Radiology

The study found a significant performance gap between medical fields. GLIDE showed a "reasonably strong" grasp of histopathology (the study of tissues under a microscope), achieving a high median score of 3 out of 4 for style. This is likely because the internet contains many colorful, pattern-heavy images similar to histology.

In contrast, the model performed poorly in radiology. Style and content scores for X-ray, MRI, and CT were significantly lower ($p < 0.05$) than all other categories. The model often produced nonsensical anatomical structures in these modalities, likely because raw radiology data is less common in general web-scraping results than histological imagery.

Semantic Confusion and CLIP Guidance

A striking finding was the model's failure in semantic disambiguation. When prompted for "Grave’s disease" (an autoimmune thyroid condition), the model generated a photograph of a literal gravestone. Similarly, a prompt for "blood vessels" occasionally resulted in images of beer glasses. These failures highlight that while the model has visual "knowledge," it lacks deep medical context.

Figure 5: Examples of "common confusions" where the model interprets medical terms literally, such as generating a gravestone for Grave's disease or beer glasses for blood vessels.

Furthermore, the study found that CLIP guidance is essential for medical tasks. While unguided models might produce higher-quality general images, CLIP was necessary to keep the model "on track" with specific medical concepts, preventing hallucinations like a heart made entirely of blood vessels.

Conclusion and Future Work

Despite its flaws, the model demonstrated "zero-shot" capabilities—the ability to synthesize images of categories it wasn't explicitly trained on by combining different concepts (e.g., "brain" in a "histopathology style"). However, anatomical inaccuracies and literal semantic failures make these models unsuitable for clinical use "as-is."

The researchers suggest that generalist models should serve as a foundation for domain-specific fine-tuning. By taking a generalist model like GLIDE and giving it a "medical residency"—training it further on high-quality, curated medical datasets—we may develop more robust and versatile medical AI systems.

Summary updated

BibTeX
@article{msc2022medical,
  author = {Jakob Nikolas Kather, MD, MSc and Jakob Nikolas Kather and Narmin Ghaffari Laleh and Sebastian Foersch and Daniel Truhn},
  title = {{Medical domain knowledge in domain-agnostic generative AI}},
  journal = {Npj Digit. Medicine},
  year = {2022},
  doi = {10.1038/s41746-022-00634-5},
}
Figure