Consider an educational scenario where interacting with a radiology image can
substantially improve learning. This demonstration shows how MedGemma might be built upon to provide
a useful tool for exploring radiology images and associated reports by translating them into
simple language, with visual cues to highlight the relevant areas of the image.
Disclaimer This
demonstration is for illustrative purposes only and does not represent a finished or approved
product. It is not representative of compliance to any regulations or standards for
quality, safety or efficacy. Any real-world application would require additional development,
training, and adaptation. The experience highlighted in this demo shows MedGemma's baseline
capability for the displayed task and is intended to help developers and users explore possible
applications and inspire further development.
keyboard_arrow_leftBack
X-Ray
Effusion
Infection
Lymphadenopathy
Nodule A
Nodule B
CT
Tumor
code_blocksDetails about this Demo
Medical Image
Loading image...
This shows a single slice of the CT. Not all elements in the report can be visualized.
Loading report details...
Generating explanation... Please wait.
What this means
Click a sentence to see the explanation here.
Select a report to view its text.
warningThis demonstration is for illustrative purposes of MedGemma's baseline
capability only. It does not represent a finished or approved product, is not intended to
diagnose or suggest treatment of any disease or condition, and should not be used for medical
advice.
Details About This Demo
The Model: This demo exclusively features Google's MedGemma-4B, a Gemma 3-based model
fine-tuned for
comprehending medical text and images, such as chest X-rays. It demonstrates MedGemma's ability to
accelerate the development of AI-powered healthcare applications by offering advanced
interpretation of medical data.
Health AI Developer Foundations (HAI-DEF) provides a collection of open-weight models and
companion resources to empower developers in building AI models for healthcare.
Enjoying the Demo? We'd love your feedback! If you found this demo helpful, please show
your appreciation by clicking the ♡ button on the HuggingFace page, linked at the top.
Explore More Demos: Discover additional demonstrations on HuggingFace Spaces or via Colabs:
CXR Foundations Demo
-
Showcases on-browser, data-efficient, and zero-shot classification of CXR images.
Path Foundations Demo
-
Highlights on-browser, data-efficient classification and outlier detection within pathology
slides.