@alibidaran on Hugging Face: "With the release of Gemma 4, I launched a new Space called MEDPAI

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

posted an update 22 days ago

Post

4073

With the release of Gemma 4, I launched a new Space called MEDPAI — a medical imaging analysis tool that combines object detection with multimodal AI.
Here's how it works:

Upload a CT scan or X-ray
Computer vision models detect and annotate findings
Gemma 4 33B generates a report or answers your questions about the image

Currently available detectors: dental analysis and bone fracture detection.
More models are in the pipeline — follow the Space to stay updated!
alibidaran/MEDPAI

unmodeled-tyler

22 days ago

This is so cool! Awesome resource for people that might be interested in a bit more info about their results!

boroll2347

20 days ago

How does the Gemma 4 general-purpose model compare with the fine-tuned Medgemma 3 models?

The Gemma 4 models have the advantage of having an open-source license, but presumably lack the medical fine-tuning.

alibidaran

19 days ago

•

edited 4 days ago

According to the MedGemma documentation, MedGemma was trained on several medical datasets that form the foundation of its classification capabilities. In this project, I leveraged computer vision object detection models to annotate medical images, with Gemma 4 then interpreting those annotations to generate meaningful insights.
I'm currently gathering datasets and additional models to benchmark performance. Stay tuned — evaluations are coming soon!

In this post