Discussion around Health AI Developer Foundations (HAI-DEF) open models.
Get started adapting models for health applications:
- Find models on Hugging Face
- Get started with documentation
Join the community and start conversations around:
11 Likes
Thats amazing, thank you 
3 Likes
THIS IS AMAZING.CAN I ASK FOR OR SUGGESTS FOUNDATIONAL MODELS OR STARTING POINTS THAT MIGHT BE USED IN OTHER MEDICATIONS? Specifically for medical use cases, I was considering Pharmacokinetic models i.e cannabinoid interactions as a neurological and physiologic predictor that could help patients with ALS, Alzheimer’s, curing pain, depression, etc.
At the moment the closest Google open models that I can think of are TxGemma, AlphaFold and AlphaGenome. Hopefully, there will be more. We encourage you to contribute to open model community, if you have relevant ongoing research.
1 Like
Thought I should share that I recently won an ambient AI hackathon deploying MedSigLIP image classification over a webcam livestream via WebRTC. I was hitting sub 50ms latencies, which would make this ideal for telemedicine applications and alert systems on scopic procedures.
4 Likes
Thank you for using MedSigLIP and sharing this great use case! Way to go!
1 Like
Is this model suitable for Dental?
1 Like
Hi @ramkumarkoppu MedGemma and MedSigLIP has not been trained on dental data. You could give it a try fine-tuning the model with your dental dataset to evaluate the performance, noting that these models are not optimized for this. Our technical report has more information and you can visit our website and ask our AI assistant for answers to questions: Health AI Developer Foundations | Google for Developers
Hello Health AI Developer Foundations Forum,
I am a current PharmD-MBA Graduate at The University of Georgia set to graduate in May of 2026. I am passionate about healthcare-technology and looking to establish a professional presence on various forums like Stack Overflow. My goal is to bridge the gap between technical literacy and health literacy as it costs nothing to help however I do not want to appear as “pandering” and incompetent. What are some of the most recommended platforms for tech enthusiasts to use, share ideas and avoid the “Echo Chamber" effect. I love information technology and theory but understand how dangerous confirmation bias can be. All in all, I want to ensure any information provided on these public forms would not cause patient harm since “Nonmaleficence” is the primary principle of healthcare but want to create solutions as an engineer would using the field of information theory to improve clinical outcomes.
Looking forward to growing in the Health-Tech field and providing my thoughts in situations others find useful!
Benjamin Brewster