<br />

These applications are developed and owned by third-party developers and manufacturers. They are built using HAI-DEF open models, which are subject to the HAI-DEF license. Real-world clinical applications [Tell us about your application](https://services.google.com/fb/forms/hai-defshowcase)

## Real-world clinical applications

[![DawaMom AI Cervical Cancer Screening](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/dawamom/dawamom-hero.jpg)
### Dawa Health
### Point-of-care cervical cancer screening in Zambia, powered by MedSigLIP \| Read More](https://developers.google.com/health-ai-developer-foundations/showcase/dawamom) [![Visilant Seeker AI Screening](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/visilant/visilant-hero.png)
### Visilant
### Expanding access to eye care in India, built with MedGemma \| Read More](https://developers.google.com/health-ai-developer-foundations/showcase/visilant) [![EaseHealth Rural Uganda Deployment](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/crane-ai/crane-hero.jpg)
### Crane AI Labs
### Offline MedGemma deployment for health workers in rural Uganda \| Read More](https://developers.google.com/health-ai-developer-foundations/showcase/crane-ai) [![Nimblemind Pain Spike Prediction](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/nimblemind/nimblemind-hero-upscaled.png)
### Nimblemind
### Predicting pain spikes in patients with opioid use disorder \| Read More](https://developers.google.com/health-ai-developer-foundations/showcase/nimblemind) [![askCPG](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/ask-cpg/heroshot.png)
### AskCPG by Qmed Asia
### Elevating national guideline inquiries with MedGemma's multimodal reasoning \| Read More](https://developers.google.com/health-ai-developer-foundations/showcase/askcpg) These solutions and tools are created by third-party developers using HAI-DEF open models, which are subject to the HAI-DEF license. The HAI-DEF program endorsed the linked write-up as of the stamped date and disclaims any obligation to monitor, review, or endorse future revisions of the material. Technical solutions and tools [Tell us about your solution](https://services.google.com/fb/forms/hai-def-showcase/)

## Technical solutions and tools

[![](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/sp-logo.jpg)
### Verifiable fine-tuning by Super Protocol
### May 27, 2026
Solving the Data Paradox: Verifiable Fine-Tuning of MedGemma 27B on Private Clinical Dialogues, Read More](https://superprotocol.com/case-studies/confidential-medgemma27b-tuning) [![](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/sp-logo.jpg)
### Deployment by Yma Health \& Super Protocol
### May 27, 2026
Yma Health Deployed AI on Real Patient Data Across Clinics and EHR Platforms, Read More](https://superprotocol.com/case-studies/yma) These prototypes are built during various community challenges we sponsor to show the art of the possible.Note that these are prototypes, and additional development and testing will be required prior to any use that impacts care. Community prototypes 🏆 Winners of the MedGemma Impact Challenge 🏆 Drawing [878 submissions](https://www.kaggle.com/competitions/med-gemma-impact-challenge/writeups), this Kaggle challenge showcased incredible community engagement.  
Read more in [our blog post.](https://blog.google/innovation-and-ai/technology/health/med-gemma-impact-challenge/)

## Community prototypes

[### Epicast
### Epicast - 1st Place \| Explore](https://www.kaggle.com/competitions/med-gemma-impact-challenge/writeups/new-writeup-1770830823676) [### Sunny
### Sunny - 2nd Place \| Explore](https://www.kaggle.com/competitions/med-gemma-impact-challenge/writeups/sunny-private-skin-health-tracker) [### FieldScreen AI
### FieldScreen AI - 3rd Place \| Explore](https://www.kaggle.com/competitions/med-gemma-impact-challenge/writeups/fieldscreen-ai) [### Tracer
### Tracer - 4th Place \| Explore](https://www.kaggle.com/competitions/med-gemma-impact-challenge/writeups/new-writeup-1771923230850) [### ClinicDX
### ClinicDX - Novel Task Track \| Explore](https://www.kaggle.com/competitions/med-gemma-impact-challenge/writeups/new-writeup-1771522062957) [### UniRad3s
### UniRad3s - Novel Task Track \| Explore](https://www.kaggle.com/competitions/med-gemma-impact-challenge/writeups/UniRad3s-reimagining-radiology-with-spot-segment) [### BridgeDx
### BridgeDx - Edge AI Track \| Explore](https://www.kaggle.com/competitions/med-gemma-impact-challenge/writeups/new-writeup-1771945306600) [### CaseTwin
### CaseTwin - Agentic Workflow Track \| Explore](https://www.kaggle.com/competitions/med-gemma-impact-challenge/writeups/new-writeup-1771703368723) [### BigTB6
### BigTB6 - Agentic Workflow Track \| Explore](https://www.kaggle.com/competitions/med-gemma-impact-challenge/writeups/bigtb6) These first-party demo applications are designed to showcase what can be built with HAI-DEF tools, and they include interactive user interfaces and full code. They are for illustrative purposes and don't represent finished products. Any real-world applications would require additional development, training, and adaptation. The experiences highlighted in these demos showcase the baseline capabilities of HAI-DEF models for the displayed tasks, and are intended to inspire further innovation. Our demo applications

## Our demo applications

[![](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/ehr_nav.gif)
### EHR navigation
### EHR navigation with MedGemma \| Explore](https://huggingface.co/spaces/google/ehr-navigator-agent-with-medgemma) [![](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/appoint_ready.gif)
### Pre-visit patient intake
### Pre-visit patient intake with MedGemma \| Explore](https://huggingface.co/spaces/google/appoint-ready) [![](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/rad_explain.png)
### Radiology explainer
### Radiology explainer with MedGemma \| Explore](https://huggingface.co/spaces/google/rad_explain) [![](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/rad_learning.png)
### Radiology learning companion
### Radiology learning companion with MedGemma \| Explore](https://huggingface.co/spaces/google/rad_learning_companion) [![](https://developers.google.com/static/health-ai-developer-foundations/showcase/images/rad_extract.gif)
### Radiology report structuring
### Radiology report structuring with LangExtract \| Explore](https://huggingface.co/spaces/google/radextract) [![](https://developers.google.com/static/health-ai-developer-foundations/images/path-foundation-logo.png)
### Digital pathology classification
### Digital pathology outlier tissue detection \& classification \| Explore](https://huggingface.co/spaces/google/path-foundation-demo) [![](https://developers.google.com/static/health-ai-developer-foundations/images/cxr-foundation-logo.png)
### Chest X-Ray classification
### Chest X-Ray classification \| Explore](https://huggingface.co/spaces/google/cxr-foundation-demo)