Bringing foundational models to the front lines of TB screening
Web page · https://research.google/blog/bringing-foundational-models-to-the-front-lines-of-tb-screening
Generative AI
First seen by Oriel on .
AI summary
Google describes a progression from validated AI chest X-ray screening to open multimodal models and local deployments intended to expand tuberculosis detection in underserved settings.
- A prospective multi-site validation reported high sensitivity across diverse populations, and Nexus AI translated the algorithms into a CE MDR Class IIb-certified clinical tool.
- The tool operates in about 40 facilities across six countries; clinicians have assessed over 25,000 radiographs, with turnaround reduced from weeks to under 45 seconds.
- MedSigLIP supports image classification and search with plain-language labels, while MedGemma models can combine radiographs with clinical history and run locally without internet access.
- Indonesia’s Ministry of Health is fine-tuning MedGemma and MedSigLIP on domestic X-rays to support annual screening of 50 million citizens while keeping patient data within national borders.
- HeAR, trained on over 300 million audio segments, enables cough-based screening using standard smartphone microphones as a complement to imaging.
#tuberculosis #medical imaging #medgemma #bioacoustics
Generated by gpt-5.6-sol. Check the original for details.
Aliases
| Source | External ID | URL |
|---|---|---|
| Google Research Blog | https://research.google/blog/bringing-foundational-models-to-the-front-lines-of-tb-screening/ | https://research.google/blog/bringing-foundational-models-to-the-front-lines-of-tb-screening/ Read discussion |