Google Research announced the deployment of its foundational AI models for tuberculosis screening across healthcare facilities in Africa, marking a significant milestone in bringing medical diagnostic AI to regions where it is needed most. In partnership with Nexus AI, a Pretoria-based medical technology company, the tool screens chest X-rays for signs of TB in approximately 45 seconds—reducing diagnostic turnaround times from weeks. The solution has achieved regulatory certification including CE MDR Class IIb designation and ISO 13485 compliance, enabling deployment across six countries: South Africa, Sierra Leone, Côte d'Ivoire, the Democratic Republic of the Congo, Zambia, and Cameroon.
Since deployment began, the AI screening tool has been evaluated on more than 25,000 chest radiographs by over 90 clinicians across approximately 40 healthcare facilities. Google's collaboration prioritizes health equity, having delivered more than 3,000 free screenings to under-resourced public clinics with a target of 10,000 sponsored screenings annually. The deployment addresses a critical diagnostic gap: tuberculosis remains one of the world's deadliest infectious diseases, claiming 1.23 million lives annually, yet millions of cases go undetected in remote and under-resourced regions where radiologists and imaging infrastructure are scarce. The effort includes ultra-portable digital X-ray kits and offline AI capabilities enabling screening even in extreme environments like deep-shaft mining operations.
Key Points
Google's AI models screen chest X-rays for tuberculosis in ~45 seconds, reducing turnaround time from weeks
Deployed across ~40 healthcare facilities in six African countries with 25,000+ radiographs evaluated to date
Partner Nexus AI achieved CE MDR Class IIb certification and ISO 13485 compliance for clinical deployment
3,000+ free screenings delivered to under-resourced clinics with goal of 10,000 annual sponsored screenings
Solution addresses critical diagnostic gap where TB kills 1.23 million annually yet many cases remain undetected in remote regions