By using this site, you agree to the Privacy Policy and Terms of Use.
Accept
MedsparkMedsparkMedspark
  • Home
  • News & Alerts
    News & AlertsShow More
    Ohio State and Microsoft build a hospital that senses itself
    By
    msadmin
    November 11, 2026
    Charta raises $55M to settle claims before doctors bill
    By
    msadmin
    November 11, 2026
    Oura spends $625M to put doctors behind its ring
    By
    msadmin
    November 11, 2026
    Google puts an agent layer under every hospital workflow
    By
    msadmin
    November 11, 2026
    XtalPi hands Pierre Fabre a molecule factory for stubborn cancer targets
    By
    msadmin
    November 10, 2026
  • Spotlight
    SpotlightShow More
    Robinhood Ventures Fund II brings retail capital to healthcare AI
    By
    msadmin
    August 13, 2026
    Adialante brings accessible MRI-based cancer screening
    By
    msadmin
    August 11, 2026
    CellType models biology so AI can discover drugs
    By
    msadmin
    August 11, 2026
    Healthcare AI startups in Robinhood Ventures Fund II
    By
    msadmin
    August 11, 2026
    OpenAI launches GPT-Rosalind for life sciences research
    By
    msadmin
    July 17, 2026
  • Articles
    ArticlesShow More
    Singapore clinics let an AI watch for referrals nobody ever sent
    By
    msadmin
    November 1, 2026
    An AI team of agents grades immunotherapy side effects
    By
    msadmin
    October 30, 2026
    Brain implant decoders fail when patients or hardware change, review warns
    By
    msadmin
    October 25, 2026
    Proteomics model picks breast cancer drugs from biopsies
    By
    msadmin
    September 10, 2026
    CRISP model reads frozen sections to steer cancer surgery
    By
    msadmin
    September 10, 2026
  • About
    • Mission
    • Services
    • Contact
  • Shop
    • All Items
    • By Category
    • Cart
  • Newsletter
Font ResizerAa
MedsparkMedspark
Font ResizerAa
  • Home
  • News & Alerts
  • Spotlight
  • Articles
  • About
  • Shop
  • Newsletter
  • Home
  • News & Alerts
  • Spotlight
  • Articles
  • About
    • Mission
    • Services
    • Contact
  • Shop
    • All Items
    • By Category
    • Cart
  • Newsletter
Follow US
News & Alerts

Google’s TB screening AI reaches 4 kilometres underground

Offline chest X-ray models from Google Research now triage miners four kilometres below the surface in South Africa, with suspected cases routed to molecular testing.

MedSpark Staff
By
msadmin
MedSpark Staff
Bymsadmin
Medical, Healthcare, & Biotech/Pharma AI News
Follow:
Published: October 24, 2026
Share
2 Min Read
SHARE

Google’s TB screening AI reaches 4 kilometres underground

Google Research has taken chest X-ray screening AI to a place where no radiologist will ever commute: the Mponeng gold mine in South Africa, roughly 4 kilometres below the surface.

The 23 October research report describes how the company worked with the Stop TB Partnership, Harmony Gold and South Africa’s Department of Health to screen miners on site. Ultra-portable digital X-ray units feed an offline version of the model, so a diagnosis does not depend on a network connection that does not exist at that depth.

Miners flagged by the system are referred for molecular testing rather than told they have tuberculosis. That distinction is central. A screening hit is a prompt to confirm, not a result.

The setting explains why Google bothered. Silica dust raises respiratory risk underground, and the trip from a deep shaft to a clinic can cost a working day or more. Moving imaging and triage into the shaft shortens the route from cough to laboratory.

Google frames the mine as one point on a longer research path, from early chest-radiograph models toward tools other health systems can adapt rather than a bespoke product for one employer.

The honest measure of success is not how many X-rays the algorithm reads. It is how many suspected cases actually reach confirmatory testing and treatment. Google has not published that number for Mponeng, and until it does the deployment stands as a promising use case rather than proof of impact.

For health systems weighing AI screening in low-connectivity settings, the design lesson is portable. Offline inference, ultra-portable hardware and a referral pathway matter as much as model accuracy.

TAGGED:AI screeningchest X-rayglobal healthGoogle ResearchMedical Imagingtuberculosis
SOURCES:TechBooky
Share This Article
Facebook Copy Link Print
MedSpark Staff
Bymsadmin
Follow:
Medical, Healthcare, & Biotech/Pharma AI News

You Might Also Like

News & Alerts

Two in three nurses say hospital AI adds work, not relief

By
msadmin
September 23, 2026
News & Alerts

Sopris buys Blackford’s imaging AI out of Bayer’s portfolio

By
msadmin
October 13, 2026
News & Alerts

Pharmacy AI suite at Singapore’s NUHS reclaims 850 staff hours a week

By
msadmin
August 1, 2026
News & Alerts

Europe approves Vara’s AI to clear normal mammograms on its own

By
msadmin
September 5, 2026

AI news, analysis, and insights for healthcare, biotech, and pharma.

Facebook Twitter Youtube Linkedin
Quick Links
  • News & Alerts
  • Articles
  • Spotlight
  • Events
About Medspark
  • Mission
  • Services
  • Contact

© Copyright 2026 MedSpark. All rights reserved.

Privacy Policy | Legal