A newly developed machine learning-based test could diagnose autoimmune diseases faster The proof-of-concept study analyzed the immune system’s built-in records of disease exposures

https://www.statnews.com/2025/02/20/machine-learning-immune-cells-test-diagnose-auto-immune-diseases/?utm_campaign=morning_rounds&utm_medium=email&_hsenc=p2ANqtz-8qbSiRYwj0MPRiM8IbxlGH2y-QrU9NjQ0A-pynqcK4Rc__ZgXaUtWWHctzEt0S5H0RSyflPe1VPrk4WZ2JOl3jF_CJbQ&_hsmi=348291083&utm_content=348291083&utm_source=hs_email The body keeps the score, but we don’t check it Every time you get sick, your body makes a tally. Researchers unveiled a new, machine learning-based tool Thursday that can diagnose Covid-19, HIV, flu, lupus, and type-1 diabetes based on the immune system’s built-in records of disease exposures: B and T cells. The proof-of-concept study is one of the first to combine B and T cell analysis and potentially paves the way for tests that could better diagnose autoimmune diseases and help researchers unravel the biological underpinnings of complex conditions. The tool is not ready for clinical use and the general idea is not new, one expert said, but it is the biggest and broadest study of the concept and will help other researchers determine which genes are driving immune response. Read more about the fascinating science from STAT’s Jonathan Wosen and Brittany Trang.