Rendeiro Team Builds AI Blood Test for 29 Organs' Biological Age

2 articles · Updated · ScienceAlert · Aug 22

25,712 tissue images from 983 deceased donors let researchers train organ-specific “tissue clocks,” then link those age estimates to blood gene-activity patterns to infer whether an organ looks older or younger than expected.
About five years was the average gap between tissue-model estimates and actual age, and older-looking tissues also showed hallmarks of aging such as shrinkage, scarring, vessel loss, fat buildup in muscle, and nerve thinning.
Eight disease groups broadly matched the organ-age signals: stroke patients showed the largest brain age gap, Crohn’s disease patients had elevated gaps across the digestive tract, and Alzheimer’s patients showed a significant increase only in the brain.
The blood-only model performed best for whole-body aging, the digestive system and the spleen, but the study cannot show whether organ aging causes disease or whether the test can spot illness before symptoms.
Clinical use is still years away—possibly close to a decade—because the work relied largely on post-mortem tissue, lacked long-term follow-up, and still needs prospective validation across populations and regulatory approval.