Tissue shape predicts telomere shortening in human tissues

Routine tissue morphology may contain signals that predict bulk-tissue telomere shortening across human organs.

A deep learning framework was trained on more than 5,000 whole-slide images from 919 individuals across 18 organs and tested telomere-length prediction in 11 tissues and about 2,800 GTEx biopsies.

The evidence is an imaging-model and validation study, so performance depends on the sampled tissues, bulk telomere labels, and model interpretation rather than direct clinical outcome testing.