CANCELLED Tissue-level Prediction of Biological Age and Pathology Incidence

In this talk I will address current challenges in understanding of cellular changes leading to tissue-specific loss of function during aging in humans. Utilizing large scale datasets (over 25,000 from 40 tissues) across the human lifespan, we employ deep learning to quantify tissue features, including cellular morphology, expression, and microanatomical organization. This cross anatomical approach provides insights into archetypal molecular and cell-level changes during healthy human aging. Our framework allows the development of 'tissue-clocks'-predictors of biological age from tissue images, decoupling chronological and biological age and validated by associations with telomere lengths and sub-clinical incidence of pathology. This research offers a human centric perspective on tissue aging, establishing healthy ranges and enabling early detection of age-associated diseases.