VitaLonga

Human study · Aging · 15 Sept 2026

A 30-second counting task may capture aging signals blood tests and scans miss

Source: News-Medical

What the research found

Study (cross-sectional). Krongauz, Segal and colleagues at the Weizmann Institute of Science report in npj Aging (doi 10.1038/s41514-026-00519-x) on standardized 30-second recordings of 6,979 Israeli adults aged 40–70 counting from 1 to 30. Sex-stratified deep-learning models predicted chronological age with mean absolute errors of 3.95 years (women) and 4.41 years (men), explaining 54% and 44% of age variance. Adding Voice Age to eight molecular, imaging, physiological, and lifestyle clocks improved age prediction in 159 of 160 evaluations. Voice Age acceleration associated with greater adiposity, sleep-disordered breathing, and lower nocturnal oxygen saturation; cardiometabolic associations were mixed and not a uniform risk score. Hebrew-speaking single-site sample; not a screening test or an individual’s aging trajectory. Educational biomarker notes, not a diagnosis.

Evidence level: Human study. A study in people, but an association isn't proof of cause.

Read the original at News-Medical

More aging news