Deep learning predicts all-cause mortality from longitudinal total-body DXA imaging

Yannik Glaser; John Shepherd; Lambert Leong; Thomas Wolfgruber; Li-Yung Lui; Peter Sadowski; Steven R. Cummings

doi:10.1038/s43856-022-00166-9

Communications Medicine (Aug 2022)

Deep learning predicts all-cause mortality from longitudinal total-body DXA imaging

Yannik Glaser,
John Shepherd,
Lambert Leong,
Thomas Wolfgruber,
Li-Yung Lui,
Peter Sadowski,
Steven R. Cummings

Affiliations

Yannik Glaser: Information and Computer Sciences, University of Hawai’i at Mānoa
John Shepherd: University of Hawai’i at Mānoa Cancer Center
Lambert Leong: University of Hawai’i at Mānoa Cancer Center
Thomas Wolfgruber: University of Hawai’i at Mānoa Cancer Center
Li-Yung Lui: San Francisco Coordinating Center, California Pacific Medical Center Research Institute
Peter Sadowski: Information and Computer Sciences, University of Hawai’i at Mānoa
Steven R. Cummings: San Francisco Coordinating Center, California Pacific Medical Center Research Institute

DOI: https://doi.org/10.1038/s43856-022-00166-9
Journal volume & issue: Vol. 2, no. 1
pp. 1 – 12

Abstract

Read online

Glaser et al. develop a deep learning system to predict all-cause mortality from total-body DXA scans. Their best predictive model integrates longitudinal body composition data with traditional mortality risk factors.

Published in Communications Medicine

ISSN: 2730-664X (Online)
Publisher: Nature Portfolio
Country of publisher: United Kingdom
LCC subjects: Medicine
Website: https://www.nature.com/commsmed/

About the journal