Tuesday, November 28 | 10:00 a.m.-10:10 a.m. | T3-SSMK05-4 | Room E352
In this Tuesday morning talk, researchers will describe how a deep-learning algorithm’s assessment of biological age on dual-energy x-ray absorptiometry (DEXA) exams can facilitate morbidity and mortality predictions.
A group led by presenter Alex Cheng sought to determine if AI could be used to estimate biological age on DEXA images and if this measure was associated with cardiovascular disease and all-cause mortality. They then used lateral and lumbar spine DEXA data from nearly 45,000 participants in the UK Biobank to train a convolutional neural network to produce what they call a Spinal-Age score, an estimate of biological age.
For the purposes of the study, the primary outcome was a composite of incident myocardial infarction, stroke, and all-cause mortality over a median of 2.68 years of follow-up, according to the researchers. After training, the model was then tested on an additional 8,951 UK Biobank participants, of which 2.8% had the primary outcome.
They found that the algorithm’s biological age estimate was associated with all-cause mortality and car diovacular events, beyond prevalent risk factors and spinal bone mineral density.
“Spinal DXA [exams] are a common test, routinely administered for osteoporosis screening,” Cheng et al wrote. “The Spinal-Age score may enable opportunistic screening for other age-related diseases.”
Sit in on this Tuesday morning session for more information.