Fifth-year PhD student Pengyu “Steven” Kan received multiple distinctions for his research on biological brain age at the 150th Annual Meeting of the American Neurological Association, held September 13–16 in Baltimore. The conference is a premier platform for showcasing groundbreaking scientific discoveries and talks from pioneers in the field.
Kan’s work, “Multiview Transformer for Brain Age Prediction,” was recognized with a Poster Award, which recognizes exceptional work presented by those in the early stages of their careers, and was additionally selected for an oral presentation, a distinction afforded to only five speakers out of the hundreds of abstracts submitted.
His research—conducted under the advisement of Craig Jones, an assistant research professor of computer science, and Kenichi Oishi, a professor of radiology and radiological science and neurology and neurosurgery at the School of Medicine—uses a deep learning model to predict biological brain age and identify critical regions of the brain associated with the aging process.
The model uses MRI brain scans and measurements of 280 different brain regions to learn patterns that relate to aging and can help predict a person’s biological brain age. By comparing people’s brain age with their actual age, Kan and his team were able to determine how this brain-age “gap” relates to cognitive function and brain disorders such as Alzheimer’s disease.
Next, the researchers plan on conducting a longitudinal study to observe how brain aging progresses in individuals over time and to explore how the brain-age gap relates to people’s lifestyle, environment, comorbidities, and genetics. By analyzing these factors, they aim to identify modifiable factors that could help reduce the risk of cognitive decline and neurodegeneration.