We use cookies on this website. By continuing to use this site without changing your cookie settings, you agree that you are happy to accept our privacy policy and for us to access our cookies on your device.
[[Researchers at the University of Southern California are using machine learning to analyze brain scans to better predict diseases associated with aging. For VOA, Genia Dulot has the story from Los Angeles.]]
Content TypePackage
LanguageEnglish
Transcript/Script LogOn BRAIN MACHINE LEARNING
HEADLINE: LogOn: Researchers Analyzing Brain Scans to Predict Alzheimer’s Disease
TEASER: Scientists at the University of Southern California are using machine learning to analyze medical imaging
PUBLISHED AT: 05/16/2023 at 10:15AM
BYLINE: Genia Dulot
CONTRIBUTOR:
DATELINE: Los Angeles, California
VIDEOGRAPHER: Genia Dulot
PRODUCER:
SCRIPT EDITORS: Stearns, Reifenrath
VIDEO SOURCE (S): VOA, Skype
PLATFORMS (mark with X): WEB __ TV _X_ RADIO _X_
TRT: 1:59
VID APPROVED BY: mia
TYPE: TVPKG
EDITOR NOTES:
((INTRO:))
[[Researchers at the University of Southern California are using machine learning to analyze brain scans to better predict diseases associated with aging. For VOA, Genia Dulot has the story from Los Angeles.]]
((NARRATOR))
University of Southern California researchers are using magnetic resonance imaging, or MRI, scans to determine the biological age of a human brain, which may not always match a person’s chronological age, says Andrei Irimia, an assistant professor at USC Leonard Davis School of Gerontology.
((Andrei Irimia, USC Leonard Davis School of Gerontology))
“If a 45-year-old has the brain of a 55-year-old due to poor lifestyle choices, then that can also increase risk for disease because brains that are biologically older are typically a lot likelier to become ill and have higher risk for neurodegenerative disease, including Alzheimer’s disease.”
((NARRATOR))
The USC team used machine learning to develop a program trained on 4,800 MRI scans of healthy brains, which was then tested on 300 patients with neurogenerative diseases. Software author Paul Bogdan integrated magnifiers that zoom in on different brain regions into the program's deep-learning architecture.
((Paul Bogdan, USC Viterbi School of Engineering))
“The role of deep learning architecture is to look at each region in a three-dimensional space, see if that particular region contributes to predicting the age, and, if it does, it also quantifies how much it can help to predict the age of the brain.”
((NARRATOR))
Bogdan says that magnification allows for richer analysis of brain aging between men and women because each gender has different changes happening in different areas of the brain. Researchers say other medical uses for machine learning may be screening circulating blood cells to identify cancers.
((Genia Dulot, for VOA News, Los Angeles)
NewsML Media TopicsArts, Culture, Entertainment and Media
Program Name
NetworkVOA
Location (dateline)
Los Angeles
Embargo DateMay 16, 2023 00:18 EDT
Byline
Genia Dulot, for VOA News
Brand / Language ServiceVoice of America - English