20 Scientific American, May 2019
SCIENCE SOURCE
ADVANCES
HEALTH TECH
Alzheimer’s AI
Algorithm predicts eventual Alzheimer’s diagnoses from brain scans
An estimated 5.7 million people in the
U.S. have Alzheimer’s disease—the most
common type of dementia—and that num-
ber is expected to more than double by
- Early diagnosis is crucial for patients
to benefit from the few therapies available.
But no single assay or scan can deliver a
conclusive diagnosis while a person is alive;
instead doctors have to conduct numerous
clinical and neuropsychological tests. So
there is growing interest in developing
artificial intelligence to identify Alzheimer’s
based on brain imaging.
Researchers at the University of Califor-
nia, San Francisco, have now successfully
trained an AI algorithm to recognize one of
the early signs of Alzheimer’s—a reduction
in the brain’s glucose consumption—in posi-
tron emission tomography (PET) imaging.
The algorithm accurately predicted an
eventual Alzheimer’s diagnosis in nearly all
the test cases, according to the study.
In PET imaging, trace amounts of a
radioactive compound are ingested or
injected into the body, producing three-
dimensional images of metabolism, circu-
lation and other cellular activity. PET is
well suited for an AI diagnostic tool be
cause Alzheimer’s causes subtle changes
in the brain’s metabolism that begin years
before neural tissue starts to degrade,
says study co-author Jae Ho Sohn, a radi-
ologist at U.C.S.F. These changes are “very
hard for radiologists to pick up,” he notes.
The algorithm was trained and tested on
2,100 PET brain images from about 1,000
people 55 years and older. The images
came from a 12-year study that tracked
people who would ultimately be diagnosed
with Alzheimer’s, as well as those with mild
memory declines and healthy control sub-
jects. The algorithm was trained on 90 per-
cent of the data and tested on the remaining
10 percent. It was then retested on a sec-
ond, independent data set from 40 patients
monitored for 10 years. The algorithm was
highly sensitive and was able to recognize
81 percent of the patients in the first test
group and 100 percent in the second who
would be diagnosed with Alzheimer’s six
years later, on average. The findings were
published in February in Radiology.
The algorithm is based on “deep learn-
ing,” a machine-learning technique that uses
artificial neural networks programmed to
learn from examples. “This is one of the first
promising, preliminary applications of deep
learning to the diagnosis of Alzheimer’s,”
says Christian Salvatore, a physicist at Italy’s
National Research Council, who was not
involved in the study. “The model performs
very well when identifying patients with
mild or late” diagnoses, he says, but catch-
ing it in the earliest stages “remains one of
the most critical open issues in this field.”
— Rod McCullom
PET scans of normal ( left ) and Alzheimer’s ( right ) brains.
© 2019 Scientific American
20 Scientific American, May 2019
SCIENCE SOURCE
ADVANCES
HEALTH TECH
Alzheimer’s AI
Algorithm predicts eventual Alzheimer’s diagnoses from brain scans
An estimated 5.7 million people in the
U.S. have Alzheimer’s disease—the most
common type of dementia—and that num-
ber is expected to more than double by
- Early diagnosis is crucial for patients
to benefit from the few therapies available.
But no single assay or scan can deliver a
conclusive diagnosis while a person is alive;
instead doctors have to conduct numerous
clinical and neuropsychological tests. So
there is growing interest in developing
artificial intelligence to identify Alzheimer’s
based on brain imaging.
Researchers at the University of Califor-
nia, San Francisco, have now successfully
trained an AI algorithm to recognize one of
the early signs of Alzheimer’s—a reduction
in the brain’s glucose consumption—in posi-
tron emission tomography (PET) imaging.
The algorithm accurately predicted an
eventual Alzheimer’s diagnosis in nearly all
the test cases, according to the study.
In PET imaging, trace amounts of a
radioactive compound are ingested or
injected into the body, producing three-
dimensional images of metabolism, circu-
lation and other cellular activity. PET is
well suited for an AI diagnostic tool be
cause Alzheimer’s causes subtle changes
in the brain’s metabolism that begin years
before neural tissue starts to degrade,
says study co-author Jae Ho Sohn, a radi-
ologist at U.C.S.F. These changes are “very
hard for radiologists to pick up,” he notes.
The algorithm was trained and tested on
2,100 PET brain images from about 1,000
people 55 years and older. The images
came from a 12-year study that tracked
people who would ultimately be diagnosed
with Alzheimer’s, as well as those with mild
memory declines and healthy control sub-
jects. The algorithm was trained on 90 per-
cent of the data and tested on the remaining
10 percent. It was then retested on a sec-
ond, independent data set from 40 patients
monitored for 10 years. The algorithm was
highly sensitive and was able to recognize
81 percent of the patients in the first test
group and 100 percent in the second who
would be diagnosed with Alzheimer’s six
years later, on average. The findings were
published in February in Radiology.
The algorithm is based on “deep learn-
ing,” a machine-learning technique that uses
artificial neural networks programmed to
learn from examples. “This is one of the first
promising, preliminary applications of deep
learning to the diagnosis of Alzheimer’s,”
says Christian Salvatore, a physicist at Italy’s
National Research Council, who was not
involved in the study. “The model performs
very well when identifying patients with
mild or late” diagnoses, he says, but catch-
ing it in the earliest stages “remains one of
the most critical open issues in this field.”
— Rod McCullom
PET scans of normal ( left ) and Alzheimer’s ( right ) brains.
sad0519Adva3p.indd 20 3/21/19 3:21 PM
In
SCIENCE
We Trust
Join the nation’s
largest association of
freethinkers, atheists
and agnostics working
to keep religion
out of government
and social policy.
For a free sample of
FFRF’s newspaper,
Freethought Today:
Call 1-800-335-4021
ffrf.us/reason
FFRF.ORG
FFRF is a 501(c)(3) educational charity.
Deductible for income tax purposes.
Text FFRF to 52886 for
legislative action alerts
The World
is in Our Hands
Untitled-388 1 20/03/2019 17:36
Untitled-10 1 3/22/19 9:50 AM