#12

Big Data May Lead to Earlier Alzheimer’s Diagnosis

A new algorithm charts a 30-year trajectory for biological risk factors.

By Linda Marsa|Thursday, December 15, 2016
DSC-B0217_04
DSC-B0217_04
A healthy brain (left) and a brain with Alzheimer's (right).
Alzheimer's Association

Diagnosing Alzheimer’s disease before irreversible damage is done is one of medicine’s major goals. Now, a Canadian team has uncovered a technique that may detect the mind-robbing disorder at its earliest stages.

Researchers at McGill University in Montreal examined more than 7,700 brain images from 1,171 people in various stages of Alzheimer’s progression, according to the study published in Nature Communications in June. They analyzed blood and cerebrospinal fluid samples, plus study participants’ cognitive skills. In 78 brain regions, they factored in the pattern of amyloid concentration, glucose metabolism, cerebral blood flow, functional activity and brain atrophy. Then, the team devised an algorithm to re-create a 30-year trajectory for each of these biological factors.

ScreenShot20161221at51954PM
ScreenShot20161221at51954PM
Thirty-year trajectories show Alzheimer's stages in patient brains: healthy state (HC), early/late mild cognitive impairment (E/LMCI) and late-onset Alzheimer's (LOAD). Each block shows a biological factor (white label), with colored bands depicting brain regions. Cooler colors indicate fewer abnormalities, warmer colors indicate more.
Y. Iturria-Medina et al./Nature Communications/ncomms11934/June 21, 2016

Scientists have long considered clumps of amyloid plaques — sticky, barnacle-like protein bundles — to be the first sign of Alzheimer’s. But the Canadian team’s analysis, which harnessed the power of big data to reveal previously undetected patterns, found that a marked decline of the brain’s blood flow is the earliest symptom.

“All the factors we looked at play an important role . . . but the vascular deterioration occurred first,” says Yasser Iturria Medina, the study’s lead author and a neuroscientist at McGill. “These findings could lead to better diagnostics and therapies that target each one of these factors.”

ADVERTISEMENT
Comment on this article
ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
DSC-V0817_01
+