By Aine Cryts
An inability to remember events, problems finishing daily tasks, mood changes—such as depression or personality changes—and confusion about physical location or the passage of time. Those are four of the early signs and symptoms of Alzheimer’s disease, researchers at Mayo Clinic reveal. Approximately 5.8 million Americans are living with Alzheimer’s, according to the Alzheimer’s Association. The key to treating Alzheimer’s is diagnosing it early.

“The hope is, future treatments could then target the disease in its earliest stages, before irreversible brain damage or mental decline has occurred,” says the Alzheimer’s Association. Enter a November 2018 study co-authored by Jae Ho Sohn, MD, a radiology resident at the University of California at San Francisco.

Published in Radiology, the study revealed that an artificial intelligence (AI) algorithm used by Sohn’s team can be used to diagnose early-stage Alzheimer’s about six years prior to the final diagnosis. The AI algorithm uses a fluorine 18 fluorodeoxyglucose PET scan of the brain. Radiology editor David A. Bluemke, MD, described Sohn’s research as “groundbreaking” in an announcement.

“It demonstrates the potential for state-of-the-art AI technology to help interpret the remarkably complex patterns on FDG PET imaging in patients with Alzheimer’s disease. Even more hopeful for Alzheimer’s patients, this same line of research will also apply to more specific PET tracers for detecting the disease,” he continues.

Alzheimer’s is a “very, very serious problem,” Sohn tells AXIS Imaging News. That’s particularly the case since the disease affects aging baby boomers. He cites the emotional and financial burdens of caring for loved ones with Alzheimer’s. A 2019 study revealed that the cost to the U.S. economy of treating dementia, which includes Alzheimer’s, is $290 billion annually.

The patients in the dataset used by Sohn’s team were 55 to 93 years of age, thus the AI algorithm is optimized to analyze patients in this age group. He tells AXIS the AI algorithm may work with other age groups, but that will require external validation.

Sohn received the Radiological Society of North America’s eighth Alexander R. Margulis Award for Scientific Excellence at the organization’s annual meeting in December.

Aine Cryts is a contributing writer for AXIS Imaging News.