Skip to main content

Machine learning may be a key in helping diagnose Alzheimer's earlier

facebook typing with your brain
Image used with permission by copyright holder
Alzheimer’s is a devastating chronic neurodegenerative disease that currently affects about 5.4 million people in the U.S. alone. Alzheimer’s patients suffer progressive mental deterioration, which eventually impairs even basic bodily functions like walking and swallowing.

While Alzheimer’s can increasingly be managed, one of the big challenges of the disease is early diagnosis. MRI machines can be used to confirm advanced cases, but by the time the disease has reached this stage, brain tissue is gone and there is no way to restore it.

Could machine-learning tools be used to help detect and identify Alzheimer’s disease before it is currently possible to do so? That is the mission of researchers from VU University Medical Centre in Amsterdam, led by Dr. Alle Meije Wink.

In a newly published study, Wink and colleagues used machine-learning tools to recognize early patterns in a special type of MRI scan that shows how much blood is reaching different parts of the brain. The researchers analyzed information from 100 patients who likely have Alzheimer’s disease, 60 patients with mild cognitive impairment, 100 with subjective cognitive decline, and a control group of 26 healthy individuals. The machine-learning tool they created was not only able to classify existing patients based on their different levels of cognitive impairment, but was also able to predict Alzheimer’s disease in previously undetected cases.

“We envision our project leading to a tool to assist the neuroradiologist who assesses a scan by providing a predicted patient group label, ideally with a confidence indication, to help and speed up the diagnostic process,” Wink tells Digital Trends. “Furthermore, we think that multimodal data analyses, which integrate imaging data from multiple sources, can improve and enhance the classification by providing higher accuracies and more comprehensive backgrounds about the nature of the diagnosis.”

The end result could be an important advance in not only machine learning, but potentially life-altering research into Alzheimer’s, too.

Editors' Recommendations

Luke Dormehl
I'm a UK-based tech writer covering Cool Tech at Digital Trends. I've also written for Fast Company, Wired, the Guardian…
Deep-learning A.I. is helping archaeologists translate ancient tablets
DeepScribe project 1

Deep-learning artificial intelligence is helping grapple with plenty of problems in the modern world. But it also has its part to play in helping solve some ancient problems as well -- such as assisting in the translation of 2,500-year-old clay tablet documents from Persia's Achaemenid Empire.

These tablets, which were discovered in modern-day Iran in 1933, have been studied by scholars for decades. However, they’ve found the translation process for the tablets -- which number in the tens of thousands -- to be laborious and prone to errors. A.I. technology can help.

Read more
Alzheimer’s stole her ability to read. An Alexa-powered chair gave it back
alexa powered magic chair  9

Making of Jacintas Chair 1

If you’ve got a smart speaker like an Amazon Echo or Google Home in your house, there’s a good chance that you’ve come to rely on it in ways that you never expected, whether that’s setting alarms in the kitchen or controlling your smart home gadgets. But it’s probably not been quite such a meaningful innovation for you as the customized Alexa-powered chair built for 69-year-old Alzheimer’s patient Jacinta Dixon.

Read more
The latest weapon in the fight against Alzheimer’s? Flickering lights
Alzheimers research glasses

Alzheimer’s is a lot of things, but it’s most notable for what it’s not. It’s a neurodegenerative disease which affects an estimated 5.5 million people in the United States alone. It’s a debilitating, life-changing disorder that’s intensely cruel in its effects, both to the immediate sufferer and their loved ones. It destroys almost one-third of a brain’s mass, gouging out enormous ravines while piling up heaps of junk amyloid plaques between the neurons. It is not curable.

At least, not yet.

Read more