New Study Uses MRI Scans to Predict Alzheimer's and Dementia

“We showed that a single MRI scan can predict dementia on average 2.6 years before memory loss is clinically detectable, which could help doctors advise and care for their patients.”

Alzheimer's is a destructive and progressive brain disorder that destroys memory and thinking skills. 

Alzheimer's is currently expensive and complicated to test for

More than 5 million Americans are affected by the disorder according to the National Institutes of Health (NIH).

Currently, a patient’s risk of developing Alzheimer’s dementia is measured using the Mini-Mental State Examination questionnaire, or by testing for the high-risk form of the gene ApoE, which increases a person’s risk of Alzheimer’s by up to 12-fold. 

These tests return around a 70- 80 percent accuracy. Other current tests include using PET scans to look for plaques of Alzheimer’s proteins in the brain.

However, this test can cost thousands of dollars and require radioactive materials not found in a typical hospital.

MRI scans are widely available in most hospitals. The scans which give doctors a look inside a patients brains were analyzed by the research team for physical signs of impending cognitive decline. 

Research leveraged inter-institution collaboration 

The researchers used a technique known as diffusion tensor imaging to look for the signs of health in the brain’s white matter which includes the cables in the brain that enables different parts of the brain to talk to one another. 

“Diffusion tensor imaging is a way of measuring the movement of water molecules along white matter tracts,” Raji said.

“If water molecules are not moving normally it suggests underlying damage to white tracts that can underlie problems with cognition.” 

[see-also]

The research used information from the Alzheimer’s Disease Neuroimaging Initiative that pools funding, data, and expertise to improve clinical trials for Alzheimer’s disease.

Raji and his team identified 10 people whose cognitive skills noticeably declined over a two year period and matched them with people the same age and gender whose skills had not declined. 

They then compared diffusion tensor MRI scans taken just before the two-year period for all 20 people.

The participants who went on to experience cognitive decline had evidence of damage to their white matter. 

The analysis was repeated in another smaller sample of people using more refined measures of white matter integrity.

By using this new analysis the researchers were able to predict cognitive decline with 89 percent accuracy when looking at the whole brain. 

Predictive test returned 95% accuracy

If just specific parts of the brain were focussed on the accuracy increased to 95%.

“We could tell that the individuals who went on to develop dementia have these differences on diffusion MRI, compared with scans of cognitively normal people whose memory and thinking skills remained intact,” Raji explained.

“What we need now, before we can bring it into the clinic, is to get more control subjects and develop computerized tools that can more reliably compare individual patients’ scans to a baseline normal standard.

With that, doctors might soon be able to tell people whether they are likely to have Alzheimer’s develop in the next few years.”

The findings were presented on Sunday, Nov. 25 at the Radiological Society of North America meeting in Chicago.