Artificial intelligence algorithm aids earlier Alzheimer’s diagnosis

06 November 2018

Radiology: A Deep Learning Model to Predict a Diagnosis of Alzheimer Disease by Using 18F-FDG PET of the Brain

Scientists from the University of California have used a deep learning algorithm alongside sophisticated positron emission tomography (PET) scans to help identify people with Alzheimer’s disease or mild cognitive impairment (MCI), on average of 76 months before diagnosis. The results are published today (6 November) in the journal Radiology.

PET scans for detecting markers of Alzheimer’s work through the injection of a compound that binds to a specific protein inside the brain before a person undergoes a PET brain imaging scan. Because this compound includes a radio-labelled ‘tag’ that can be picked up by the scan, it’s then possible to see how much of the marker of interest is present, and where it is located in the brain.

The US team of researchers used PET scans from people enrolled in an existing database to look at glucose levels in the brain, a marker that sheds light on brain metabolism. The researchers followed these individuals to see if they were diagnosed with Alzheimer’s disease or MCI and used 90 percent of the scans to train a deep learning algorithm about which scans could best predict a diagnosis.

The scientists tested the deep learning algorithm on the remaining 10% of PET scans from the existing database and additionally 40 people who the researchers recruited themselves. The researchers found that in the additional group of people, the machine learning technique was able to tell with 100% accuracy if someone would get a diagnosis of Alzheimer’s or MCI.  The algorithm was 82% accurate in cases where it decided the PET scan was not indicative of the disease.

Dr Carol Routledge, Director of Research at Alzheimer’s Research UK, said:

“The diseases that cause dementia begin in the brain up to 20 years before any symptoms start to show, presenting a vital window of opportunity for us to intervene before widespread damage occurs. This study highlights the potential of machine learning to assist with the early detection of diseases like Alzheimer’s, but the findings will need to be confirmed in much larger groups of people before we can properly assess the power of this approach.

“Currently in the UK, the use of PET scanning is mainly limited to research studies and clinical trials, to ensure that potential new medicines are tested in the right people. PET scans are a powerful tool, but they are expensive and require specialist facilities and expertise.

“Although recent advances in artificial intelligence offer a critical opportunity for tackling the challenge of diagnosing Alzheimer’s disease much earlier, this will require a major effort and significant investment. Alzheimer’s Research UK believes a technological approach, using big data and machine learning, could lead to huge benefits for people affected by dementia and their families.”