Colorised brain MRI

AI could detect Alzheimer's disease from brain scans

Image credit: Dreamstime

Researchers at Massachusetts General Hospital (MGH) have developed an accurate Alzheimer's detection method that relies on routinely collected clinical brain images.

Artificial intelligence (AI) could help clinicians identify patients who would benefit from the treatment of Alzheimer's disease. 

The team at MGH used deep learning a type of machine learning and artificial intelligence that uses large amounts of data and complex algorithms to train models and improve detection of the brain disease. 

The AI was trained on brain magnetic resonance images (MRIs) collected from patients with and without Alzheimer’s disease who were seen at MGH before 2019.

Next, the group tested the model across five datasets – MGH post-2019, Brigham and Women’s Hospital pre- and post-2019, and outside systems pre- and post-2019 – to see if it could accurately detect Alzheimer’s disease based on real-world clinical data, regardless of hospital and time.

Overall, the research involved 11,103 images from 2,348 patients at risk for Alzheimer’s disease and 26,892 images from 8,456 patients without Alzheimer’s disease.

The results showed the AI was able to detect Alzheimer’s disease risk with 90.2 per cent accuracy and regardless of other variables, such as age.

“Alzheimer's disease typically occurs in older adults, and so deep learning models often have difficulty in detecting the rarer early-onset cases,” said Matthew Leming, PhD, an investigator on the team. 

“We addressed this by making the deep learning model ‘blind’ to features of the brain that it finds to be overly associated with the patient's listed age.”

A key aspect of the research was the fact that the AI could detect Alzheimer's disease from scans that were very different from those it was trained on. 

This was not the case of similar AIs developed in the past, which often failed to recognise MRIs collected on a scanner manufactured by a different company from that of the scans in the dataset it was trained on. 

The model used an uncertainty metric to determine whether patient data were too different from what it had been trained on for it to be able to make a successful prediction.

“This is one of the only studies that used routinely collected brain MRIs to attempt to detect dementia," said Leming. 

"While a large number of deep learning studies for Alzheimer’s detection from brain MRIs have been conducted, this study made substantial steps towards actually performing this in real-world clinical settings as opposed to perfect laboratory settings.” 

The findings of the team were published in the journal PLOS ONE

Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.

Recent articles