Scientists at McGill University in Canada created a piece of software that can analyze images taken during an amyloid PET scan and provide an estimate of the chances of patients developing dementia. Alzheimer’s is associated with the buildup of amyloid plaques within the brain, but interpreting their location and concentrations into clinically useful information is difficult. Knowing how individual patients will fare can help with choosing therapies, structuring targeted clinical trials, and with developing new drugs.

Discover The World's MOST COMPREHENSIVE Mental Health Assessment Platform

Efficiently assess your patients for 80+ possible conditions with a single dynamic, intuitive mental health assessment. As low as $12 per patient per year.

The software, which relies on artificial intelligence and machine learning techniques, is able to spot the onset of dementia up to two years before actual symptoms develop. It was given a bunch of scans and related information from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) and it was made to figure out and learn which image characteristics are important and how they affect the chances of developing dementia. The final algorithm achieved an 84% rate of correctly predicting which patients will eventually have dementia.

The new technology may have major implications for how Alzheimer’s is diagnosed, monitored, and treated. The researchers are already making their software available to other scientists, but clinical use will only become possible once larger trials are done and the relevant regulatory authorities have given their approvals.

Hot daily news right into your inbox.