According to a study published in the journal PLOS ONE, researchers from the Massachusetts General Hospital (MGH) have developed a novel artificial intelligence-based tool that can detect Alzheimer’s disease with an accuracy of 90% from routinely collected clinical brain images.

The findings of the study showed that the tool was able to detect Alzheimer’s disease regardless of other variables, such as age. Dr. Matthew Leming, a research fellow at MGH’s Center for Systems Biology, stated that deep learning algorithms frequently struggle to identify the rarer cases of early-onset Alzheimers disease because it typically affects older persons.

In the study, the researchers used 11,103 brain magnetic resonance images (MRIs) from 2,348 patients at risk for Alzheimer’s disease and 26,892 images from 8,456 patients without Alzheimer’s disease. The model was able to detect the risk of Alzheimer’s disease with 90.2% accuracy.

Dr. Leming commented that dealing with data that is vastly different from the training set is another typical difficulty in disease identification, particularly in real-world settings. However, this study made significant progress toward carrying out the experimentation in realistic clinical situations as opposed to ideal laboratory conditions.