Algorithm Beats Experts in Alzheimer's Diagnosis

Researchers are using a new deep learning algorithm that harnesses magnetic resonance imaging (MRI) brain scans, age, gender, and scores on the Mini-Mental State Examination to more accurately diagnose the risk of Alzheimer's disease, reports Futurity . The algorithm's developers at Boston University School of Medicine first worked with data from the Alzheimer's Disease Neuroimaging Initiative, then validated their findings using data from three other cohorts. "We also obtained postmortem data on a small group of people who were evaluated for the presence of any Alzheimer's disease-related pathology," said Boston University Professor Vijaya Kolachalama. "Interestingly, we found that our model predictions of Alzheimer's risk correlated with neuropathology scores on these cases." Kolachalama added that comparing the algorithmic model's predictive ability to that of an international team of expert neurologists determined that "the model performed slightly better than the average expert neurologist." The most significant finding is that computers can accurately detect debilitating conditions like Alzheimer's from readily available data, like MRI brain scans. This implies that such technologies have a wide-reaching potential, especially in settings where expert clinicians are in short supply.