Artificial Intelligence Can Now Detect Alzheimer's Disease
Alzheimer’s disease is one of the most debilitating and devastating diseases affecting adults today. In fact, it’s the sixth-leading cause of death among adults in the United states and one in 10 people age 65 and older has Alzheimer's dementia.
Suffice it to say that Alzheimer’s is a major concern for those getting up there in age. While science has yet to develop a cure for the insidious disease, researchers have made a monumental breakthrough that can aid in early detection of the disease.
A team of scientists at the University of Bari, in Italy, created an algorithm for detecting minute structural changes in the brain caused by Alzheimer’s. To train the computer system, the team fed it 67 MRI (magnetic resonance imaging) scans -- 38 from Alzheimer's patients and 29 from healthy control subjects. This helps the artificial intelligence system to “learn” or recognize common patterns in neurological activity in both healthy individuals and those with Alzheimer’s.
Following the training, researchers tested the computer system by having it process brain scans from 148 subjects. Of the 148 scans, 48 were scans of individuals with Alzheimer’s, another 48 were scans of those suffering some form of mild cognitive impairment that eventually developed full-blown Alzheimer's, and the rest were of normal brains.
The computer was able to correctly identify or “diagnose” 86% of the scans showing Alzheimer’s disease and 84% of the scans showing mild cognitive impairment, leading researchers to believe they had developed an effective means for early detection and diagnosis of the disease.
While the research community is still far away from developing a cure, this ability to identify symptoms or signs of Alzheimer’s as early as 10 years prior to its full-blown development can do wonders for slowing the progression of the disease, the researchers believe.
One area that limited the development of the system was that researchers only had access to scans held in USC LA's Alzheimer's Disease Neuroimaging Initiative database. Researchers believe that with a greater number of samples, as well as some additional tweaking to the AI’s algorithm, they can enhance the accuracy of the computer system to make it a reliable and noninvasive means for early detection.
This isn’t the first time AI has been explored as a means for detecting Alzheimer’s though. Researchers at VU University Medical Centre in Amsterdam have been using MRI brain scans in an effort to diagnose Alzheimer’s earlier. Several other universities in California have also been working on various AI systems that can detect the disease via eye examination.
Is AI the future of medical diagnoses? Do you find them more reliable than the trained eye (and brain) of a human physician?
Leave a comment down below with your thoughts.