Machine learning algorithm outperforms tests in diagnosing early Alzheimer's

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon

Diagnosing Alzheimer’s in its early stages is key to providing patients with proper delaying or preventive drugs. A new study, published in Scientific Reports, explains how a machine learning algorithm could be used to predict the onset of the disease.

Developed by researchers at Case Western Reserve University, the computer machine learning algorithm takes into account measurements from magnetic resonance imaging (MRI) scans, the state of the hippocampus, the brain's glucose metabolism, proteomics, genomics, mild cognitive impairment and other factors to predict a patient’s risk of developing Alzheimer’s.

"Many papers compare the healthy to those with the disease, but there's a continuum," said Anant Madabhushi, F. Alex Nason professor II of biomedical engineering at Case Western Reserve. "We deliberately included mild cognitive impairment, which can be a precursor to Alzheimers, but not always. The algorithm assumes each parameter provides a different view of the disease, as if each were a different set of colored spectacles.”

The study enrolled 149 patients, taken from the Alzheimer's Disease Neuroimaging Initiative, and analyzed data on two sets. The first set tested the algorithm's ability to differentiate healthy patients from those who were not, while the second set tested its ability to identify which patients had mild cognitive impairment and who had Alzheimer’s. The algorithm was able to outperform other methods of detecting Alzheimer’s diseases before symptoms interfered with patients' daily lives.