Researchers from an Ohio University have developed a machine learning programme that could prove beneficial for diagnosing Alzheimer’s disease.
A team at Case Western Reserve University developed the programme, which they state appears to outperform other methods for diagnosing the disease before symptoms interfere with patients’ quality of life.
The programme uses an algorithm that integrates a range of Alzheimer’s indicators to predict who has the disease. The algorithm assesses measurements from magnetic resonance imaging (MRI) scans, features of the hippocampus, glucose metabolism rates, proteomics, genomics, mild-cognitive impairment and other metrics.
Anant Madabhushi, F. Alex Nason professor II of biomedical engineering at Case Western Reserve, said: “Many papers compare the healthy to those with the disease, but there’s a continuum. We deliberately included mild cognitive impairment, which can be a precursor to Alzheimer’s, but not always. The algorithm assumes each parameter provides a different view of the disease, as if each were a different set of coloured spectacles.”
The programme evaluates the variables across two stages. First, the programme uses the parameters to identify healthy and unhealthy people. It then selects unhealthy variables to identify mild cognitive impairment in people to help assess who has Alzheimer’s disease.
The programme outperformed individual indicators as well as combinative methods that don’t use selective assessment.
The team are continuing to test the approach and are planning on using the software to test data with collaborating neurologists. If the programme is effective at predicting early Alzheimer’s the team expects to pursue clinical trials for validation.