Case Western Reserve University scientists have developed a new machine learning program that looks like it may be able to diagnose Alzheimer's disease before symptoms begin to interfere with day-to-day activities.
While there is no cure for Alzheimer’s, a number of drugs can delay or prevent symptoms from worsening for up to five years or more, Early diagnosis and treatment—the goal of the new computer-based program—is key to allowing those with the disease to remain independent longer, according to the researchers.
The computer program integrates a range of Alzheimer's disease indicators, including mild cognitive impairment. In two successive stages, the algorithm selects the most pertinent to predict who has Alzheimer's.
The team published its study (“Cascaded Multi-view Canonical Correlation (CaMCCo) for Early Diagnosis of Alzheimer’s Disease via Fusion of Clinical, Imaging and Omic Features”) in Scientific Reports.
“The introduction of mild cognitive impairment (MCI) as a diagnostic category adds to the challenges of diagnosing Alzheimer’s Disease (AD). No single marker has been proven to accurately categorize patients into their respective diagnostic groups,” write the investigators.
“In this work, we present a combined framework, cascaded multiview canonical correlation (CaMCCo), for fusion and cascaded classification that incorporates all diagnostic categories and optimizes classification by selectively combining a subset of modalities at each level of the cascade.”
“Many papers compare the healthy to those with the disease, but there's a continuum,” said Anant Madabhushi, Ph.D., 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 scientists tested the algorithm using data from 149 patients collected via the Alzheimer's Disease Neuroimaging Initiative. The CaMCCo integrates measurements from magnetic resonance imaging scans, features of the hippocampus, glucose metabolism rates in the brain, proteomics, genomics, mild cognitive impairment, and other parameters.
Dr. Madabhushi's lab has repeatedly found that integrating dissimilar information is valuable for identifying cancers. This is the first time he and his team have done so for diagnosis and characterization of Alzheimer's disease.
“The algorithm assumes each parameter provides a different view of the disease, as if each were a different set of colored spectacles,” explained Dr. Madabhushi.
The program then assesses the variables in a two-stage cascade. First, the algorithm selects the parameters that best distinguish between someone who's healthy and someone who's not. Second, the algorithm selects from the unhealthy variables those that best distinguish who has mild cognitive impairment and who has Alzheimer's disease.
“The remaining views are combined to give the best picture,” according to Dr. Madabhushi.
In predicting which patients in the study had Alzheimer's disease, CaMCCo reportedly outperformed individual indicators as well as methods that combine them all without selective assessment. It also was better at predicting who had mild cognitive impairment than other methods that combine multiple indicators.
The researchers say they will fine-tune their technique with data from multiple sites. They also plan to use the software in an observational mode. Thus, as a collaborating neurologist compiles tests on patients, the computer would run the data.
If CaMCCo proves useful in predicting early Alzheimer's, Dr. Madabhushi expects to pursue a clinical trial for prospective validation.