‘Only the beginning’: AI algorithm shows promise for predicting CVD on a budget
A machine learning tool developed by researchers at Michigan State University can accurately predict an individual’s height within an inch and may offer more precise CVD and cancer risk assessment in the future, according to a report featured this month in Genetics.
The DNA algorithm, built by Stephen D. H. Hsu, MSU’s vice president for research and graduate studies, and his colleagues in East Lansing, Michigan, has been validated for three outcomes: predicting height, bone density and a person’s level of education. The tool calculates its predictions based on a patient’s genome.
Hsu and co-authors said in the journal they developed the algorithm by drawing data from the U.K. Biobank, which provided the researchers with information about the genetic makeup of half a million adults in England. The algorithm studied the genetics and height of each person, eventually teaching itself to construct genomic predictors for human traits based on that knowledge.
Those predictors explained 40 percent, 20 percent and 9 percent of total variance for height, heel bone density and educational attainment, respectively, in validation tests, according to the study. Height was predicted accurately in all cases within roughly an inch, while bone density and educational attainment weren’t as precise. They were, however, accurate enough to identify outlying individuals, like those who could be at risk for osteoporosis or might struggle in school.
Hsu et al. wrote they’re working to improve the DNA tool in studies involving larger datasets. With the cost of DNA sequencing decreasing, they said, the algorithm could be easily applied in the future to more serious illnesses, eventually predicting future risk of heart disease, cancer and other major killers.
“Our team believes this is the future of medicine,” Hsu said in a release. “For the patient, a genomic test can be as simple as a cheek swab, with a cost of about $50. Once we calculate the predictors for genetically based diseases, early intervention can save billions of dollars in treatment costs, and, more importantly, save lives. This is only the beginning.”
After graduating from Indiana University-Bloomington with a bachelor’s in journalism, Anicka joined TriMed’s Chicago team in 2017 covering cardiology. Close to her heart is long-form journalism, Pilot G-2 pens, dark chocolate and her dog Harper Lee.