Novel AI Tool Plus Patient’s Genetic Information Up To 1,000% More Accurate At Predicting Autoimmune Disease Risk Than Current Models, Study Suggests

Healio (4/3, Volansky) reports a study suggests that “a risk prediction score that uses machine learning and a patient’s genetic information may identify autoimmune conditions up to 1,000% more accurately than current models.” Researchers developed the Genetic Progression Score (GPS) tool “using data from electronic health records and large genetic studies of individuals with rheumatoid arthritis and systemic lupus erythematosus.” When they compared GPS with currently existing risk prediction tools, researchers “found that the novel score was between 25% and 1,000% more accurate in determining which patients had symptoms that would progress to more advanced disease.” They explained, “AI and genetics are particularly useful here since we may be able to find patterns of genetics or other molecular markers that indicate the risk progression and apply targeted treatment regime.” Findings from the study were published in Nature Communications.