Dascena developed and evaluated a machine learning algorithm for Acute Kidney Injury (AKI) prediction that demonstrates high sensitivity and specificity up to 72 hours in advance of onset. The outcomes are published in the Canadian Journal of Kidney Health and Disease.
In this multi-center study, data were collected from inpatient wards at Stanford Medical Center and intensive care unit patients at Beth Israel Deaconess Medical Center. Performance of the machine learning algorithm was compared to performance of Sequential Organ Failure Assessment (SOFA) score for AKI identification, and was validated by measuring Area Under the Receiver Operating Characteristic (AUROC). The algorithm AUROC values outperformed the comparable SOFA scoring system at all measured prediction windows, up to 72 hours prior to AKI onset. These results suggest that a machine learning–based AKI prediction tool may lead to earlier clinical interventions, as they offer important prognostic capabilities for determining which patients are likely to suffer AKI.
Read the paper here.