For the study, Ollie Fielding (pulseData, in New York) and his colleagues deployed a machine learning model to identify patients at risk of requiring kidney replacement therapy, such as dialysis or kidney transplantation.
An electronic health record database of 110,998 patients was used to create a machine learning model to predict progression to kidney failure. The system calculates weekly risk scores for patients, and for those with high-risk scores, an alert is sent so that treatment discussions can be made by a multidisciplinary team of clinicians.
Since the deployment of the machine learning model, 54% of patients started dialysis under optimal conditions. This is almost 3-times the national average of 20% and 14% better than the 47.3% rate prior to use of the machine learning model.
"Using artificial intelligence can help you focus care efforts on the right patients at the right time, but the human effort and clinical care delivery experts are required to improve outcomes for patients truly. Predictive analytics applied on a large scale can augment a highly focused care team," said Fielding.
"There is huge potential to change the healthcare dynamic by providing care before bad events rather than after. The possibility to shape the delivery of kidney care is tremendously exciting."