This research has been performed To make an AI model distinguishing possibly avoidable blood draws for serum potassium among pediatric patients following heart medical procedure. All patients admitted to the heart ICU at Boston Children’s Hospital between January 2010 and December 2018 with a length of remain more prominent than or equivalent to 4 days and more noteworthy than or equivalent to two recorded serum potassium estimations. We gathered factors identified with potassium homeostasis, including serum science, hourly potassium admission, diuretics, and pee yield.

Utilizing set up AI procedures, including arbitrary woods classifiers, and hyperparameter tuning, we made models foreseeing whether a patient’s potassium would be ordinary or strange dependent on the latest potassium level, drugs directed, pee yield, and markers of renal capacity. We built up various models dependent on various age-classifications and worldly closeness of the latest potassium estimation.

Reference link-