Introduction

Based on the H2O automated machine learning (AutoML) platform, we developed and validated a AutoML models to identify subphenotypes of acuate kidney disease (AKI)

The objective of this Shiny AKI application is to utilize clinical data to identify AKI subphenotypes based on AutoML.

AKI identification

Age

First day urine output (mL)

Temperature (°C)

Platelets (10^9/L)

Bun (mg/dL)

INR

PTT

SOFA

Lactate (mmol/L)

Diabetes (0:NO, 1:Yes)

chronic.kidney.disease (0:NO, 1:Yes)