Skip to main content
Fig. 2 | BioData Mining

Fig. 2

From: Algorithm-based detection of acute kidney injury according to full KDIGO criteria including urine output following cardiac surgery: a descriptive analysis

Fig. 2

Visualization of the automated AKI detection process on an exemplary patient during ICU stay. In the first row the graphic depicts the cumulative maximum of the AKI detection by any information derived from the creatinine, urine and dialysis signal. Instead, the first row of the columns below shows the result of the AKI detection based on a single method. The remaining plots in each column from left to right show: the relative change in creatinine value from the selected baseline along with critical limits, the raw creatinine value along with red markers indicating critical increases from the observed minimum in the last 48 hours, the calculated urine output over different time windows in hours along with critical limits, the urine volume normalized by the body weight and a binary indicator of the need for dialysis. In all partial images, the x-axis describes the days since ICU admission

Back to article page