Skip to main content

Table 5 Performance of derived ETSM on ICUC in the experiment of predicting AKI 24 hours ahead

From: Utilizing imbalanced electronic health records to predict acute kidney injury by ensemble learning and time series model

Model

AUC

Sensitivity

F1-score

AP

 

(95% CI)

(95% CI)

(95% CI)

(95% CI)

ETSM

0.810 ±0.002

0.746 ±0.003

0.577 ±0.003

0.594 ±0.004

ETSM-ex

0.737 ±0.002*

0.629 ±0.004*

0.470 ±0.002*

0.470 ±0.003*

ETSM-bool

0.759 ±0.002*

0.654 ±0.005*

0.512 ±0.003*

0.530 ±0.004*

ETSM-times

0.803 ±0.002*

0.726 ±0.003*

0.579 ±0.003

0.647 ±0.003

  1. Note: CI = confident interval
  2. *indicates ETSM significantly outperforms the baseline with p <0.01 using Student t-test