Skip to main content

Table 4 Average ± standard deviation of accuracy, AUC, precision and recall of models in predicting functional status

From: Computational Barthel Index: an automated tool for assessing and predicting activities of daily living among nursing home patients

 

Re-evaluation models (MREdτ)

Evaluation models (MEdτ)

Prediction time Ï„

Accuracy

AUC

Precision

Recall

Accuracy

AUC

Precision

Recall

Current

.900 ± .007

.947 ± .006

.910 ± .011

.907 ± .041

.743 ± .029

.795 ± .010

.743 ± .046

.800 ± .128

3 Months

.815 ± .020

.876 ± .011

.849 ± .019

.816 ± .094

.727 ± .037

.761 ± .006

.734 ± .049

.783 ± .161

6 Months

.759 ± .029

.808 ± .014

.784 ± .029

.737 ± .165

.720 ± .038

.746 ± .009

.721 ± .045

.729 ± .238

12 Months

.737 ± .035

.772 ± .022

.742 ± .049

.699 ± .226

.716 ± .039

.725 ± .016

.696 ± .073

.701 ± 264