Skip to main content

Table 1 Evaluation measures for DETACH, Zhou [25], and van Hees [11] algorithms applied to the reference testing dataset. Data represent means (± 95% CI)

From: Detecting accelerometer non-wear periods using change in acceleration combined with rate-of-change in temperature

 

Precision

Recall

F1-Score

Accuracy

DETACH

0.942 (0.844 to 1.000)

0.942 (0.883 to 1.000)

0.942 (0.880 to 1.000)

0.996 (0.994 to 1.000)

Zhou

0.538 (0.395 to 0.681)

0.824 (0.742 to 0.906)

0.651 (0.551 to 0.751)

0.971 (0.958 to 0.983)

van Hees

0.710 (0.554 to 0.866)

0.590 (0.451 to 0.728)

0.644 (0.535 to 0.753)

0.978 (0.973 to 0.984)