Fig. 1 | npj Digital Medicine

Fig. 1

From: A computer vision system for deep learning-based detection of patient mobilization activities in the ICU

Fig. 1The alt text for this image may have been generated using AI.

Algorithm performance for detecting the occurrence of mobility activities. a Per-class specificity and sensitivity, evaluated at the frame-level. b Per-class receiver operating characteristic curves (ROC). These ROC curves demonstrate the trade-off between sensitivity (the true positive rate) and 1-specificity (the false-positive rate), as the detection thresholds are varied. The area under the ROC curve (AUC) is an aggregate measure of detection performance, and indicates the probability that the model will rank a positive example more highly than a negative example (a model whose predictions are 100% correct will have an AUC of 1.0)

Back to article page