Table 1 Evaluation metrics for the two different temporal model types: MS-TCN and LTContext

From: Robotic scrub nurse to anticipate surgical instruments based on real-time laparoscopic video analysis

Model

wAP

wAR

wAA

wAF1

MS-TCN

74.51 ± 1.86

69.59 ± 1.45

87.81 ± 0.79

70.29 ± 1.65

LTContext

77.15 ± 1.93

70.31 ± 0.96

88.32 ± 0.64

71.54 ± 1.07

  1. The evaluation metrics are computed class-wise, for each instrument and working trocar individually, and then averaged over all surgeries in the test set, yielding the weighted-averaged precision (wAP), recall (wAR), accuracy (wAA), and F1 score (wAF1). The weighted-averaged metrics are calculated by taking the mean of all per-class scores while considering the number of actual occurrences of each class in the test data set. The averaged metrics over tenfold are reported (%) with the corresponding standard deviation ( ± ).
Back to article page