Table 1 CA and MAE prediction results for force estimation from identical video data after 100% learning of each sequence’s data.

From: Vision-based estimation of manipulation forces by deep learning of laparoscopic surgical images obtained in a porcine excised kidney experiment

 

P1

P2

P3

Non-cropping

Cropping

Non-cropping

Cropping

Non-cropping

Cropping

CA (%)

98.6

96.9

99

98.6

98.6

95.6

MAE (N)

0.04

0.108

0.045

0.083

0.05

0.134

  1. Significant values are in [bold].
  2. P Pig, CA Classification accuracy, MAE Mean absolute error.
Back to article page