Skip to main content

Table 3 The average Dice scores, Jaccard Index and Haursdorff distances (±standard deviation) of the test dataset evaluated at 21,000 iterations by the neural network model trained with all image channels combined

From: Automatic segmentation of human knee anatomy by a convolutional neural network applying a 3D MRI protocol

Class

DSC

Jaccard

Hausdorff Distance

Bone medulla

0.997±0.002

0.994±0.004

42.63±24.79

PCL

0.973±0.015

0.947±0.028

98.30±55.61

ACL

0.964±0.022

0.931±0.042

107.00±71.58

Muscle

0.998±0.001

0.997±0.002

66.94±42.77

Cartilage

0.966±0.018

0.934±0.034

111.90±18.25

Cortical bone

0.980±0.010

0.961±0.019

53.43±28.26

Artery

0.943±0.038

0.894±0.065

68.50±43.88

Collateral ligament

0.919±0.069

0.856±0.109

114.27±39.07

Tendon

0.982±0.005

0.965±0.009

81.40±21.66

Meniscus

0.955±0.032

0.916±0.057

185.60±53.54

Adipose tissue

0.998±0.010

0.997±0.003

16.568±6.685

Vein

0.980±0.010

0.962±0.018

30.36±8.00

Nerve

0.921±0.071

0.860±0.106

114.11±46.61

Average

0.967±0.040

0.940±0.068

83.92±58.15