Table 2 The Kappa index for the WM/GM segmentation of the simulated database with different Rician noise levels and a 20% inhomogeneity
From: Magnetic resonance image tissue classification using an automatic method
Methods | White matter (%) | Grey matter (%) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
Noise level | 0 | 1 | 5 | 7 | 9 | 0 | 1 | 3 | 5 | 7 | 9 |
EM | 86.1 | 91.5 | 92.2 | 90.1 | 86.4 | 83.1 | 90.8 | 92.5 | 92 | 89.1 | 84.2 |
SPM 5 | 91.05 | 94.2 | 93.6 | 90.2 | 86.3 | 91.2 | 93.4 | 93.3 | 92.1 | 90 | 86.6 |
HMC | 97.8 | 95.7 | 93.9 | 937 | 91.7 | 97 | 94.5 | 95.9 | 93.8 | 92.7 | 91.1 |
NL-FCM | 95.6 | 94.2 | 91.5 | 89.8 | 83.2 | 95.4 | 94.1 | 93.8 | 92.9 | 89.9 | 79.3 |
FCM | 97.1 | 96 | 92 | 88 | 84 | 97 | 96 | 91 | 86 | 87 | 83 |
Proposed-Method | 96.8 | 94.9 | 93.7 | 93.9 | 92.1 | 95.7 | 95.4 | 92.8 | 92.6 | 92.1 | 91.2 |