Grbf

Radius p

Parameter

Value

FIGURE 9 Comparison of the segmentation results for a frontal coronal cross-section of data set 2. (a) T1 weighted image. (b) Segmentation by vector quantization and subsequent manual interactive assignment of codebook vectors. (c) Segmentation by a GRBF classifier.

TABLE 4 Statistics of the training data sets used for the GRBF classifierO

# Voxels 3177 Gray matter Volume [cm3] 7.69

# Voxels 4664 White matter Volume [cm3] 11.29

# Voxels 500 CSF Volume [cm3] 1.21

Volume [cm3] 20.19

Recognition rate [%] 98.2

3620 2887 3383

3810 2986 3429

746 799 638

8179 6672 7450

19.81 16.16 18.03

Tissue

Data set 1

Data set 2

Data set 3

Data set 4

"The table shows the number of voxels labeled as "gray matter,'' "white matter,'' and "CSF," the resulting volumes, and the corresponding percentage of the whole training data set. Furthermore, the ratio of correct GRBF classification results for the training data set is listed.

FIGURE 10 Comparison of the segmentation results for a central coronal cross-section of data set 2. (a) T1 weighted image. (b) Segmentation by vector quantization and subsequent manual interactive assignment of codebook vectors. (c) Segmentation by a GRBF classifier.

TABLE 5 Segmentation results for vector quantization of the gray-level feature space according to Section 7 and subsequent manual interactive assignment of the codebook vectors to the tissue classes according to Section 8.1a

Tissue

0 0

Post a comment