Skip to main content
. Author manuscript; available in PMC: 2010 Jul 1.
Published in final edited form as: Neuroimage. 2009 Feb 21;46(3):717–725. doi: 10.1016/j.neuroimage.2009.02.012

Table 2.

Accuracy of the four segmentation methods (eMRF, MRF-SVM, atlas-based segmentation and MRF) for each of the 21 segmented brain structures is evaluated using the VOP (top row) for each segmentation and VDP (second row) metrics. The percentage improvement in segmentation accuracy is computed for eMRF versus MRS-SVM, eMRF versus atlas-based segmentation, and eMRF versus MRF. The improvements in VOP and VOD are calculated separately. A ‘+’ sign always means that eMRF method outperforms the other methods for the specific neuroanatomical structure and ‘−’ means that eMRF underperforms.

CORT CPED HC CPU GP ICAP PAG INFC MED THAL MIDB
eMRF VOP (%) 91.10 73.15 86.20 87.62 79.64 73.14 90.26 85.32 91.93 93.53 93.27
VDP (%) 3.34 8.32 5.56 3.98 7.54 11.05 3.55 5.26 7.12 2.48 2.97

MRS-SVM VOP (%) 93.46 63.31 83.92 85.72 74.47 63.91 80.90 83.21 90.27 92.07 90.48
VDP (%) 1.63 11.50 5.87 3.89 11.34 12.66 7.97 7.81 10.03 2.78 3.84

Improvement VOP (%) −2.53 15.55 2.71 2.22 6.95 14.44 11.57 2.54 1.84 1.59 3.09
VDP (%) 104.24 −27.60 −5.33 2.16 −33.51 −12.78 −55.44 −32.64 −28.98 −10.61 −22.57

Atlas-based segmentation VOP (%) 84.19 64.31 84.07 82.46 68.19 60.83 84.47 75.67 84.24 91.16 88.56
VDP (%) 12.24 11.76 11.91 11.54 13.51 11.40 11.14 14.87 12.50 11.57 12.47

Improvement VOP (%) 8.20 13.75 2.54 6.26 16.79 20.25 6.85 12.76 9.12 2.60 5.32
VDP (%) 72.74 29.20 53.34 65.53 44.16 3.13 68.11 64.60 43.05 78.54 76.18

MRF VOP (%) 90.77 67.69 87.69 88.46 78.46 73.85 85.38 83.08 86.15 93.08 90.77
VDP (%) 4.35 12.17 6.96 6.96 7.83 13.04 2.61 4.35 10.43 3.48 4.35

Improvement VOP (%) 0.36 8.06 −1.70 −0.95 1.50 −0.96 5.71 2.70 6.70 0.49 2.76
VDP (%) 23.24 31.62 20.08 42.84 3.61 15.31 −36.19 −21.05 31.76 28.58 31.68

AC CBLM VEN PON SNR INTP OLFB OPT TRI CC Avg.

eMRF VOP (%) 50.76 92.68 71.83 80.03 78.98 71.61 82.50 68.67 73.83 65.75 80.09
VDP (%) 25.55 3.73 16.83 12.49 10.16 26.09 10.79 5.67 7.33 20.57 9.54

MRS-SVM VOP (%) 25.47 96.29 52.85 74.38 60.79 58.18 91.23 59.25 69.72 57.65 73.69
VDP (%) 68.49 1.48 24.92 14.30 25.65 62.22 7.85 19.93 17.27 24.28 16.46

Improvement VOP (%) 99.29 −3.75 35.91 7.60 29.91 23.08 −9.57 15.90 5.89 14.04 8.68
VDP (%) 62.70 −151.51 32.48 12.65 60.39 58.06 −37.49 71.58 57.57 15.26 42.04

Atlas-based segmentation VOP (%) 50.99 85.91 67.30 70.68 73.18 75.77 82.19 44.74 59.72 49.55 72.77
VDP (%) 11.51 12.49 12.40 14.55 12.93 15.23 12.40 12.05 12.36 12.28 12.53

Improvement VOP (%) −0.45 7.89 6.73 13.24 7.92 −5.49 0.38 53.49 23.62 32.69 10.05
VDP (%) −121.98 70.13 −35.70 14.17 21.41 −71.32 12.98 52.99 40.71 −67.49 23.84

MRF VOP (%) 55.38 93.08 70.77 73.08 68.46 76.92 84.62 53.08 63.85 71.54 77.91
VDP (%) 19.13 3.48 19.13 17.39 15.65 13.91 16.52 14.78 10.43 22.61 10.93

Improvement VOP (%) −8.35 −0.42 1.50 9.52 15.36 −6.91 −2.50 29.39 15.63 −8.10 2.79
VDP (%) −33.54 −7.22 12.04 28.18 35.09 −87.55 34.68 61.68 29.77 9.00 12.71
HHS Vulnerability Disclosure