Information Technology Reference
In-Depth Information
should be noted that any parallel programming or graphical acceleration option
were not used in this work. However, if needed, the execution time can be reduced
using such methods.
The end-plate slices are segmented successfully thanks to the shape registration
process although the shape model is obtained using in-plane slices. Further works
are suggested to include the rotational transformation in the shape registration
process for the sagittal plane to enhance the results and capture more
fine details in
the end-plate slices. In some cases, small portions of spinal processes and ribs are
segmented erroneously. Future work can be including to investigate to reduce the
misclassi
cations.
Possible works to estimate how the segmentation quality affect the BMD mea-
surements and fracture analysis can be analyzed. To assist the VB fracture analysis,
an automated point correspondence detection algorithm, such as scale-invariant
feature transform (SIFT), can be tested to detect the VB height changes. In this
problem, the corresponding points on the same patient, which is scanned at speci
c
time intervals, should be detected successfully.
References
1. Klinder T, Ostermann J, Ehm M, Franz A, Kneser R, Lorenz C (2009) Automated model-
based vertebra detection, identification, and segmentation in CT images. Med Image Anal
13:471
482
2. Kang Y, Klaus E, Kalender WA (2003) A new accurate and precise 3-D segmentation method
for skeletal structures in volumetric CT data. IEEE Trans Med Imaging 5:586
-
598
3. Kalender WA, Felsenberg D, Genant HK, Fischer M, Dequeker J, Reeve J (1995) The
European Spine Phantom
-
—
a tool for standardization and quality control
in spinal bone
92
4. Mastmeyer A, Engelke K, Fuchs C, Kalender WA (2006) A hierarchical 3D segmentation
method and the de
nition of vertebral body coordinate systems for QCT of the lumbar spine.
Med Image Anal 10(4):560
-
577
5. Kaminsky J, Klinge P, Bokemeyer M, Luedemann W, Samii M (2004) Specially adapted
interactive tools for an improved 3D-segmentation of the spine. Comput Med Imaging
Graph 28(3):119
-
127
6. Roberts M, Cootes T, Adams J (2005) Vertebral shape: automatic measurement with
dynamically sequenced active appearance models. Proc Int Conf Med Image Comput Comput
Assist Interv 2:733
mineral measurements by DXA and QCT. Eur J Radiol 20(2):83
-
740
7. Tan S, Yao J, Ward MM, Yao L, Summers RM (2008) Computer aided evaluation of
ankylosing spondylitis using high-resolution CT. IEEE Trans Med Imaging (TMI) 27
(9):1252
-
1267
8. Criscob JJ, Sebastiana TB, Teka H, Kimia BB (2003) Segmentation of carpal bones from CT
images using skeletally coupled deformable models. Med Image Anal 7(1):21
-
45
9. Kim Y, Kim D (2009) A fully automatic vertebra segmentation method using 3D deformable
fences. Comput Med Imaging Graph 33:343
-
352
10. Leventon ME, Grimson WEL, Faugeras O (2000) Statistical shape in
-
fl
uence in geodesic active
1324
11. Chen Y, Thiruvenkadam S, Huang F, Wilson D, Geiser EA, Tagare HD (2001) On the
incorporation of shape priors into geometric active contours. In: IEEE VLSM, pp 145
-
152
contours. In: Proceedings of IEEE CVPR, pp 1316
-
