Information Technology Reference
In-Depth Information
Fig. 10 Trabecular bone specimens of T10 from a normal (a) and an osteoporotic (b) subject.
Color-coded 3D visualizations of the isotropic scaling indices
ʱ
of micro-CT scans with a spatial
resolution of 26
-values representing the local dimensionality of the bone voxels are
increasing from blue over green to red color. Rod-like structures have lower local dimensionality
than plate-like structures. a shows greater local dimensionality compared to b resulting from the
transformation of trabecular bone from plate- to rod-like structures due to osteoporosis
μ
m
.
ʱ
cortical bone contributes substantially to the mechanical properties of the bone [
80
].
Information about bone geometry and bone density distribution is used for MDCT-
based FEMs at the spine [
81
]. A representative FEM of a vertebra is displayed in
Fig.
11
.
MDCT-based trabecular bone microstructure parameters and FEMs at the spine
were successfully validated in-vitro by using biomechanically determined failure
load as gold standard [
68
84
]. Trabecular bone microstructure
parameters and FEMs provided better diagnostic performance for differentiating
subjects with versus without osteoporotic vertebral fracture than BMD measure-
ments. Ito et al. [
67
] computed MDCT-based trabecular bone microstructure
parameters in the vertebrae of 82 postmenopausal women in-vivo including 39
women with and 43 without osteoporotic vertebral fracture. The microstructure
parameters revealed higher relative risk for prevalent osteoporotic vertebral fracture
than vertebral BMD as determined by DXA. The authors reported reproducibility
errors for the trabecular microstructure parameters in cadaver specimens in the
range from 0.67 to 12.30 %. Vertebral FEMs were not only able to detect prevalent
osteoporotic vertebral fractures better than BMD measurements, but also predicted
incident fractures in-vivo [
85
,
86
]. Wang et al. [
86
] performed baseline MDCT-
based FEMs to determine L1 vertebral compressive strength and a load-to-strength
ratio in 306 men aged 65 years and older. An incident osteoporotic vertebral
fracture was diagnosed in 63 subjects during follow-up of averaged 6.5 years. The
area-under-the-curve for areal BMD (AUC = 0.76) was signi
70
,
78
,
79
,
82
-
-
cantly lower than for
