| Projects
HA Scaffolds/Micro-CT:
Quantification of soft tissue on hydroxyapatite scaffolds
using Micro-CT
publication: Hilldore et al 2007
Hydroxyapatite (HA) bone
scaffolds with
porosity spanning multiple
lengthscales are
investigated for use in load-bearing,
site specific, and critical sized
defects. In order to
to determine the extent of tissue ingrowth, scaffolds cultured
in vitro or implanted in vivo must be carefully
characterized. Techniques currently employed include scanning
electron microscopy and histology; however, these are
destructive in nature and can only represent the data in 2D.
Here, we have imaged Os-stained cells and tissue on HA scaffolds
for the first time using a nondestructive evaluation technique
called x-ray micro-computed tomography (micro-CT).
 An improved method to analyze data
obtained using micro-CT is presented. The example case was of a
hydroxyapatite scaffold cultured in vitro using mouse D1 ORL
UVA cells for one and four weeks (Figure 1). The histogram obtained
from the micro-CT data was decomposed into a Gaussian attenuation
distribution for each material in the sample, including scaffold,
tissue, and background (Figure 2). This was done by local sampling
of attenuation data and by using a non-linear curve fit, which
produced R2 values greater than 0.998 for the
simulations. Using this method, the material volumes are determined
prior to image segmentation by integrating the curves that simulate
each material. Also, materials with small volume fractions can be
accurately quantified. These are improvements over other reported
methods that use global thresholds prior to calculating the volume
of the sample and cannot be used to calculate small volumes.
Finally, ‘smart’ thresholds were selected based on the simulation in
order to visualize the distribution of the materials in the sample.
Because the attenuation distribution is known and well correlated to
the original data, the accuracy of the visualization can be
determined and the threshold adjusted accordingly. Two methods of
‘smart’ thresholding are presented for the example case (Figure 3).
This method is currently being used to study bone ingrowth in HA
scaffolds from an in vivo study.
 |
Fig. 1. Top view of a representative scaffold cultured for four
weeks and subsequently stained by OTO. The parallel lines show the
approximate region imaged using micro-CT. |
 |
Fig.
2. Simulations of data using Gaussian curves to represent the
background, border, thin tissue, scaffold, and thick tissue. (a) A
one week sample and (c) a four week sample. (b) and (d) show more
detail at smaller numbers of voxels of (a) and (c), respectively,
between 0.75 to 4.5 cm-1. The legend applies to (a-d). [Click
to Enlarge] |
 |
Fig.
3. Histograms and tissue visualizations for a partial cross section
of a four week sample. (a) Threshold locations for the threshold
method, (c) visualization of the threshold method where red shows
the scaffold and blue represents the thick tissue. (b) Threshold
locations for the overlap method, (d) visualization of the overlap
method, where the green voxels are enlarged, which shows the
overlap. [Click
to Englarge] |
Projects
|
