Jung-Yoo Choi et al. Comparison of micro-CT and histomorphometry in the measurement of bone-implant contact ratios. Oral Surgery, Oral Medicine, Oral Pathology and Oral Radiology.
January 4, 2019
How Analyze was Used
“After scanning, image segmentation was performed using the Analyze 12.0 software (AnalyzeDirect, Overland Park, KS, USA). To reduce image noise and preserve detail in three dimensions, the image series were filtered with a 5 × 5 × 5 median filter using the ‘Spatial Filter’ module. The dataset was then manually reoriented using the ‘Oblique Sections’ function in Analyze to allow for the visualization of the standard coronal, sagittal, and horizontal planes through the implants. Using the ‘Image Calculator’ module, the images were reformatted to cubic volume with a resliced 0.03-mm image thickness of the surface area on the implant area for the x-axis and z-axis (Figure 1A). Image segmentation was performed using a combination of the ‘Semi-automatic’ and ‘Manual Technique’ settings. The ‘Volume Edit’ module was opened, and the ‘Semi-automatic’ tab was selected. The implant was segmented by point seeding using the ‘Object Extractor’ model. The upper and lower limits of the threshold range were adjusted manually to ensure that their respective values were descriptive of the implants. The object was then extracted, and volume rendering for the implant began. The segment of bone growth was segmented primarily by point seeding using the ‘Threshold’ model. The above process was repeated. Subsequently, a 3D rendering of the implant and bone growth could be generated, which could be saved as an object map. To allow reformatting of a surface volume for a 0.03-mm image thickness on the implant, the ‘Oblique Sections’ module was opened, and 3D image rotation was selected. To rotate the oblique plane into the coronal orientation, the plane was rotated by 10 degrees using the ‘Yaw Counter Clockwise’ and ‘Fly Value’ tools. Then, to save the resample of the entire data set in this oblique orientation, ‘Reformat Entire Volume’ was selected in the Method panel, and the transformed volume size was ‘Maintained.’ The resampled data set was saved to the selected tab by choosing ‘Generate Slices.’ The above process was repeated 18 times. Finally, a 3D rendering of bone growth could be generated, from which volumes could be calculated using the ‘ROI’ module (Figure 1C). The software manipulations used for the 3D reconstructed image are summarized in Figure 1D. The values obtained for the volume parameters were the mean ± standard deviation (SD). The 3D BIC area was calculated by dividing the bone formation area by the total cylinder area of the implant and then expressing the ratio as a percentage. First, the cylinder area was calculated from circular constant (π) × diameter (3 mm) × height (4 mm). Then, the bone formation area was calculated by dividing the bone volume by the bone formation thickness, which was the threshold thickness of 0.03 mm for image processing. This geometric and infinitesimal approach bypassed the partial volume effect that occurs because the resolution limitation of CT produces a voxel representing the average CT value of many different substances. The partial volume effect makes material boundaries or interfaces blurred, causing the quantitative interpretation of the interfacial surface area to be very difficult”
Three dimensional imaging, Micro-computed tomography, Bone-to-implant contact, Titanium implant
Seoul National University, Seoul, Korea