Calibrations – Punumbra-spread, Distance, Area/Volume Accuracies-
Punumbra-spread = 2.4 pixels (or 1.30mm for pixel size=0.54mm) across using an 1.2mm X-ray tube-spot.
Distance – At high-contrast & 1% low-contrast edge, both accurate to 1/50th of a pixel in (delta)r (i.e. (delta)r in (delta)r/r);
and,
Area/Volume – At high-contrast & 1% low-contrast edge, accurate to 1/18th & 1/10th of a pixel respectively in (delta)r (i.e. 2(pi)r(delta)r/(pi)r2 or
(delta)(area) = 2(pi)r(delta)r (Volume = Area x Slice-thickness).
High-Contrast Resolution – Mean Modulation Transfer Function (MTF) (Ref. 2) at the high-contrast edge were improved from 3.38 lp/cm to 9.61 or 10.04 lp/cm at 50% normalized modulation level after 7x or 9x magnification and enhancement respectively.
Processing-Time- 9x Magnification & enhancement of Region-of-Interest (ROI) of size of: 107x107 took 1.17 seconds by a 2.4GHz-CPU.
Figures 1 & 2 show the original phantom image and the reduction of the penumbra effect at the guide-wire edges.
Six Field-Examples:-
1. Figures 3, 4, & 5 show the original image of the fractured bone (femur) with ROI, the 9x ROI images of `before' and 'after' enhancement side-by-side, and the enhanced image with the annotation for the minimum gap measurement.
Minimum gap distance = 0.61 ± 0.02mm.
(i) In this case, average blurring at the Edge-Response-Function caused by the penumbra-spread effect = 0.84mm (pixel-size = 0.35mm). This implies that the measurement of the above minimum gap distance would not have been possible without using this technique,
and,
(ii) Selective enhancement of the high-contrast fracture edges did clean-up the effect of the medium-contrast streaking artifact from the steel pin.
[Applications - reduce streaking artifact and sharper edge-profiles for clearer diagnosis and accurate measurements of minute fracture(s) for quantitative assessment in fracture bone healing progress ].
2. A ROI was centered on to the fractured radius bone of the forearm and magnified 9x. Figures 6, 7 & 8 show the original image with ROI, the 9x ROI `before' and `after' enhancement side-by-side, and the enhanced image with the annotation for the minimum gap measurement.
(i) Minimum gap distance = 0.89 ± 0.02mm.
and,
(ii) Selective enhancement of the high-contrast fracture edges over the
medium-contrast artifact did clean up the streaking artifact from the steel pin.
[Applications – detection of minute fractures for clearer diagnosis and
quantitative assessment of fracture bone healing progress].
3. Figures 9, 10, & 11 show the original image of fractured thigh femur with inserted intramedullary fixation nail with ROI, the 7x ROI images of 'before' and 'after' enhancement placed side-by-side, and the 7x ROI enhanced image with annotations for the gap measurements:-
(i) Minimum gap measurements were:-
(a) Fracture gap 1 = 1.15 ± .02mm;
(b) Fractured gap 2 = 1.23 ± .02mm;
(c) Fractured gap 3 = 1.43 ± .02mm; and,
(ii) Traces of streaking artifact were still visible in spite of the most advance 64 row data used. Selective enhancement of the high-contrast bone edges over the medium-contrast did reduce the artifact further.
Noise filter may be used to completely remove the remaining streaking artifact and the background noises. But the trade-off is the removal of the useful high spatial frequency information. Figure 12 illustrates the case.
[Applications – reduce artifact and the detection of sharper edge-profiles of minute fractures for clearer diagnosis and quantitative assessment of fracture bone healing progress].
4. A region of Interest (ROI) was centered on the tear of the Posterior Horn of the Medial Meniscus. Figures 13, 14, & 15 show the original image of MR knee scan with ROI centered on the tear region, the 7x ROI `before' and `after' enhancement side-by-side, and 7xROI enhanced image with annotations for measurements.
Measurements within the tear enclosure were:-
(i) Length of Tear = 11.73 ± 0.02mm;
and,
(ii) Maximum Width = 3.34 ± 0.02mm.
[Applications - The accurate measurements of minute tear effects within the MR knee images are possible for treatment planning and monitoring].
5. Slice#8 of multiple slices of images of humerus fracture was analyzed. Figures 16, 17, & 18 show the original image with ROI, 7xROI 'before' and 'after' enhancement side-by-side, and the 7x ROI enhanced image with annotation to measure the fractured gap.
The fractured gap was invisible either in the original image or within the image of 7xROI 'before' enhancement. But after the enhancement, the fracture was clearly visible with the fractured gap measured by a ruler as:-
Fractured gap = 1.22 ± .02mm.
[Applications – Quantitative detection of invisible hairline fractures are now possible for diagnosis and fracture bone healing progress ].
6. Figures 19, 20, & 21 show the original image with ROI of a metatarsal fracture, the 7xROI 'before' and 'after' enhancement, and the enhanced image with annotations to measure the fractured gaps:-
Fracture #1 = 0.64 ± .02mm;
Fracture #2 = 2.02 ± .02mm.
[Applications – The detection of sharper edge-profiles of minute fractures for clearer diagnosis and quantitative assessment of fracture bone healing progress].