| Author |
Minimizing artifacts from hip implants on CT images?
|
|
|
| Hi,
I am working on a project on artifacts in CT images and I am looking
for information on the subject.
I am looking at the accuracy in CT images taken of the hip with a hip
implant in place. The implants are made of CoboltCrom or Titanium and
cause a great deal of artifacts in the image.
Are there any "rules of thumb" when taking images of this kind? And
also do anyone know of work done on this subject before?
Best regards
Pål Eilertsen
| |
| Conway Yee 2005-03-19, 5:38 pm |
| pal.eilertsen@gmail.com (P?l) writes:
> I am working on a project on artifacts in CT images and I am looking
> for information on the subject.
>
> I am looking at the accuracy in CT images taken of the hip with a hip
> implant in place. The implants are made of CoboltCrom or Titanium and
> cause a great deal of artifacts in the image.
>
> Are there any "rules of thumb" when taking images of this kind? And
> also do anyone know of work done on this subject before?
One of the assumptions in filtered backprojection is that the
absorption cross-section of each material is constant. This is
assumption is violated because of the non-monochromatic nature of
CT. The absorption cross-section changes significantly as a function
of wavelength. Thus, synchotron based CT images should have less
artifact.
Conway Yee
| |
|
| Hi,
I am not entirely sure I understand what you mean here. What are a
synchotron based CT. We ordinerily get images from sequential or helix
CTs. They are of the DICOM format.
What I am looking for is information on CT machine settings that might
improve the quality of the images or algorithms for post accusition
processing.
Pål
Conway Yee <yee@bronze.lcs.mit.edu> wrote in message news:<bqw1xadoury.fsf@bronze.lcs.mit.edu>...
> pal.eilertsen@gmail.com (P?l) writes:
>
> One of the assumptions in filtered backprojection is that the
> absorption cross-section of each material is constant. This is
> assumption is violated because of the non-monochromatic nature of
> CT. The absorption cross-section changes significantly as a function
> of wavelength. Thus, synchotron based CT images should have less
> artifact.
>
> Conway Yee
| |
| Conway Yee 2005-04-05, 6:14 pm |
| pal.eilertsen@gmail.com (P?l) writes:
> I am not entirely sure I understand what you mean here. What are a
> synchotron based CT. We ordinerily get images from sequential or helix
> CTs. They are of the DICOM format.
"sequential" and helical CT scanners generate the x-rays using a x-ray
tube. The radiation is inherently polychromatic. The assumption in
filtered backprojection is that the x-ray cross section is constant.
In other words, the probability that photons will be absorbed in a
given tissue is constant per unit distance travelled. Unfortunately,
this is NOT the case. The probability that a photon is absorbed is
dependent on the energy of the photon. Higher energy photons are less
likely to be absorbed. A good book to learn about filtered
backprojection is Kak & Slaney (originally IEEE Press but I think is
is now SIAM). I THINK it is still available for download.
Conway Yee
| |
|
| Hi,
I am not entirely sure I understand what you mean here. What are a
synchotron based CT. We ordinerily get images from sequential or helix
CTs. They are of the DICOM format.
What I am looking for is information on CT machine settings that might
improve the quality of the images or algorithms for post accusition
processing.
Pål
Conway Yee <yee@bronze.lcs.mit.edu> wrote in message news:<bqw1xadoury.fsf@bronze.lcs.mit.edu>...
> pal.eilertsen@gmail.com (P?l) writes:
>
> One of the assumptions in filtered backprojection is that the
> absorption cross-section of each material is constant. This is
> assumption is violated because of the non-monochromatic nature of
> CT. The absorption cross-section changes significantly as a function
> of wavelength. Thus, synchotron based CT images should have less
> artifact.
>
> Conway Yee
| |
| Conway Yee 2005-04-05, 6:14 pm |
| pal.eilertsen@gmail.com (P?l) writes:
> I am not entirely sure I understand what you mean here. What are a
> synchotron based CT. We ordinerily get images from sequential or helix
> CTs. They are of the DICOM format.
"sequential" and helical CT scanners generate the x-rays using a x-ray
tube. The radiation is inherently polychromatic. The assumption in
filtered backprojection is that the x-ray cross section is constant.
In other words, the probability that photons will be absorbed in a
given tissue is constant per unit distance travelled. Unfortunately,
this is NOT the case. The probability that a photon is absorbed is
dependent on the energy of the photon. Higher energy photons are less
likely to be absorbed. A good book to learn about filtered
backprojection is Kak & Slaney (originally IEEE Press but I think is
is now SIAM). I THINK it is still available for download.
Conway Yee
|
|
|
|