Artefact at the edge of the FOV

[s_mousavi]

Hello All,

I have reconstructed a the image of a cylindrical phantom using CASToR using the following cases:

1. With normalization:
   a. List-mode data
   b. Histogram mode data
2. Without normalization:
   a. List-mode data
   b. Histogram mode data

The input data is the same for both the list and histogram modes. When reconstructing with normalization, the normalization factors are add to the same data. After reconstruction, the artefact at the edge of the FOV for the histogram mode input datafile has higher intensity compared to the artefact reconstructed from the list mode input data in both items 1 and 2. In addition, the pixel intensity value when reconstructing with normalization is almost 90 times less that the intensities when we construct without normalization. See the profile screenshots attached. (I have used Joseph and incrementalSiddon, however, the general trend is the same.)

Are these observations OK in CASToR or I am making a mistake somewhere?

1. With Normalization Histogram Mode Data (Joseph):

2. With Normalization Histogram Mode Data (IncrementalSiddon):

3. With Normalization List Mode Data (IncrementalSiddon):*

4. Without Normalization Histogram Mode Data (IncrementalSiddon):

5. Without Normalization Histogram Mode Data (Joseph):

Best,

Seyyed

[Maxime_Toussaint]

Hi,

For the difference between list mode and histo., is it possible that you use the field min angle difference and/or Maximum axial difference mm? Both of them are, if my understanding is correct, only applied on list-mode data. If that is indeed the case, your reconstructions in hist. includes more LOR than the one in list-mode. Normaly, these LORs would be outside the image FOV, so it should not have that much an impact on the reconstruction. Nevertheless, it is a possible reason for the difference!

Bests,
Maxime Toussaint