At present, we either do TOF or non-TOF backprojection. However, the former is slow and the latter is not quite accurate at the edges of the LOR. @NikEfth and I think this can be relatively easily solved by adding a 3rd option where we do a backprojection with 1 very large TOF bin. In principle, the TOF kernel should adjust.
Relevant code modification is probably mostly here and a few occurences of use_tofsens.
At present, we either do TOF or non-TOF backprojection. However, the former is slow and the latter is not quite accurate at the edges of the LOR. @NikEfth and I think this can be relatively easily solved by adding a 3rd option where we do a backprojection with 1 very large TOF bin. In principle, the TOF kernel should adjust.
Relevant code modification is probably mostly here and a few occurences of
use_tofsens
.