Dear all,
I'm currently working on a project involving projection-based analysis in Dual-Energy CT (DECT), particularly focusing on educational applications and algorithm development. Unfortunately, I do not have access to raw projection data (e.g., sinograms or list-mode data), which are often proprietary and vendor-restricted.
Instead, I’m exploring the feasibility of generating synthetic sinograms from reconstructed DECT images (such as DICOM slices), with the goal of conducting image-domain-based simulation of projection data. This would allow me to test certain concepts such as spectral response, material decomposition, or reconstruction methods from a didactic standpoint.
My main questions are:
Any insights, shared experiences, or references would be greatly appreciated. Thank you!
Your question, though very interesting, requires taking apart quite a few steps in the simulation process. Since even in the existing DECT images, the spectral response of the chain is already fully encoded, it is difficult to separate that out with scans of multiple materials. As a result, if I were doing this for didactic purposes, I would focus on 2 material decomposition, not multi-material. I would also just simulate the objects and work from there. There are multiple other limitations, but that would be a much longer discussion and more appropriate for a PhD dissertation. We are always looking for motivated students!
- Badges
- Science topic
- Similar topics
- Biological Science
- Microbiology
- Virology
- Virus