Related Experiment Videos
Propagation of errors from the sensitivity image in list mode reconstruction
1Department of Nuclear Medicine and Functional Imaging, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA. jqi@lbl.gov
IEEE Transactions on Medical Imaging
|September 21, 2004
Summary
List mode image reconstruction requires accurate sensitivity images. This study analyzes how errors in sensitivity images affect reconstructed images, providing a frequency-dependent error expression for improved accuracy in medical imaging.
Area of Science:
- Medical Imaging
- Image Reconstruction
- Positron Emission Tomography (PET)
Background:
- List mode reconstruction offers advantages by eliminating empty sinogram bins.
- Accurate sensitivity image calculation is crucial but challenging for modern scanners due to numerous lines of response (LORs).
Purpose of the Study:
- To analyze error propagation from sensitivity images to reconstructed images in list mode reconstruction.
- To derive a method for determining the required accuracy of sensitivity image calculations.
Main Methods:
- Theoretical analysis based on the fixed point condition and Kuhn-Tucker conditions for nonnegativity constraint.
- First-order Taylor series approximation to derive a closed-form expression for image error.
- Computer simulations to validate theoretical findings.
Main Results:
- A frequency-dependent error response from the sensitivity image to the reconstructed image was derived.
- A simple expression quantifies the relationship between sensitivity image error and reconstructed image error.
- Simulation results confirmed the accuracy of the theoretical analysis.
Conclusions:
- The derived expression aids in setting appropriate accuracy requirements for sensitivity image computation.
- This work provides a theoretical basis for optimizing list mode image reconstruction in PET.
- Understanding error propagation is key to improving image quality and diagnostic accuracy.