Anti-aliasing weighting functions for single-slice helical CT
Patrick J La Rivière1, Xiaochuan Pan
1Department of Radiology, The University of Chicago, 5841 S. Maryland Ave., MC-1037, Chicago, IL 60637, USA. pjlarivi@midway.uchicago.edu
IEEE Transactions on Medical Imaging
|December 11, 2002
Summary
New helical computed tomography methods, 180AA and 180BSP, reduce longitudinal aliasing artifacts. These spatial-domain approaches offer improved image quality and resolution compared to standard methods, enabling faster processing.
Area of Science:
- Medical Imaging
- Computed Tomography
- Image Reconstruction
Background:
- Longitudinal aliasing in helical computed tomography (CT) degrades image resolution and distorts structures.
- Existing Fourier-based interpolation methods eliminate aliasing but are computationally intensive and not pipelined.
Purpose of the Study:
- To develop computationally efficient, pipelined methods for reducing longitudinal aliasing in helical CT.
- To approximate the performance of a Fourier-based approach using spatial-domain techniques.
Main Methods:
- Derived two spatial-domain weighting functions: 180AA (direct approximation) and 180BSP (B-spline approximation).
- Evaluated aliasing suppression and resolution properties using simulated CT data.
- Compared performance against the standard 180LI approach.
Main Results:
- Both 180AA and 180BSP methods approximate the aliasing suppression of the Fourier-based approach.
- These methods offer improvements in aliasing suppression and resolution over the standard 180LI technique.
- The new methods allow for pipelined implementation, addressing computational limitations.
Conclusions:
- 180AA and 180BSP provide effective, computationally feasible solutions for longitudinal aliasing in helical CT.
- These spatial-domain methods enhance image quality and reconstruction efficiency.
- Represent a significant advancement over current clinical standards for helical CT reconstruction.
