Iterative deblurring for CT metal artifact reduction.
G Wang1, D L Snyder, J A O'Sullivan
1Mallinckrodt Inst. of Radiol., Washington Univ., St. Louis, MO.
IEEE Transactions on Medical Imaging
|January 1, 1996
Summary
This study adapted iterative deblurring methods for metal artifact reduction in computed tomography (CT). Both expectation maximization (EM) and algebraic reconstruction technique (ART) algorithms improved image quality over traditional methods.
Area of Science:
- Medical Imaging
- Image Reconstruction
- Computational Imaging
Background:
- Metal artifacts significantly degrade image quality in medical computed tomography (CT).
- Iterative deblurring techniques offer potential for artifact reduction.
Purpose of the Study:
- To adapt and compare expectation maximization (EM) and algebraic reconstruction technique (ART) algorithms for metal artifact reduction in CT.
- To evaluate the performance of these iterative methods against filtered backprojection.
Main Methods:
- Implementation of EM and ART iterative deblurring algorithms for metal artifact reduction.
- Experimental validation using synthetic noise-free and noisy projection data from dental phantoms.
- Comparison of image quality and convergence rates with filtered backprojection.
Main Results:
- Both EM and ART algorithms demonstrated superior image quality compared to filtered backprojection.
- The EM algorithm exhibited faster convergence than the ART algorithm.
- EM-based deblurring yielded better image clarity but increased noise compared to ART.
Conclusions:
- Iterative deblurring methods, particularly EM and ART, are effective for metal artifact reduction in CT.
- Algorithm choice (EM vs. ART) involves a trade-off between image clarity and noise levels.
- Computational complexity is comparable for EM and ART, dominated by reprojection and backprojection steps.
