Room: Track 1
Purpose: evaluate the resolution properties of a statistical-based hybrid iterative reconstruction algorithm (AIDR 3D) and a new model-based iterative reconstruction algorithm (FIRST) for single-energy CT (SECT) and dual-energy CT (DECT) acquisitions. The task-based modulation transfer function (MTF-task) was calculated to compare spatial resolution across a series of dose levels and contrast levels for single-energy and dual-energy acquisitions.
Methods: ACR phantom was modified to fill the air space with a urethane rubber mixture to produce a -35 HU low-contrast disk. The phantom was scanned on a Canon Aquilion Genesis scanner using 120-kVp SECT with CTDIvol of 2.8 mGy, 19.7 mGy, and 29.4 mGy. The DECT scan was acquired at 18.1 mGy, and a 120 kVp-equivalent image was created for comparison with the SECT scans. Each acquisition was repeated 10 times, and all images were reconstructed using the AIDR 3D and FIRST reconstruction algorithms with settings appropriate for abdominal CT exams. The MTF-task was calculated using the circular-edge technique averaged across 10 1-mm slices from each of the 10 acquisitions. Measurements were performed for the Teflon, acrylic, polyethylene, and low-contrast insert (LCI) in ACR Phantom Module 1.
Results: for the AIDR 3D SECT acquisitions appeared equivalent across all dose and contrast levels. Conversely, the FIRST SECT reconstructions showed that spatial resolution improved with increasing contrast. The DECT 120 kVp-equivalent images also showed an improvement in the MTF-task using the FIRST algorithm over AIDR. The MTF-task for the AIDR 3D SECT acquisition was equivalent to with the AIDR 3D DECT 120 kVp-equivalent image at similar dose levels. However, the SECT MTF-task outperformed the DECT MTF-task using the FIRST reconstruction algorithm.
Conclusion: offers significant improvements in MTF-task over AIDR for all acquisitions. Although the AIDR MTF-task is unaffected by dose and contrast levels, the FIRST MTF-task is dependent on both.
Funding Support, Disclosures, and Conflict of Interest: This work was funded by Canon Medical Systems USA, Inc.