Room: AAPM ePoster Library
Purpose: To compare the Iterative Metal Artifact Reduction (iMAR) and Automated iMAR (AiMAR) algorithms for use in CT-guided microwave ablations.
Methods: A microwave antenna (Certus 140 2.54GHz Ablation Probe, NeuWave Medical) was suspended in a water bath at 60°, 80°, and 90° relative to the z-axis of the CT. A 31.2mm diameter sphere containing 2 mg/mL iodine was placed directly below the antenna. The phantom was scanned on a Siemens Force with scan parameters mimicking the use of biopsy mode for positioning. Two tube voltages (120 kVp and 150 kVp with tin filter [Sn]) were tested. Images were reconstructed with seven anatomical settings of commercially-available iMAR and an AiMAR prototype, which simplifies workflow by prescribing a strength of 1-5. Signal dropout at 90° was assessed by measuring the full-width half-minimum (FWHM) of the signal void on a line profile. Mean CT numbers in the sphere were compared for each angle and kVp.
Results: Without MAR, the FWHM of the signal dropout with the probe at 90° was 3.68 mm at 120 kVp and 2.65 mm at 150Sn kVp. The optimal iMAR anatomical setting by FWHM was Neuro Coils (120 kVp: 2.65 mm; 150Sn kVp: 1.88 mm), and the optimal AiMAR setting was Strength 2 (120 kVp: 3.07 mm; 150Sn kVp: 2.04 mm). At angles of 60° and 80°, all settings of iMAR and AiMAR settings above strength 1 were able to correct the CT number in the sphere to within 6 HU of the mean CT number without metal present at both tube voltages.
Conclusion: Artifact width is comparable between the best anatomical iMAR settings and AiMAR strengths 2 and above, but AiMAR simplifies workflow. For exams in which image contrast is not a limiting factor, using high kVp with additional filtration can result in significantly improved metal artifact reduction.
Funding Support, Disclosures, and Conflict of Interest: This work was funded by Siemens Healthineers.