Dose reduction for coronary calcium scoring with hybrid and model-based iterative reconstruction: an ex vivo study

Abstract

Purpose: To determine the influence of dose reduction on coronary calcium scoring using hybrid and model-based iterative reconstruction (IR) techniques.

Methods: Fifteen ex vivo hearts were scanned in a phantom representing an average adult person at routine dose and three levels of dose reduction; 27, 55 and 82% reduced-dose, respectively. All images were reconstructed using filtered back-projection (FBP), hybrid IR (iDose4, levels 1, 4 and 7) as well as model-based IR iterative model reconstruction (IMR, levels 1, 2 and 3). Agatston, mass and volume scores found with iDose4 and IMR were compared to FBP reconstruction (routine dose) as well as objective image quality.

Results: With FBP calcium scores remained unchanged at 82% reduced dose. With IR Agatston scores differed significantly at routine dose, using IMR level 3 and iDose4 level 7, and at 82% reduced dose, using IMR levels 1-3 and iDose4 level 7. The maximum median difference was 5.3%. Mass remained unchanged at reduced dose levels while volume was significantly lower at 82% reduced dose with IMR (maximum median difference 5.0%). Objective image quality improved with IR, at 82% reduced dose the CNR of iDose4 level 7 was similar to the reference dose CNR, and IMR levels 1-3 resulted in an even higher CNR.

Conclusion: Calcium scores were not affected by radiation-dose reduction with FBP and low levels of hybrid IR. Objective image quality increased significantly using hybrid and model-based IR. Therefore low level hybrid IR has the potential to reduce radiation-dose of coronary calcium scoring with up to 82%.