Abstract

Purpose Model-based iterative reconstruction (MBIR) is a CT reconstruction algorithm that has the potential to provide high quality diagnostic images at very low radiation doses. The purpose of this study is to compare radiation dose, image quality and noise using an ultra low radiation dose (ULD) protocol reconstructed with MBIR to our current standard radiation dose (SD) protocol. Materials and Methods A prospective evaluation of 20 patients (20M;0F; mean age=75 yrs) undergoing surveillance post-EVAR enhanced CT was performed. All patients underwent SD and ULD venous phase imaging (SD/ULD: noise index = 36/51, min mA=100/50, max mA=450/225, kV=100/100). SD images were reconstructed using ASIR (Adaptive Statistical Iterative Reconstruction) per standard of care and compared with ULD images reconstructed with ASIR and MBIR. The volume CT Dose Index (CTDI) was recorded. Quantitative measurements of image noise using the HU standard deviation in 5 separate locations were recorded and averaged. Subjective image assessment (1=worst, 5 = best) of two radiologists blinded to scanning protocol and reconstruction algorithm was performed evaluating diagnostic confidence, image noise, vessel sharpness and aneurysm size. Results The CTDIvol was 73% lower using the ULD protocol (3.4 mGy vs 12.8 mGy for SD). Quantitative image noise measurements were 34% lower with ULD MBIR compared to SD ASIR and 55% lower when compared to ULD ASIR. ULD MBIR and SD ASIR showed no significant difference in averaged subjective noise (3.8 vs 4.0, p=.25). Although vessel sharpness (3.3 vs 4.1, p Conclusion An ULD MBIR protocol performed at 73% lower dose than standard of care imaging provides identical assessment of aneurysm size, high diagnostic confidence and lower image noise and should be considered for post EVAR CT follow-up.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call