Head CT: Image quality improvement with ASIR-V using a reduced radiation dose protocol for children.
Eur Radiol
; 27(9): 3609-3617, 2017 Sep.
Article
em En
| MEDLINE
| ID: mdl-28116512
ABSTRACT
OBJECTIVE:
To investigate the quality of images reconstructed with adaptive statistical iterative reconstruction V (ASIR-V), using pediatric head CT protocols.METHODS:
A phantom was scanned at decreasing 20% mA intervals using our standard pediatric head CT protocols. Each study was then reconstructed at 10% ASIR-V intervals. After the phantom study, we reduced mA by 10% in the protocol for <3-year-old patients and applied 30% ASIR-V and by 30% in the protocol for 3- to 15-year-old patients and applied 40% ASIR-V.RESULTS:
Increasing the percentage of ASIR-V resulted in lower noise and higher contrast-to-noise ratio (CNR) and preserved spatial resolution in the phantom study. Compared to a conventional-protocol, reduced-dose protocol with ASIR-V achieved 12.8% to 34.0% of dose reduction and showed images of lower noise (9.22 vs. 10.73, P = 0.043) and higher CNR in different levels (centrum semiovale, 2.14 vs. 1.52, P = 0.003; basal ganglia, 1.46 vs. 1.07, P = 0.001; and cerebellum, 2.18 vs. 1.33, P < 0.001). Qualitative analysis showed higher gray-white matter differentiation and sharpness and preserved overall diagnostic quality in the images with ASIR-V.CONCLUSIONS:
Use of ASIR-V allowed a 12.8% to 34.0% dose reduction in each age group with potential to improve image quality. KEY POINTS ⢠It is possible to reduce radiation dose and improve image quality with ASIR-V. ⢠We improved noise and CNR and decreased radiation dose. ⢠Sharpness improved with ASIR-V. ⢠Total radiation dose was decreased by 12.8% to 34.0%.Palavras-chave
Texto completo:
1
Base de dados:
MEDLINE
Assunto principal:
Encéfalo
/
Tomografia Computadorizada por Raios X
Tipo de estudo:
Evaluation_studies
/
Guideline
/
Observational_studies
/
Qualitative_research
Limite:
Adolescent
/
Child
/
Child, preschool
/
Female
/
Humans
/
Male
Idioma:
En
Ano de publicação:
2017
Tipo de documento:
Article