Model-based iterative reconstruction in paediatric head computed tomography: a pilot study on dose reduction in children

Introduction
To evaluate the potential of model-based iterative reconstruction (MBIR) on dose reduction and image quality in children undergoing computed tomography (CT) head examinations.

Material and methods
This prospective study was approved by the institutional ethics committee. A total of 88 children (age range of 5 to 16 years) with a history of seizures underwent contrast-enhanced CT scan. Forty-one children underwent CT study according to the MBIR technique, while 47 children underwent CT of the head with the non-MBIR protocol. Images were reviewed by 2 blinded paediatric radiologists in a random order. Mean dose-length product, CT dose index (CTDI) volume, and mean effective dose were recorded for both groups. Image quality, image noise, and diagnostic acceptability of 2 image sets were also recorded.

Results
In the MBIR group, the mean dose-length product was reduced by 79.8%; the mean CTDI volume was reduced by 88.5%, while the mean effective dose was reduced by 81% when compared to the non-MBIR group. No significant difference was seen in diagnostic acceptability, image noise, and image quality between the 2 groups.

Conclusions
MBIR technique is highly effective in reducing radiation dose in paediatric head CT examinations without any significant difference in image quality, image noise, and diagnostic acceptability.