Paediatric CT protocol optimisation of the abdo-pelvic region

Abstract

Background: Local radiologists had expressed concerns regarding patient dose and image quality for paediatric abdomen-pelvis scans on a GE VCT scanner relative to a Siemens Definition system. Objective: To improve the image quality of a GE Lightspeed VCT scanner relative to the Siemens Definition scanner, whilst keeping the radiation dose to a minimum for paediatric abdomen-pelvis scans. Research Design: Dose and image quality measurements were collected for the full range of weight-based protocols on both scanners using a 32cm CTDI phantom and Catphan 500 image quality phantom. For dose measurements on both scanners axial protocols were created. For image quality, the GE scanner has the iterative reconstruction option (ASIR) available. Results: The radiation dose from the GE scanner was originally higher than from the Siemens scanner prior to optimisation especially for patients weighing between 22.5 - 55 kg with percentage differences ranging from 50% - 96.2%. These doses were reduced by 34 - 50% post optimisation to fall within 21.65% of the Siemens doses. Before any optimisation was done the noise on the GE scanner was between 39% greater to 26% lower than the Siemens scanner and post optimisation the same noise was reduced to 7% greater and 8% lower than the noise on the Siemens system with an increased overall performance of ~ 9%. With Adaptive Statistical Iterative Reconstruction (ASIR) there were dose savings of 34.7% - 50.5% for the 22.5 - 55 kg patients and a noise reduction of ~ 6 - 8% with the same percentage increase for the CNR. Conclusion: The GE Lightspeed VCT scanner achieved broadly similar image quality (noise and contrast-to-noise ratio (CNR)) and better overall performance (figure-of-merit (FOM)) at reduced or similar radiation doses as the Siemens Definition scanner (reference scanner). Recommendations: Adequate training and education about paediatric doses must be provided to radiologists and operators. Another recommendation would be implementing the changes in the GE protocols. Finally, a change in the clinical workflow might also be beneficial to efficiently use both scanners.