EP22.16: Comparison of a 2D and 3D measurement of the intracranial translucency using post‐processing software

Abstract

The aim of the study is to determine the ability to accurately measure the Intracranial Translucency (fourth ventricle) using a stored volume data set compared to the standardised 2 dimensional (2D) image. This is a study on women who attended the Perinatal Ultrasound Department for their nuchal translucency scan, at which time a 2D measurement of the intracranial translucency was recorded. This is considered the gold standard measurement. A 3 dimensional (3D) sweep of the fetal brain was also performed on the same patient by a qualified sonographer or doctor working within the Perinatal Ultrasound unit. The image criteria used were the same as recommended by the Fetal Medicine Foundation for the measurement of the nuchal translucency. Post processing analysis of the 3D volume data set was performed by the four study investigators, using Viewpoint software, who were blinded to the standard 2D measurement of the intracranial translucency. Evaluation for consistency of 3D measurements was undertaken and summarised with an Inter-rater Correlation Coefficient (ICC). The mean measure of the four investigators was compared to the 2D measurement by ICC and visualised with a Bland-Altman plot. In this analysis by four operators of 173 3D data sets there was a low level of agreement, ICC = 0.227 95%CI (0.15, 0.313). The average 3D measurement correlated poorly with the 2D gold standard, ICC = 0.22 95%CI (0.076, 0.355). Our study shows that measurement of the Intracranial Translucency using a 3D data set does not appear to be a reliable method of assessment. This reliability may improve as the image acquisition technology develops.