OC18.04: Embryonic staging, biometry and volumetry using virtual reality

Abstract

The aim of this study is to stage human embryos using the internal and external characteristics of the Carnegie staging system in an in-vivo setting. We also analyzed the relation of the stages with gestational age (GA), crown-rump length (CRL), embryonic volume and brain ventricle volume. In a prospective cohort study we performed weekly three-dimensional (3D) ultrasound examinations in 112 healthy pregnant women between 6 + 0 and 10 + 1 weeks GA. This resulted in a total of 405 3D ultrasound datasets. These datasets were visualized in an I-Space virtual reality (VR) system using the V-Scope software. V-Scope is used to create a ‘hologram’ of the ultrasound image and allows depth perception and interaction with the rendered objects. The embryo's were staged using the internal and the external characteristics of the Carnegie staging system. The CRL measurements were performed with a tracing-tool and the volume measurements were performed with a semi-automatic segmentation algorithm. The outcomes were analyzed using repeated measurements ANOVA. The embryos could be staged in 74% (n = 299) of the cases. We were able to stage as early as stage 12. Especially the limb development, the curving of the embryo and the brain development were used in this in-vivo staging process. The CRL was measurable in 91% (n = 368) of the datasets, ranging from 2.5 to 41.1 mm. The embryonic volume in 74% (n = 301) of the datasets, ranging from 2.0 to 506.0 mm3. The brain ventricle volume could be measured in 37% (n = 149) of the datasets starting from 7 + 0 weeks GA, ranging from 10.4 to 213.7 mm3. Scatterplots of stages versus GA, CRL, embryonic volume and brain ventricle volume show that log-transformations of both axes result in approximate linear relationships (P < 0.001). This study shows for the first time that embryos can be staged in-vivo, purely based on the internal and external morphologic characteristics, using VR. This staging and the correlation with other growth parameters can be used in the differentiation between normal and abnormal early development.