The reliability of virtual organ computer‐aided analysis (VOCAL) for the semiquantification of ovarian, endometrial and subendometrial perfusion

Publisher Website
Google Scholar
Abstract
Objectives Three‐dimensional power Doppler angiography (3D‐PDA) has been largely used for the subjective assessment of vascular patterns but semiquantification of the power Doppler signal is now possible. We examined the intraobserver and interobserver reliability of the semiquantification of ovarian, endometrial and subendometrial blood flow using 3D‐PDA, virtual organ computer‐aided analysis (VOCAL™) and shell‐imaging. Methods 3D‐PDA was used to acquire 20 ovarian and 20 endometrial volumes from 40 different patients at various stages of in vitro fertilization treatment. VOCAL was then used to delineate the 3D areas of interest and the ‘histogram facility’ employed to generate three indices of vascularity: the vascular index, the flow index and the vascularization flow index. Intraobserver and interobserver reliability was assessed by two‐way, mixed, intraclass correlation coefficients (ICCs) and general linear modeling was used to examine for differences in the mean values between each observer. Results The intraobserver reliability for both observers was extremely high and there were no differences in reliability between the observers for measurements of both volume and vascularity within the ovary or endometrium and its shells. With the exception of the outside subendometrial shell volumes, there were no significant differences between the two observers in the mean values obtained for either endometrial or ovarian volume and vascularity measurements. The interobserver reliability of measurements was equally high throughout with all measurements obtaining a mean ICC of above 0.985. Conclusions 3D‐PDA and shell‐imaging offer a reliable, practical and non‐invasive method for the assessment of ovarian, endometrial and subendometrial blood flow. Future work should concentrate upon confirming the reliability of data acquisition and the validity of the technique before its predictive value can be truly tested in prospective clinical studies. Copyright © 2003 ISUOG. Published by John Wiley & Sons, Ltd.

This publication has 23 references indexed in Scilit: