OP31.05: the Accuracy and Validity of the XI VOCAL™ Three‐dimensional Ultrasound Volumetric Measurement Using an in Vitro Model

Ultrasonography (US) to follow tumor volume fluctuation during the therapy of the oncologic patient should be an accurate method. The aim of the study was to determine the accuracy and validity of measurements of regular- and irregular-shape phantom objects of known volume using the novel XI VOCAL™ (eXtended Imaging Virtual Organ Computer-aided AnaLysis™) three-dimensional (3D) US volumetry. Three-dimensional US volumetric measurements by XI VOCAL were performed on 20 different 3D-US datasets of ultrasound images obtained from 4 water-filled objects at the Department of Obstetrics and Gynecology of the University of Pécs, Hungary. For each 3D dataset 20 volumes were calculated using XI VOCAL (with 5, 10, 15 and 20 slices) technique. All measurements were performed independently by two observers who were blind to the actual volumes of the objects. Accuracy and validity were assessed by examining the percentage difference from the true volume. The true volumes of the objects were determined by a modified Archimedes water displacement technique. The differences between the calculated volume by 3D-US technique and of the true volume were typically negative (underestimation). Most of the calculated volumes were significantly smaller compared to the true volumes assessed by the ANOVA. Inaccurate correlations were found for both observers in an in vitro setting with the use of different multi-sliced XI VOCAL 3D-US volumetric measurements. The accuracy of XI VOCAL depends on the observer's interpretation with respect to where the object starts and finishes. This interpretation can be the most difficult aspect of volume measurement by the XI VOCAL technique and could be responsible for the significant bias of our results.