Three-dimensional ultrasound volumetric measurements: is the largest number of image planes necessary for outlining the region-of-interest?

Volumetric measurement is useful in the clinical practice as it can accurately assess the size of organs and lesions, which aids diagnosis and monitoring of treatment response. This in vitro study was undertaken to evaluate and compare the accuracy of three-dimensional ultrasound (3D-US) volumetric measurements using different numbers of image planes for outlining the region-of-interest. A total of 20 tissue phantoms (10 with regular and 10 with irregular tissue specimens) with known volume (2.2 - 14.7 ml) were constructed. The tissue phantoms were scanned with a commercially-available ultrasound machine in conjunction with a 3D add-on system and the volume of the tissue specimen of the phantoms was measured by manual outlining of the boundaries using different numbers of image plane (16, eight, four and two image planes). Results showed that there was a high accuracy in 3D-US volumetric measurements, when 16 and eight image planes were used (best-fit slope between measured and true volume: 0.9224-0.9761 and 0.9099-0.956, respectively). Results also showed that there was no significant difference in the volume measured with 16 and eight image planes for both phantoms with regular- and irregular-shaped tissue specimens (p > 0.05). The measurement differences between 16 and eight image planes ranged from -0.07 to 0.14 ml for phantoms with regular-shaped tissue and from -0.24 to 0.1 ml for phantoms with irregular-shaped tissue. 3D-US has a high accuracy in volumetric measurements of regular- and irregular-shaped structures. Although highest accuracy in volume measurements was found when 16 image planes were used, similar level of accuracy can be achieved with eight image planes but the measurement time can be reduced by 50%. Therefore, in 3D-US volumetric measurements, the largest number of image planes may not be necessary for outlining the region-of-interest.