N A C Smeets1, B Winkens, S G Oei. 1. Department of Gynecology and Obstetrics, Atrium Medical Centre, Parkstad, The Netherlands. n.smeets @ atriummc.nl
Abstract
BACKGROUND: The goal of this study was to calculate the accuracy of three-dimensional volume measurements with the rotational multiplanar technique VOCAL (Volume Organ Computer-aided AnaLysis). METHODS: An ultrasound phantom with 13 objects (volume 10.2-40.5 cm(3)) was created. After data acquisition, the volumes of the objects were measured with the VOCAL technique. RESULTS: A linear mixed model analysis showed a significant linear (B = -0.008, 95% CI -0.014, -0.002, p = 0.005) and a significant quadratic trend (B = 0.0001, 95% CI 0.000004, 0.0002, p = 0.040). The absolute error increases significantly with the increasing volume of interest with 0.044 cm(3) for each cm(3) rise of the volume of interest. The actual volume increased from an initial value of 10 cm(3) to values of 20, 30 and 40 cm(3), resulting in a decrease in the mean estimated percentage error from 15.4 to 10.6%, 8.0% and 7.5%. CONCLUSION: The results of this study showed that volume measurements with three-dimensional ultrasound and the VOCAL software can be used for volume measurements in vivo. However, it is important to assess the measurement error and to consider these error values when interpreting the results in daily practice.
BACKGROUND: The goal of this study was to calculate the accuracy of three-dimensional volume measurements with the rotational multiplanar technique VOCAL (Volume Organ Computer-aided AnaLysis). METHODS: An ultrasound phantom with 13 objects (volume 10.2-40.5 cm(3)) was created. After data acquisition, the volumes of the objects were measured with the VOCAL technique. RESULTS: A linear mixed model analysis showed a significant linear (B = -0.008, 95% CI -0.014, -0.002, p = 0.005) and a significant quadratic trend (B = 0.0001, 95% CI 0.000004, 0.0002, p = 0.040). The absolute error increases significantly with the increasing volume of interest with 0.044 cm(3) for each cm(3) rise of the volume of interest. The actual volume increased from an initial value of 10 cm(3) to values of 20, 30 and 40 cm(3), resulting in a decrease in the mean estimated percentage error from 15.4 to 10.6%, 8.0% and 7.5%. CONCLUSION: The results of this study showed that volume measurements with three-dimensional ultrasound and the VOCAL software can be used for volume measurements in vivo. However, it is important to assess the measurement error and to consider these error values when interpreting the results in daily practice.