Automated Fetal Femur Length Measurement by Five-Dimensional Ultrasound

Abstract

Accurate gestational dating is of paramount importance and the cornerstone for management of pregnancies, easily reproducible sonographic fetal biometric parameters for gestational dating are clinically important for the optimal obstetric management of pregnancies. This is especially true in determining timing of a variety of gestational tests, assessing adequacy of growth and timing of delivery for the optimal obstetric outcome (Chavez et al., 2007). Five-dimensional Long Bone (5D LB), a new technique that automatically archives, reconstructs images, and measures lengths of fetal long bones (Hur et al, 2015). Furthermore, 3D-ultrasound provides the ability to store volume data that can be manipulated after the patient has left the examination room and also be transmitted electronically to evaluate elsewhere. Despite these advances of 3D-ultrasound, the manipulation of volume data requires a learning curve and is also operator dependent. Automated 3D imaging will allow an operator independent and standardized approach to evaluate complex anatomical evaluation and improve the efficiency by reducing the time to complete the ultrasound examination. Buoyed by the recent advancements in ultrasound technology, numerous studies have been conducted to develop a more efficient and precise automated ultrasound system for biometry (Hur et al, 2015). The aim of this study is to evaluate the accuracy of Five-Dimensional Ultrasound in automated measurement of the fetal femur length at third trimester. This study was carried out at the Fetal Care Unit at Ain Shams University Maternity Hospital from April 2015 to December 2015, Ninety pregnant women were recruited from the Fetal Care Unit who fulfilled the inclusion criteria. Verbal consent was obtained from participants who were included into the study. All women were subjected to examination by Two-Dimensional Ultrasound and Five-Dimensional Ultrasound. Results: In this study, 5D US estimation of gestational age showed non-significant difference (P>0.05) with 2D US for estimation of gestational age. In this study, the accuracy of 2D and 5D US was compared by running the paired Student t test on the estimated mean error (systematic error) for either assessment tool.: p-value of signed error (weeks) = 0.931, p-value of Signed percentage error (%) = 0.929, p-value of absolute (unsigned) error (weeks) = 0.204 and p-value of absolute (unsigned) percentage error (%) = 0.177. In this study, The precision of 2D and 5D US was compared by running the Pitman t test for comparison of paired variances on the variance of the error of 2D US or 5D US: p-value of signed error (weeks) = 0.735, p-value of Signed percentage error (%) = 0.902, p-value of absolute (unsigned) error (weeks) = 0.174 and p-value of absolute (unsigned) percentage error (%) = 0.201. In this study, The Bland-Altman method was used to examine inter-method agreement between 2D US and 5D US for estimation of gestational age. The Mean of difference between 2D and 5D US was -0.01 with SD of mean difference was 1.49, 95% CI for lower limit of agreement from -3.47 to -2.40 and 95% CI for upper limit of agreement from 2.37 to 3.44.