Evaluation and Comparison of Current Fetal Ultrasound Image Segmentation Methods for Biometric Measurements: A Grand Challenge

Abstract

This paper presents the evaluation results of the methods submitted to Challenge US: Biometric Measurements from Fetal Ultrasound Images, a segmentation challenge held at the IEEE International Symposium on Biomedical Imaging 2012. The challenge was set to compare and evaluate current fetal ultrasound image segmentation methods. It consisted of automatically segmenting fetal anatomical structures to measure standard obstetric biometric parameters, from 2D fetal ultrasound images taken on fetuses at different gestational ages (21 weeks, 28 weeks, and 33 weeks) and with varying image quality to reflect data encountered in real clinical environments. Four independent sub-challenges were proposed, according to the objects of interest measured in clinical practice: abdomen, head, femur, and whole fetus. Five teams participated in the head sub-challenge and two teams in the femur sub-challenge, including one team who tackled both. Nobody attempted the abdomen and whole fetus sub-challenges. The challenge goals were two-fold and the participants were asked to submit the segmentation results as well as the measurements derived from the segmented objects. Extensive quantitative (region-based, distance-based, and Bland-Altman measurements) and qualitative evaluation was performed to compare the results from a representative selection of current methods submitted to the challenge. Several experts (three for the head sub-challenge and two for the femur sub-challenge), with different degrees of expertise, manually delineated the objects of interest to define the ground truth used within the evaluation framework. For the head sub-challenge, several groups produced results that could be potentially used in clinical settings, with comparable performance to manual delineations. The femur sub-challenge had inferior performance to the head sub-challenge due to the fact that it is a harder segmentation problem and that the techniques presented relied more on the femur's appearance.

Difference of Gaussians revolved along elliptical paths for ultrasound fetal head segmentation

“…To obtain outer-to-outer measurements OFD and BPD definitions, as in the manual analysis in the ISBI 2012 challenge data [22], we should increment the center-to-center measures by the calvarial thickness T: OFD := OFD cc + T, BPD := BPD cc + T.…”

“…This DoGEll method won the first prize of the head subchallenge in the Challenge US: Biometric Measurements from Fetal Ultrasound Images, held in conjunction with and supported by the IEEE International Symposium on Biomedical Imaging (ISBI) 2012 in Barcelona, Spain [22]. The contributions of the present paper are two-fold.…”

“…These methods have mostly been based on the segmentation of the ultrasound images [22]. Approaches for the fetal skull segmentation have been based on deformable spline [12] and contour models [4,20], supervised learning [2], and ellipse fitting through (randomized) Hough transform [10,17,27].…”

“…The method has not been published in peer-reviewed literature. A short account of it has been presented in the Challenge proceedings [9] and one paragraph summary was presented in [22]. Second, we present a new, improved implementation of the DoGEll method typically running in under 5 seconds per image, thus yielding a 60 fold savings in computation time as compared to the one winning the segmentation challenge.…”

Automatic image quality assessment and measurement of fetal head in two-dimensional ultrasound image

“…1 The main goals of the fetal US scan are to estimate the gestational age (GA) and weight, confirm growth patterns, and show the presence of possible abnormalities. A set of standard fetal US biometrics is used in routine practice, which includes: crown-rump length, biparietal diameter (BPD), occipital-frontal diameter (OFD), head circumference (HC), femur length (FL), and abdominal circumference.…”