CHU Brest, Brest, 29200, France.
LaTIM, INSERM, UMR 1101, SFR IBSAM, 29200, Brest, France.
Int J Comput Assist Radiol Surg. 2018 Nov;13(11):1707-1716. doi: 10.1007/s11548-018-1856-x. Epub 2018 Sep 7.
A new algorithm, based on fully convolutional networks (FCN), is proposed for the automatic localization of the bone interface in ultrasound (US) images. The aim of this paper is to compare and validate this method with (1) a manual segmentation and (2) a state-of-the-art method called confidence in phase symmetry (CPS).
The dataset used for this study was composed of 1738 US images collected from three volunteers and manually delineated by three experts. The inter- and intra-observer variabilities of this manual delineation were assessed. Images having annotations with an inter-observer variability higher than a confidence threshold were rejected, resulting in 1287 images. Both FCN-based and CPS approaches were studied and compared to the average inter-observer segmentation according to six criteria: recall, precision, F1 score, accuracy, specificity and root-mean-square error (RMSE).
The intra- and inter-observer variabilities were inferior to 1 mm for 90% of manual annotations. The RMSE was 1.32 ± 3.70 mm and 5.00 ± 7.70 mm for, respectively, the FCN-based approach and the CPS algorithm. The mean recall, precision, F1 score, accuracy and specificity were, respectively, 62%, 64%, 57%, 80% and 83% for the FCN-based approach and 66%, 34%, 41%, 52% and 43% for the CPS algorithm.
The FCN-based approach outperforms the CPS algorithm, and the obtained RMSE is similar to the manual segmentation uncertainty.
提出了一种新的基于全卷积网络（FCN）的算法，用于自动定位超声（US）图像中的骨界面。本文的目的是将该方法与（1）手动分割和（2）称为相位对称置信度（CPS）的最新方法进行比较和验证。
该研究使用的数据集由三名志愿者采集的 1738 张 US 图像组成，由三位专家进行手动勾画。评估了这种手动勾画的组内和组间可变性。具有高于置信度阈值的组间可变性的注释的图像被拒绝，导致 1287 张图像。研究了基于 FCN 的方法和 CPS 方法，并根据六个标准（召回率、精确度、F1 得分、准确性、特异性和均方根误差（RMSE））与平均组间分割进行了比较。
90%的手动注释的组内和组间可变性小于 1 毫米。基于 FCN 的方法和 CPS 算法的 RMSE 分别为 1.32 ± 3.70 毫米和 5.00 ± 7.70 毫米。基于 FCN 的方法的平均召回率、精确度、F1 得分、准确性和特异性分别为 62%、64%、57%、80%和 83%，而 CPS 算法的分别为 66%、34%、41%、52%和 43%。
基于 FCN 的方法优于 CPS 算法，并且得到的 RMSE 与手动分割的不确定性相似。
