Experimental setup for objective evaluation of uterine motion analysis by ultrasound speckle tracking. (13th March 2018)

• Record Type: Journal Article
• Authors: Sammali, F; Blank, C; Xu, L; Huang, Y; Kuijsters, N P M; Schoot, B C; Mischi, M
• Is Part Of: Biomedical physics & engineering express. Volume 4:Number 3(2018)
• Publication Date: 2018-03-13
• Subjects: experimental setup -- speckle tracking -- motion analysis -- uterine motion -- block matching -- uterus; Medical physics -- Periodicals; Biophysics -- Periodicals; Biomedical engineering -- Periodicals; Medical sciences -- Periodicals; 610.153
• DOI: 10.1088/2057-1976/aab053
• Languages: English
• ISSNs: 2057-1976

Abstract

Abstract: Recent research indicates uterine contractions to play an important role for the success of conception. Unfortunately, the lack of tools for quantitative analysis limits our understanding of the uterine contractility outside pregnancy. More recently, several ultrasound methods based on speckle tracking have gained attention for assessment of the uterine contractility; however, the absence of a ground truth hampers the optimization of the proposed tracking methods. To overcome this limitation, here we present an experimental setup based on a human ex-vivo uterus that generates controlled uterine motion for assessing the accuracy of motion tracking methods. Uterine motion was obtained by driving an electromagnetic actuator that produces controlled, sinusoidal displacement of a syringe piston connected to a balloon catheter filled with saline and inserted into the uterine cavity. This way, controlled uterine motion was generated while maintaining the original speckle characteristics. For validation, six needle markers were inserted in three ex-vivo uteri and tracked over time by optical flow. The motion of these markers was then compared with the driving signal of the electromagnetic actuator to assess the reproducibility of the produced uterine motion. Correlation coefficient ( r ), mean square error (MSE), and Hausdorff distance ( H d ) were the adopted objective metrics to evaluate the agreement between the signals. A use case is also presented where the proposed setup was tested to evaluate the speckle-tracking performance of block-matching by sum of absolute differences (SAD) with different block sizes. SAD with block size = 1.724 × 1.724 mm 2 showed the highest agreement with the reference motion of the needle markers. To conclude, the proposed experimental setup generates controlled, rhythmic and reproducible uterine tissue motion representing a reliable ground truth for objective evaluation of speckle-tracking methods aimed at quantitative analysis of uterine motion outside pregnancy.