Speckle tracking in a phantom and feature-based tracking in liver in the presence of respiratory motion using 4D ultrasound
Emma J Harris,Naomi R Miller,Jeffrey C Bamber,J Richard N Symonds-Tayler,Philip M Evans
DOI: https://doi.org/10.1088/0031-9155/55/12/007
IF: 3.5
2010-05-26
Physics in Medicine and Biology
Abstract:We have evaluated a 4D ultrasound-based motion tracking system developed for tracking of abdominal organs during therapy. Tracking accuracy and precision were determined using a tissue-mimicking phantom, by comparing tracked motion with known 3D sinusoidal motion. The feasibility of tracking 3D liver motion in vivo was evaluated by acquiring 4D ultrasound data from four healthy volunteers. For two of these volunteers, data were also acquired whilst simultaneously measuring breath flow using a spirometer. Hepatic blood vessels, tracked off-line using manual tracking, were used as a reference to assess, in vivo, two types of automated tracking algorithm: incremental (from one volume to the next) and non-incremental (from the first volume to each subsequent volume). For phantom-based experiments, accuracy and precision (RMS error and SD) were found to be 0.78 mm and 0.54 mm, respectively. For in vivo measurements, mean absolute distance and standard deviation of the difference between automatically and manually tracked displacements were less than 1.7 mm and 1 mm respectively in all directions (left-right, anterior-posterior and superior-inferior). In vivo non-incremental tracking gave the best agreement. In both phantom and in vivo experiments, tracking performance was poorest for the elevational component of 3D motion. Good agreement between automatically and manually tracked displacements indicates that 4D ultrasound-based motion tracking has potential for image guidance applications in therapy.
engineering, biomedical,radiology, nuclear medicine & medical imaging