A particle filter-based target tracking algorithm for MR-guided respiratory compensation: robustness and accuracy assessment

Publication date: Available online 12 October 2017
Source:International Journal of Radiation Oncology*Biology*Physics
Author(s): Alexandra E. Bourque, Stéphane Bedwani, Jean-François Carrier, Cynthia Ménard, Pim Borman, Clemens Bos, Bas Raaymakers, Nikolai Mickevicius, Eric Paulson, Rob. H.N. Tijssen
PurposeTo assess overall robustness and accuracy of a modified particle filter based tracking algorithm for MR-guided radiation therapy treatments.MethodsAn improved particle filter based tracking algorithm is implemented, which employs a normalized cross-correlation function as the likelihood calculation. With a total of 5 healthy volunteers and 8 patients, the robustness of the algorithm is tested on 24 dynamic MRI time-series with varying resolution, contrast, and signal-to-noise ratio. The complete data set includes data acquired with different scan parameters on a number of MRI scanners with varying field strengths including the 1.5T MR-linac. Tracking errors are computed by comparing the results obtained by the particle filter algorithm to experts' delineations.ResultsThe ameliorated tracking algorithm is able to accurately track abdominal as well as thoracic tumors, whereas the previous Bhattacharyya distance based implementation fails in over 50% of the cases. The tracking error, combined over all MRI acquisitions, is (1.1 ± 0.4) mm, which demonstrates high robustness against variations in contrast, noise and image resolution. Finally, the effect of the input/control parameters of the model is very similar across all cases suggesting a class-based optimization is possible.ConclusionThe modified particle filter tracking algorithm is highly accurate and robust against varying image quality. This makes the algorithm a promising candidate for automated tracking on the MR-linac.

Teaser

A modified particle filter based algorithm allows tracking of abdominal and thoracic anatomies of interest. A total of 5 healthy volunteers and 8 patients were scanned with various image contrasts and resolutions. The results demonstrate an accurate and robust technique with promising clinical potential.


http://ift.tt/2z7FXpd