Energy based clutter filtering for vector flow imaging

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

To obtain accurate blood flow velocity estimates it is important to remove the clutter signal originating from tissue. Conventionally, the clutter signal has been separated from the blood signal based on the difference of their spectral frequencies. However, this approach is not enough for obtaining vector flow measurements, since the spectra overlaps at high beam-to-flow angles. In this work a distinct approach is proposed, where the energy of the velocity spectrum is used to differentiate among the two signals. The energy based method is applied by limiting the amplitude of the velocity spectrum function to a predetermined threshold. The effect of the clutter filtering is evaluated on a plane wave (PW) scan sequence in combination with transverse oscillation (TO) and directional beamforming (DB) for velocity estimation. The performance of the filter is assessed by comparison of the velocity estimates of the proposed filter against a conventional moving average clutter filter. The effect of tissue motion is investigated using a Field II simulation of a straight vessel with moving wall, while the direct effect of the filter on the velocity estimates is evaluated on a CFD model of a carotid bifurcation with a fixed vessel wall. The results show that the proposed filter outperformed the moving average during moving vessel wall conditions, where standard deviations from the velocity magnitudes and angles were kept consistently below 6% and 6° compared to 63% and 48° on the moving average filter. The results on the CFD showed that on non-moving conditions the velocity estimates had minor statistical differences with errors on the magnitude of -7.95±10.1% and angles of 0.15±6.65° for the proposed filter compared to -5.83±9.08% and -0.12±4.48°.

Original languageEnglish
Title of host publication2017 IEEE International Ultrasonics Symposium, IUS 2017
Number of pages4
PublisherIEEE Computer Society Press
Publication date31 Oct 2017
Article number8092639
ISBN (Electronic)9781538633830
Publication statusPublished - 31 Oct 2017
Event2017 IEEE International Ultrasonics Symposium, IUS 2017 - Washington, United States
Duration: 6 Sep 20179 Sep 2017


Conference2017 IEEE International Ultrasonics Symposium, IUS 2017
LandUnited States
SeriesIEEE International Ultrasonics Symposium, IUS

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

ID: 331496742