Tracking performance in ultrasound super-resolution imaging
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
Tracking performance in ultrasound super-resolution imaging. / Taghavi, Iman; Andersen, Sofie B.; Hoyos, Carlos A.Villagomez; Schou, Mikkel; Oygard, Sigrid H.; Gran, Fredrik; Hansen, Kristoffer L.; Sorensen, Charlotte M.; Nielsen, Michael B.; Stuart, Matthias B.; Jensen, Jorgen Arendt.
IUS 2020 - International Ultrasonics Symposium, Proceedings. IEEE, 2020. 9251605.Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Tracking performance in ultrasound super-resolution imaging
AU - Taghavi, Iman
AU - Andersen, Sofie B.
AU - Hoyos, Carlos A.Villagomez
AU - Schou, Mikkel
AU - Oygard, Sigrid H.
AU - Gran, Fredrik
AU - Hansen, Kristoffer L.
AU - Sorensen, Charlotte M.
AU - Nielsen, Michael B.
AU - Stuart, Matthias B.
AU - Jensen, Jorgen Arendt
PY - 2020/9/7
Y1 - 2020/9/7
N2 - Tracking plays an important role in super-resolution (SR) ultrasound imaging, as it improves the quality and sharpness of the final SR images. Moreover, tracking enables quantification of clinically important parameters, such as blood flow velocity. However, the tracking performance degrades in the presence of complex particle patterns and localization uncertainty due to noise and motion. This work presents and discusses multiple approaches for tracking evaluation and compares a nearest-neighbor (NN) with a Kalman tracker through simulations and an in vivo experiment. It is shown that in the presence of a localization uncertainty with a standard deviation (SD) of ?/5, the bias and SD of the velocity estimates reach -1.04 ± 0.9 mm/s and -0.12 ± 0.72 mm/s in the NN and Kalman tracker, respectively (relative to the peak velocity of 10 mm/s). The precision of individual track positions is estimated for an in vivo experiment as 37.95 ± 21.37 µm and 23.9 ± 11.82 µm for the NN and Kalman trackers, respectively. The results indicate that the Kalman tracker achieves a better velocity estimation and reduces localization uncertainty.
AB - Tracking plays an important role in super-resolution (SR) ultrasound imaging, as it improves the quality and sharpness of the final SR images. Moreover, tracking enables quantification of clinically important parameters, such as blood flow velocity. However, the tracking performance degrades in the presence of complex particle patterns and localization uncertainty due to noise and motion. This work presents and discusses multiple approaches for tracking evaluation and compares a nearest-neighbor (NN) with a Kalman tracker through simulations and an in vivo experiment. It is shown that in the presence of a localization uncertainty with a standard deviation (SD) of ?/5, the bias and SD of the velocity estimates reach -1.04 ± 0.9 mm/s and -0.12 ± 0.72 mm/s in the NN and Kalman tracker, respectively (relative to the peak velocity of 10 mm/s). The precision of individual track positions is estimated for an in vivo experiment as 37.95 ± 21.37 µm and 23.9 ± 11.82 µm for the NN and Kalman trackers, respectively. The results indicate that the Kalman tracker achieves a better velocity estimation and reduces localization uncertainty.
U2 - 10.1109/IUS46767.2020.9251605
DO - 10.1109/IUS46767.2020.9251605
M3 - Article in proceedings
AN - SCOPUS:85097910746
BT - IUS 2020 - International Ultrasonics Symposium, Proceedings
PB - IEEE
T2 - 2020 IEEE International Ultrasonics Symposium, IUS 2020
Y2 - 7 September 2020 through 11 September 2020
ER -
ID: 253691159