Detail publikace

Feature Drift Resilient Tracking of the Carotid Artery Wall Using Unscented Kalman Filtering With Data Fusion

DORAZIL, J. REPP, R. KROPFREITER, T. PRÜLLER, R. ŘÍHA, K. HLAWATSCH, F.

Originální název

Feature Drift Resilient Tracking of the Carotid Artery Wall Using Unscented Kalman Filtering With Data Fusion

Anglický název

Feature Drift Resilient Tracking of the Carotid Artery Wall Using Unscented Kalman Filtering With Data Fusion

Jazyk

en

Originální abstrakt

An analysis of the motion of the common carotid artery (CCA) provides effective indicators for cardiovascular diseases. Here, we propose a method for tracking CCA wall motion from a B-mode ultrasound video sequence. An unscented Kalman filter based on a suitably devised state-space model fuses measurements produced by an optical flow algorithm and a CCA wall localization algorithm. This approach compensates for feature drift, which is a detrimental effect in optical flow algorithms. The proposed method is demonstrated to outperform a state-of-the-art tracking method based on optical flow.

Anglický abstrakt

An analysis of the motion of the common carotid artery (CCA) provides effective indicators for cardiovascular diseases. Here, we propose a method for tracking CCA wall motion from a B-mode ultrasound video sequence. An unscented Kalman filter based on a suitably devised state-space model fuses measurements produced by an optical flow algorithm and a CCA wall localization algorithm. This approach compensates for feature drift, which is a detrimental effect in optical flow algorithms. The proposed method is demonstrated to outperform a state-of-the-art tracking method based on optical flow.

Dokumenty

BibTex


@inproceedings{BUT164233,
  author="Jan {Dorazil} and Rene {Repp} and Thomas {Kropfreiter} and Richard {Prüller} and Kamil {Říha} and Franz {Hlawatsch}",
  title="Feature Drift Resilient Tracking of the Carotid Artery Wall Using Unscented Kalman Filtering With Data Fusion",
  annote="An analysis of the motion of the common carotid artery (CCA) provides effective indicators for cardiovascular diseases. Here, we propose a method for tracking CCA wall motion from a B-mode ultrasound video sequence. An unscented Kalman filter based on a suitably devised state-space model fuses measurements produced by an optical flow algorithm and a CCA wall localization algorithm. This approach compensates for feature drift, which is a detrimental effect in optical flow algorithms. The proposed method is demonstrated to outperform a state-of-the-art tracking method based on optical flow.
",
  booktitle="Proceedings of 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)",
  chapter="164233",
  doi="10.1109/ICASSP40776.2020.9054703",
  howpublished="online",
  year="2020",
  month="may",
  pages="1095--1099",
  type="conference paper"
}