Self-supervised context-aware correlation filter for robust landmark tracking in liver ultrasound sequences

Lin Ma; Junjie Wang; Shu Gong; Libin Lan; Li Geng; Siping Wang; Xin Feng

doi:10.1515/bmt-2022-0489

Self-supervised context-aware correlation filter for robust landmark tracking in liver ultrasound sequences

Biomed Tech (Berl). 2024 Feb 7. doi: 10.1515/bmt-2022-0489. Online ahead of print.

Authors

Lin Ma¹, Junjie Wang¹, Shu Gong², Libin Lan¹, Li Geng³, Siping Wang¹, Xin Feng¹

Affiliations

¹ College of Computer Science and Engineering, Chongqing University of Technology, Chongqing, China.
² Department of Gastroenterology, Children's Hospital of Chongqing Medical University, Chongqing, China.
³ City University of New York NYCCT, New York, USA.

PMID: 38353097
DOI: 10.1515/bmt-2022-0489

Abstract

Objectives: Respiratory motion-induced displacement of internal organs poses a significant challenge in image-guided radiation therapy, particularly affecting liver landmark tracking accuracy.

Methods: Addressing this concern, we propose a self-supervised method for robust landmark tracking in long liver ultrasound sequences. Our approach leverages a Siamese-based context-aware correlation filter network, trained by using the consistency loss between forward tracking and back verification. By effectively utilizing both labeled and unlabeled liver ultrasound images, our model, Siam-CCF , mitigates the impact of speckle noise and artifacts on ultrasonic image tracking by a context-aware correlation filter. Additionally, a fusion strategy for template patch feature helps the tracker to obtain rich appearance information around the point-landmark.

Results: Siam-CCF achieves a mean tracking error of 0.79 ± 0.83 mm at a frame rate of 118.6 fps, exhibiting a superior speed-accuracy trade-off on the public MICCAI 2015 Challenge on Liver Ultrasound Tracking (CLUST2015) 2D dataset. This performance won the 5th place on the CLUST2015 2D point-landmark tracking task.

Conclusions: Extensive experiments validate the effectiveness of our proposed approach, establishing it as one of the top-performing techniques on the CLUST2015 online leaderboard at the time of this submission.

Keywords: image-guided radiation therapy; liver ultrasound landmark tracking; respiratory motion estimation; self-supervised context-aware correlation filter.