Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

EchoTracker: Advancing Myocardial Point Tracking in Echocardiography

Md Abulkalam Azad, Artem Chernyshov, John Nyberg, Ingrid Tveten, Lasse Lovstakken, Håvard Dalen, Bjørnar Grenne, Andreas Østvik

TL;DR

EchoTracker addresses the challenges of long-range myocardial tissue tracking in echocardiography by introducing a two-stage coarse-to-fine, TAP-inspired framework that jointly leverages coarse trajectory initialization and iterative refinement using multi-scale cost volumes. The method uses a lightweight pruned 2D encoder for coarse features, followed by a fine-stage with local multicrop-pyramid feature extraction and a 1D ResNet to update trajectories, enabling robust tracking across the cardiac cycle. Empirical results on four GE Vivid E95 datasets show EchoTracker outperforms state-of-the-art point-tracking methods in technical metrics and yields a 25% relative improvement in global longitudinal strain (GLS) measurements in a clinical test–retest setting. This approach offers a pathway to improved diagnostic and prognostic value in echocardiography and is made available as open-source code.

Abstract

Tissue tracking in echocardiography is challenging due to the complex cardiac motion and the inherent nature of ultrasound acquisitions. Although optical flow methods are considered state-of-the-art (SOTA), they struggle with long-range tracking, noise occlusions, and drift throughout the cardiac cycle. Recently, novel learning-based point tracking techniques have been introduced to tackle some of these issues. In this paper, we build upon these techniques and introduce EchoTracker, a two-fold coarse-to-fine model that facilitates the tracking of queried points on a tissue surface across ultrasound image sequences. The architecture contains a preliminary coarse initialization of the trajectories, followed by reinforcement iterations based on fine-grained appearance changes. It is efficient, light, and can run on mid-range GPUs. Experiments demonstrate that the model outperforms SOTA methods, with an average position accuracy of 67% and a median trajectory error of 2.86 pixels. Furthermore, we show a relative improvement of 25% when using our model to calculate the global longitudinal strain (GLS) in a clinical test-retest dataset compared to other methods. This implies that learning-based point tracking can potentially improve performance and yield a higher diagnostic and prognostic value for clinical measurements than current techniques. Our source code is available at: https://github.com/riponazad/echotracker/.

EchoTracker: Advancing Myocardial Point Tracking in Echocardiography

TL;DR

EchoTracker addresses the challenges of long-range myocardial tissue tracking in echocardiography by introducing a two-stage coarse-to-fine, TAP-inspired framework that jointly leverages coarse trajectory initialization and iterative refinement using multi-scale cost volumes. The method uses a lightweight pruned 2D encoder for coarse features, followed by a fine-stage with local multicrop-pyramid feature extraction and a 1D ResNet to update trajectories, enabling robust tracking across the cardiac cycle. Empirical results on four GE Vivid E95 datasets show EchoTracker outperforms state-of-the-art point-tracking methods in technical metrics and yields a 25% relative improvement in global longitudinal strain (GLS) measurements in a clinical test–retest setting. This approach offers a pathway to improved diagnostic and prognostic value in echocardiography and is made available as open-source code.

Abstract

Tissue tracking in echocardiography is challenging due to the complex cardiac motion and the inherent nature of ultrasound acquisitions. Although optical flow methods are considered state-of-the-art (SOTA), they struggle with long-range tracking, noise occlusions, and drift throughout the cardiac cycle. Recently, novel learning-based point tracking techniques have been introduced to tackle some of these issues. In this paper, we build upon these techniques and introduce EchoTracker, a two-fold coarse-to-fine model that facilitates the tracking of queried points on a tissue surface across ultrasound image sequences. The architecture contains a preliminary coarse initialization of the trajectories, followed by reinforcement iterations based on fine-grained appearance changes. It is efficient, light, and can run on mid-range GPUs. Experiments demonstrate that the model outperforms SOTA methods, with an average position accuracy of 67% and a median trajectory error of 2.86 pixels. Furthermore, we show a relative improvement of 25% when using our model to calculate the global longitudinal strain (GLS) in a clinical test-retest dataset compared to other methods. This implies that learning-based point tracking can potentially improve performance and yield a higher diagnostic and prognostic value for clinical measurements than current techniques. Our source code is available at: https://github.com/riponazad/echotracker/.
Paper Structure (14 sections, 2 figures, 4 tables)

This paper contains 14 sections, 2 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Tracking Any Point (TAP)
  3. Methods
  4. EchoTracker
  5. Initialization.
  6. Iterative Reinforcement.
  7. Experiments and Results
  8. Datasets
  9. Implementation details
  10. Evaluation
  11. Evaluation Metrics.
  12. Technical Performance.
  13. Clinical Performance.
  14. Conclusion

Figures (2)

  • Figure 1: An illustration of tracking queried points (highlighted in red) from the first frame throughout one heart cycle.
  • Figure 2: EchoTracker is a two-fold coarse-to-fine model. Initially, it estimates coarse trajectories (yellow points) based on the cost volume for the given query points (red). It then imposes iterative reinforcement to obtain the fine trajectories (green points).