Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Ensemble Learning for Microbubble Localization in Super-Resolution Ultrasound

Sepideh K. Gharamaleki, Brandon Helfield, Hassan Rivaz

TL;DR

This work tackles the challenge of accurately localizing microbubbles in super-resolution ultrasound by leveraging ensemble learning on a transformer-based MB detector (DEDETR). By combining five DEDETR models trained on varied patch sizes through four fusion strategies, the approach achieves higher precision and recall with reduced RMSE, notably with Weighted Box Fusion (WBF) providing the best gains on simulated data and consistent improvements in in vivo data. The method improves the quality of SR-US vascular maps while highlighting trade-offs in compute time and complexity, suggesting careful consideration for clinical implementation. Overall, the ensemble framework enhances MB localization robustness and reliability, promising more accurate SR-US imaging of microvasculature and potential diagnostic benefits.

Abstract

Super-resolution ultrasound (SR-US) is a powerful imaging technique for capturing microvasculature and blood flow at high spatial resolution. However, accurate microbubble (MB) localization remains a key challenge, as errors in localization can propagate through subsequent stages of the super-resolution process, affecting overall performance. In this paper, we explore the potential of ensemble learning techniques to enhance MB localization by increasing detection sensitivity and reducing false positives. Our study evaluates the effectiveness of ensemble methods on both in vivo and simulated outputs of a Deformable DEtection TRansformer (Deformable DETR) network. As a result of our study, we are able to demonstrate the advantages of these ensemble approaches by showing improved precision and recall in MB detection and offering insights into their application in SR-US.

Ensemble Learning for Microbubble Localization in Super-Resolution Ultrasound

TL;DR

This work tackles the challenge of accurately localizing microbubbles in super-resolution ultrasound by leveraging ensemble learning on a transformer-based MB detector (DEDETR). By combining five DEDETR models trained on varied patch sizes through four fusion strategies, the approach achieves higher precision and recall with reduced RMSE, notably with Weighted Box Fusion (WBF) providing the best gains on simulated data and consistent improvements in in vivo data. The method improves the quality of SR-US vascular maps while highlighting trade-offs in compute time and complexity, suggesting careful consideration for clinical implementation. Overall, the ensemble framework enhances MB localization robustness and reliability, promising more accurate SR-US imaging of microvasculature and potential diagnostic benefits.

Abstract

Super-resolution ultrasound (SR-US) is a powerful imaging technique for capturing microvasculature and blood flow at high spatial resolution. However, accurate microbubble (MB) localization remains a key challenge, as errors in localization can propagate through subsequent stages of the super-resolution process, affecting overall performance. In this paper, we explore the potential of ensemble learning techniques to enhance MB localization by increasing detection sensitivity and reducing false positives. Our study evaluates the effectiveness of ensemble methods on both in vivo and simulated outputs of a Deformable DEtection TRansformer (Deformable DETR) network. As a result of our study, we are able to demonstrate the advantages of these ensemble approaches by showing improved precision and recall in MB detection and offering insights into their application in SR-US.

Paper Structure

This paper contains 6 sections, 2 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Results
  4. Conclusion
  5. Compliance with ethical standards
  6. Acknowledgments

Figures (4)

  • Figure 1: Full-view SR maps of the simulation test dataset for different methods.
  • Figure 2: Zoomed-in boxes from different simulation SR maps, showing the results from each method (indicated by columns) for each of the (A), (B), and (C) boxes.
  • Figure 3: Full-view SR maps of the in vivo test dataset for different methods.
  • Figure 4: Zoomed-in boxes from different in vivo SR maps, showing the results from each method (indicated by columns) for each of the (A), (B) and (C) boxes.