Related Experiment Video
Updated: Jul 26, 2025

04:33
Three-Dimensional Ultrasonic Needle Tip Tracking with a Fiber-Optic Ultrasound Receiver
Published on: August 21, 2018
10.4K
Learning-based needle tip tracking in 2D ultrasound by fusing visual tracking and motion prediction
Wanquan Yan1, Qingpeng Ding1, Jianghua Chen1
1Department of Mechanical and Automation Engineering and T Stone Robotics Institute, The Chinese University of Hong Kong, Hong Kong.
Medical Image Analysis
|June 12, 2023
Summary
This study introduces a novel learning-based needle tip tracking system for ultrasound procedures. The system enhances accuracy and robustness in biological tissues, improving safety in clinical practice.
Area of Science:
- Medical Imaging
- Robotics
- Computer Vision
Background:
- Ultrasound (US)-guided procedures commonly use visual trackers for needle tip tracking.
- Existing visual trackers struggle in biological tissues due to noise and occlusion.
Purpose of the Study:
- To develop a robust and accurate learning-based needle tip tracking system for US-guided procedures.
- To overcome limitations of traditional visual trackers in challenging biological environments.
Main Methods:
- A hybrid system combining visual tracking and motion prediction modules.
- Visual tracking module with enhanced discriminability masks and template updates.
- Motion prediction module utilizing a Transformer network for historical position data analysis.
- Data fusion module to integrate outputs from both tracking and prediction modules.
Main Results:
- Demonstrated significant improvement over state-of-the-art trackers in motorized needle insertion experiments (e.g., 78% vs. <60% success rate).
- Achieved >18% higher tracking success rate than the second-best system in manual needle insertion experiments.
- Showcased robustness in varying conditions, including temporary needle tip disappearance.
Conclusions:
- The proposed system offers improved computational efficiency, robustness, and accuracy for needle tip tracking.
- Enhances safety in current US-guided needle operations.
- Potential for integration into robotic systems for tissue biopsy.

