Related Experiment Video
Updated: May 12, 2026
09:11
Use of MRI-ultrasound Fusion to Achieve Targeted Prostate Biopsy
Published on: April 9, 2019
Needle segmentation using 3D Hough transform in 3D TRUS guided prostate transperineal therapy.
Wu Qiu1, Ming Yuchi, Mingyue Ding
1Department of Biomedical Engineering, School of Life Science and Technology, Huazhong University of Science and Technology, Wuhan, Hubei 430074, China. wqiu@imaging.robarts.ca
Medical Physics
|April 6, 2013
Summary
A new needle segmentation method accurately localizes implants in 3D transrectal ultrasound (TRUS) images for prostate cancer therapy. This robust approach improves needle positioning accuracy, aiding effective cryotherapy and brachytherapy procedures.
Area of Science:
- Medical Imaging
- Computational Anatomy
- Surgical Navigation
Background:
- Prostate adenocarcinoma is a prevalent cancer requiring precise interventional therapies like cryotherapy and brachytherapy.
- Accurate needle placement is critical for the success of these prostate cancer treatments.
- Existing methods may lack the precision needed for optimal needle guidance in three-dimensional (3D) procedures.
Purpose of the Study:
- To develop and validate a robust and efficient needle segmentation algorithm.
- To enhance the localization accuracy of needles in 3D transrectal ultrasound (TRUS) images.
- To provide an aid for precise needle implantation in image-guided prostate therapies.
Main Methods:
- A three-step process involving image cropping, histogram-based binarization, and 3D Hough transform for needle axis localization.
- Optimal threshold analysis of intensity distribution to determine needle endpoint position.
- Coarse-fine searching strategy implemented for improved computational efficiency.
Main Results:
- The algorithm achieved high endpoint localization accuracy: 0.5 mm (agar phantom), 0.7 mm (chicken breast phantom), and 1 mm (in vivo patient images).
- Demonstrated robustness across varying parameters like insertion angle, length, and threshold levels.
- Mean execution time for segmentation was approximately 2 seconds for a 3D TRUS image.
Conclusions:
- The proposed needle segmentation algorithm is accurate and robust for 3D TRUS-guided prostate therapies.
- The method is suitable for real-time or near-real-time applications in prostate brachytherapy and cryotherapy.
- This technique can improve the precision and success rates of interventional prostate cancer treatments.
