Related Experiment Video
Updated: Aug 4, 2025
09:11
Use of MRI-ultrasound Fusion to Achieve Targeted Prostate Biopsy
Published on: April 9, 2019
21.6K
2D/3D Deep Registration Along Trajectories With Spatiotemporal Context: Application to Prostate Biopsy Navigation
IEEE Transactions on Bio-Medical Engineering
|April 6, 2023
Summary
This study introduces a novel spatiotemporal registration network (SpT-Net) for accurate prostate cancer biopsy targeting. The method enhances navigation by continuously tracking biopsy location relative to the prostate, improving diagnostic and therapeutic accuracy.
Area of Science:
- Medical Imaging
- Computer Vision
- Biomedical Engineering
Background:
- Accurate biopsy targeting is critical for prostate cancer diagnosis and treatment.
- Current transrectal ultrasound (TRUS) guidance faces challenges due to prostate motion and navigation limitations.
Purpose of the Study:
- To develop and evaluate a novel deep registration method for precise 2D/3D ultrasound image guidance.
- To enhance the continuous tracking of biopsy tool trajectories within the prostate.
Main Methods:
- A spatiotemporal registration network (SpT-Net) was proposed to localize live 2D ultrasound images against a reference volume.
- The network incorporated temporal context from prior trajectories and probe tracking, with various spatial context inputs evaluated.
- Ablation studies and cumulative error analysis along simulated trajectories validated the model's clinical realism.
Main Results:
- A model combining local spatial information with temporal context outperformed more complex spatiotemporal combinations.
- The best-performing SpT-Net demonstrated robust, real-time 2D/3D ultrasound registration performance on trajectories.
Conclusions:
- The developed method achieves registration performance that meets clinical requirements and outperforms existing state-of-the-art techniques.
- This approach shows significant promise for improving prostate biopsy navigation and other ultrasound-guided procedures.
