Related Experiment Video
Updated: Jun 28, 2026
09:30
Localizing Function-specific Targets for Transcranial Magnetic Stimulation in the Absence of Navigation Equipment
Published on: May 23, 2025
An active contour-based atlas registration model applied to automatic subthalamic nucleus targeting on MRI: method
Valérie Duay1, Xavier Bresson, Javier Sanchez Castro
1Signal Processing Laboratory (LTS5), EPFL, CH-1015 Lausanne.
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|November 6, 2008
Summary
This study introduces a novel atlas registration framework for precise subthalamic nucleus (STN) targeting in deep brain stimulation (DBS) surgery. The new method achieves expert-level accuracy with significantly reduced computational time.
Area of Science:
- Neurosurgery
- Medical Imaging
- Computational Anatomy
Background:
- Accurate subthalamic nucleus (STN) targeting is crucial for effective deep brain stimulation (DBS) surgery.
- Previous methods relied on visible brain structures like ventricles for STN localization.
- Existing atlas registration techniques face challenges in balancing accuracy and computational efficiency.
Purpose of the Study:
- To develop a non-parametric atlas registration framework for automatic STN targeting in DBS.
- To improve the accuracy and reduce the computational time of current state-of-the-art targeting methods.
- To integrate segmentation and registration for enhanced STN localization.
Main Methods:
- A novel non-parametric atlas registration framework combining optical flow and active contour models.
- Simultaneous segmentation and registration of brain structures, specifically the lateral and third ventricles, for STN prediction.
- Validation against expert variability and leading registration algorithms.
Main Results:
- The proposed model achieved mean STN localization errors comparable to the best performing registration algorithms.
- The method's accuracy was statistically similar to the variability observed among human targeting experts.
- A significant reduction in computational time was observed compared to existing methods.
Conclusions:
- The new framework offers a clinically viable solution for automatic STN targeting in DBS surgery.
- The simultaneous segmentation and registration approach enhances both accuracy and efficiency.
- This advancement holds potential for improving surgical planning and outcomes in DBS procedures.
