Related Experiment Video
Updated: May 25, 2026
02:09
Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
Published on: April 12, 2024
Model-based 3D/2D deformable registration of MR images.
Bahram Marami1, Shahin Sirouspour, David W Capson
1Department of Electrical and Computer Engineering, McMaster University, Hamilton, Ontario L8S 4K1, Canada. maramib@mcmaster.ca
Summary
This study introduces an automatic method for registering 3D MRI scans to 2D intraoperative images, accurately tracking large tissue deformations without markers. This advance aids in procedures like MR-guided breast biopsy.
Area of Science:
- Medical Imaging
- Computational Mechanics
- Image Registration
Background:
- Accurate registration of preoperative 3D magnetic resonance imaging (MRI) to intraoperative 2D images is crucial for guiding interventions in deformable tissues.
- Existing methods often rely on fiducial markers, feature extraction, or segmentation, which can be cumbersome or inaccurate for highly deformable tissues.
Purpose of the Study:
- To develop and evaluate an automatic registration method for 3D MRI to 2D intraoperative images of deformable tissues.
- To enable accurate image guidance in procedures involving significant tissue deformation, such as MR-based breast biopsy.
Main Methods:
- Proposed an algorithm employing a dynamic continuum mechanics model for tissue deformation.
- Utilized similarity measures including correlation ratio, mutual information, and sum of squared differences for registration.
- Registration was performed using only information from preoperative 3D MRI and intraoperative 2D images, without requiring fiducial markers, feature extraction, or segmentation.
Main Results:
- Experimental results with a biopsy training breast phantom demonstrated the method's ability to perform well despite large deformations.
- The algorithm successfully registered 3D preoperative MRI data to 2D intraoperative images automatically.
Conclusions:
- The proposed automatic registration method is effective for deformable tissues, even with substantial deformations.
- This technique holds significant promise for improving the precision and safety of clinical applications like MR-based breast biopsy.
