NA-MIC Project Weeks
Back to Projects List
Multimodal registration
Key Investigators
- Leroux Gaelle (University of Michigan, USA)
- Claret Jeanne (University of Michigan, USA)
- Cevidanes Lucia (University of Michigan, USA)
- Allemand David (Kitware, USA)
- Prieto Juan Carlos (University of North Carolina, USA)
Project Description
The “Multimodal Registration Project” aims to develop a system capable of aligning different medical imaging modalities to enhance diagnostic accuracy and patient outcomes. By focusing on the registration of Cone Beam Computed Tomography (CBCT) with Magnetic Resonance Imaging (MRI), we seek to create a unified imaging model that offers comprehensive insights into patient anatomy and pathology.
Objective
- The objective is to register MRI images with CBCT data accurately.
Approach and Plan
- Collected a robust dataset comprising MRI and CBCT files.
- Compare the performance of three registration approaches : Generative Adversarial Network (GNA), agent-based action learning for non-rigid registration and Elastix’s free-form deformation method.
- Validate the model’s accuracy through rigorous testing against established benchmarks.
Progress and Next Steps
- Data collection phase.
- Reviewing available papers and code.
Illustrations
No response
Background and References
No response