Related Experiment Video
Updated: Jun 28, 2026
10:25
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019
Symmetric nonrigid image registration: application to average brain templates construction
Vincent Noblet1, Christian Heinrich, Fabrice Heitz
1Laboratoire des Sciences de l'Image, de l'Informatique et de la Télédétection, LSIIT, UMR CNRS-ULP 7005, Bd Sébastien Brant, 67412 Illkirch, France. noblet@lsiit.u-strasbg.fr
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 symmetric image registration method. It maps images to a shared space, improving consistency for applications like creating average brain templates from MRI scans.
Area of Science:
- Medical Imaging
- Computer Vision
- Computational Anatomy
Background:
- Image registration is crucial for comparing and analyzing medical images.
- Traditional methods often create asymmetric mappings, leading to inconsistencies.
- Developing consistent registration is vital for accurate medical image analysis.
Purpose of the Study:
- To present a symmetric image registration formulation.
- To establish a common coordinate system for multiple images.
- To enable efficient construction of average brain templates from 3D MRI data.
Main Methods:
- Developed a symmetric registration framework mapping images to a midpoint coordinate system.
- Applied the framework to register multiple 3D nonrigid brain MR images.
- Utilized the registration strategy for creating average brain templates.
Main Results:
- The symmetric approach yields consistent mappings between images.
- Demonstrated efficient registration of a large set of 3D brain MR images.
- Successfully constructed average brain templates using the proposed method.
Conclusions:
- Symmetric image registration offers a more consistent approach compared to traditional methods.
- The proposed framework is efficient for large-scale image mapping.
- This method facilitates the creation of accurate average brain templates for research.
