Related Experiment Video
Updated: Aug 8, 2026
10:25
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019
Simultaneous registration and segmentation of anatomical structures from brain MRI
1Department of Computer & Information Sciences & Engr., University of Florida, Gainesville, FL 32611, USA. fewang@cise.ufl.edu
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|May 12, 2006
Summary
This study introduces a new variational method for image segmentation that simultaneously performs non-rigid registration and segmentation. This approach effectively handles images with differing intensity distributions, improving accuracy in medical imaging analysis.
Area of Science:
- Medical image analysis
- Computational imaging
- Applied mathematics
Background:
- Image segmentation and registration are crucial for medical image analysis.
- Existing methods often struggle with images of distinct intensity distributions.
- Simultaneous non-rigid registration and segmentation remain a challenge.
Purpose of the Study:
- To present a novel variational formulation for registration-assisted image segmentation.
- To develop a unified approach for simultaneous non-rigid registration and segmentation.
- To address limitations of existing methods in handling diverse image intensity distributions.
Main Methods:
- A novel variational formulation leading to coupled nonlinear partial differential equations (PDEs).
- Efficient numerical schemes for solving the nonlinear PDEs.
- A unified framework for simultaneous registration and segmentation.
Main Results:
- The proposed algorithm successfully performs simultaneous non-rigid registration and segmentation.
- The method accommodates image pairs with significantly different intensity distributions.
- Demonstrated performance on synthetic and real data with quantitative accuracy estimates for registration.
Conclusions:
- The unified variational approach offers a robust solution for registration-assisted image segmentation.
- The method's ability to handle intensity variations enhances its applicability in medical imaging.
- This work provides an efficient and accurate tool for complex image analysis tasks.
