Related Experiment Video
Updated: May 26, 2026
05:33
Three-Dimensional Shape Modeling and Analysis of Brain Structures
Published on: November 14, 2019
Hierarchical statistical shape models of multiobject anatomical structures: application to brain MRI.
Juan J Cerrolaza1, Arantxa Villanueva, Rafael Cabeza
1Department of Electrical and Electronic Engineering, Public University of Navarra, Pamplona, Spain. juanjose.cerrolaza@unavarra.es
IEEE Transactions on Medical Imaging
|December 24, 2011
Summary
This study introduces a novel hierarchical segmentation method for magnetic resonance (MR) brain images. The new approach improves segmentation accuracy and robustness for subcortical structures.
Area of Science:
- Medical Imaging
- Computational Anatomy
- Neuroscience
Background:
- Accurate segmentation of subcortical brain structures in MR images is vital for medical diagnosis and research.
- Traditional methods like active shape models (ASMs) struggle with complex shape variations and inter-object relationships.
Purpose of the Study:
- To develop a more efficient and accurate segmentation framework for subcortical brain structures.
- To leverage a multiresolution approach to model inter-object relationships effectively.
Main Methods:
- A novel hierarchical segmentation framework based on the wavelet transform.
- Decomposition of inter-object relationships into multiple levels of detail.
- Application to an eight-object structure in axial MR brain images.
Main Results:
- The hierarchical segmentation significantly enhanced segmentation accuracy compared to traditional methods.
- The framework demonstrated remarkable robustness irrespective of training set size.
- Efficient characterization of object relationships and local contexts was achieved.
Conclusions:
- The proposed hierarchical segmentation method offers a significant advancement in MR image analysis.
- This technique provides a robust and accurate solution for segmenting complex subcortical brain structures.
- The wavelet-based multiresolution framework effectively models inter-object dependencies for improved segmentation.
