Related Experiment Video
Updated: May 28, 2026
05:33
Three-Dimensional Shape Modeling and Analysis of Brain Structures
Published on: November 14, 2019
3D segmentation of rodent brain structures using hierarchical shape priors and deformable models
Shaoting Zhang1, Junzhou Huang, Mustafa Uzunbas
1CBIM, Rutgers, The State University of New Jersey, Piscataway, NJ, USA.
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|October 19, 2011
Summary
This study introduces a novel method for segmenting multiple rodent brain structures simultaneously using deformable models and hierarchical shape priors. The approach enhances accuracy and robustness, even with limited training data.
Area of Science:
- Neuroimaging
- Computational Neuroscience
- Medical Image Analysis
Background:
- Accurate segmentation of rodent brain structures is crucial for neuroscience research.
- Existing methods may struggle with simultaneous segmentation of multiple structures or require extensive training data.
Purpose of the Study:
- To develop and validate a novel method for simultaneous segmentation of multiple rodent brain structures.
- To improve segmentation accuracy and robustness through the use of hierarchical shape priors.
Main Methods:
- A combined framework of deformable models and hierarchical shape priors was employed.
- The deformation module utilized gradient and appearance information for shape transformation.
- Principal Component Analysis (PCA) was used hierarchically to model global (inter-structure) and local (intra-structure) shape statistics.
Main Results:
- The proposed method effectively segmented multiple rodent brain structures, including the cerebellum, striatum, and hippocampus.
- Hierarchical shape priors adaptively constrained deformation, enhancing model robustness and segmentation accuracy.
- The method demonstrated effectiveness with a small training dataset size, preserving shape details.
Conclusions:
- The developed method offers an effective solution for simultaneous rodent brain structure segmentation.
- Hierarchical shape priors significantly improve segmentation performance and robustness.
- This approach is advantageous for studies with limited annotated neuroimaging data.
