Related Experiment Videos
Articulated rigid registration for serial lower-limb mouse imaging.
Xenophon Papademetris1, Donald P Dione, Lawrence W Dobrucki
1Departments of Biomedical Engineering, Yale University, New Haven, CT 06520-8042, USA. xenophon.papademetris@yale.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 continuous piecewise rotational transformation model to accurately capture joint articulation, avoiding folding and stretching issues seen in rigid models. The new method successfully models hip, knee, and ankle movement in CT images.
Area of Science:
- Medical imaging
- Computational anatomy
- Biomechanical modeling
Background:
- Piecewise rigid models for joint articulation exhibit discontinuities, causing folding and stretching artifacts.
- Accurate modeling of joint articulation is crucial for medical image analysis and registration.
Purpose of the Study:
- To develop a novel piecewise rotational transformation model for continuous joint articulation.
- To address the limitations of existing rigid models in capturing complex joint movements.
Main Methods:
- Developed a provably continuous piecewise rotational transformation.
- Integrated the transformation model within the robust point matching framework.
- Applied the model to synthetic data and serial X-ray CT mouse images.
Main Results:
- The model successfully captures the articulation of multiple joints (hip, knee, ankle) in mouse CT images.
- Demonstrated avoidance of folding and stretching artifacts inherent in rigid models.
- Validated the model's performance on both synthetic and real-world medical imaging data.
Conclusions:
- The proposed continuous transformation model offers a robust solution for joint articulation in medical imaging.
- Potential applications include initializing non-rigid registrations and improving image-guided interventions.
- Future work may extend the model to human data for various clinical applications.