Related Experiment Video
Updated: May 16, 2026
07:33
Non-invasive Skeletal Muscle Quantification in Small Animals Using Micro-computed Tomography
Published on: November 8, 2024
Segmentation and visual analysis of whole-body mouse skeleton microSPECT
Artem Khmelinskii1, Harald C Groen, Martin Baiker
1Division of Image Processing, Department of Radiology, Leiden University Medical Center, Leiden, The Netherlands.
Plos One
|November 16, 2012
Summary
This study presents a novel method for aligning whole-body SPECT/CT mouse datasets, overcoming data variability for improved drug development research. The technique significantly reduces comparison errors, enabling more accurate analysis of small animal imaging data.
Area of Science:
- Medical Imaging
- Small Animal Models
- Nuclear Medicine
Background:
- Whole-body SPECT/CT imaging is crucial for preclinical cancer research and drug development.
- Data heterogeneity in whole-body datasets (volume, modality, positioning) complicates comparative analysis.
- Existing methods struggle with variability, hindering efficient cross-sectional and multi-modal study comparisons.
Purpose of the Study:
- To develop and evaluate a method for aligning and comparing multiple whole-body mouse SPECT datasets in a common reference frame.
- To eliminate acquisition variability in cross-sectional and multi-modal small animal imaging studies.
- To facilitate side-by-side visualization and comparison of SPECT/CT data.
Main Methods:
- Utilized six whole-body SPECT/CT datasets from BALB/c mice using bone-targeting tracers.
- Employed an articulated MOBY whole-body mouse atlas as a common reference.
- Registered individual bones sequentially to extracted SPECT skeletons using an anatomical hierarchy and constrained degrees of freedom.
- Applied the Articulated Planar Reformation (APR) algorithm for data visualization and comparison.
Main Results:
- Quantitative evaluation showed a significant decrease in mean Euclidean distance from 11.5±12.1 to 2.6±2.1 voxels after registration.
- The algorithm achieved satisfactory segmentation with minimal user intervention.
- Demonstrated robustness with incomplete skeletal data and facilitated intuitive exploration of multi-modal SPECT/CT data.
Conclusions:
- The proposed registration and visualization method effectively addresses data heterogeneity in whole-body mouse SPECT/CT imaging.
- This approach enhances the accuracy and efficiency of comparative analysis in preclinical research.
- It supports robust, intuitive exploration of cross-sectional and multi-modal small animal imaging data.
