Related Experiment Video
Updated: Feb 17, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
43.7K
THE IMPACT OF MATCHING FUNCTIONAL ON ATROPHY MEASUREMENT FROM GEODESIC SHOOTING IN DIFFEOMORPHISMS
Greg M Fleishman1, Paul M Thompson2
1UCLA Bioengineering, 420 Westwood Plaza, 5121 Engineering V, UCLA, CA 90095-1600.
Proceedings. IEEE International Symposium on Biomedical Imaging
|December 5, 2017
Summary
This study explores image matching functionals for brain atrophy quantification using geodesic shooting in diffeomorphisms. Mutual information yielded comparable clinical correlations and improved sample size estimates for clinical trials.
Area of Science:
- Neuroimaging
- Medical Image Analysis
- Computational Anatomy
Background:
- Longitudinal registration is crucial for mapping brain atrophy and tissue loss over time in healthy and demented individuals.
- The geodesic shooting in diffeomorphisms framework offers a powerful tool for analyzing these changes.
- However, the impact of different image matching functionals within this framework on atrophy quantification remains under-explored.
Purpose of the Study:
- To investigate the influence of four distinct image matching functionals on brain atrophy quantification using the geodesic shooting in diffeomorphisms framework.
- To assess how the choice of matching functional affects the correlation between atrophy scores and clinical variables.
- To evaluate the impact of matching functional selection on sample size estimations for clinical trials focused on slowing brain atrophy.
Main Methods:
- Utilized the geodesic shooting in diffeomorphisms framework for longitudinal image registration.
- Compared four different image matching functionals (e.g., mutual information, sum of squared differences) to drive the registration process.
- Correlated resulting atrophy scores with clinical variables and estimated sample sizes (N80) for hypothetical clinical trials.
Main Results:
- The choice of matching functional significantly impacts the registration outcomes for atrophy quantification.
- Mutual information, typically used for linear and multi-modal registration, demonstrated comparable correlations with clinical variables.
- Mutual information also provided more favorable sample size estimates for detecting treatment effects in clinical trials.
Conclusions:
- The selection of an appropriate image matching functional is critical for accurate brain atrophy quantification within the geodesic shooting in diffeomorphisms framework.
- Mutual information emerges as a robust choice, balancing clinical relevance with statistical power for future clinical trial design.
- This finding has implications for optimizing neuroimaging analysis in dementia research and drug development.
