Related Experiment Videos
Characterization and propagation of uncertainty in diffusion-weighted MR imaging
T E J Behrens1, M W Woolrich, M Jenkinson
1Oxford Centre for Functional Magnetic Resonance Imaging of the Brain (FMRIB), Oxford, UK. behrens@fmrib.ox.ac.uk
Magnetic Resonance in Medicine
|October 31, 2003
Summary
This study introduces a probabilistic framework for diffusion model analysis, enhancing tractography by quantifying connection probabilities. The method accurately estimates brain connectivity, aligning with invasive tracer studies.
Area of Science:
- Neuroimaging
- Computational Neuroscience
- Biophysics
Background:
- Diffusion MRI models are crucial for understanding brain structure.
- Estimating local fiber direction and global connectivity remains challenging.
- Probabilistic methods offer a robust approach to address these challenges.
Purpose of the Study:
- To develop a fully probabilistic framework for estimating local probability density functions in diffusion models.
- To apply this framework to diffusion tensor and partial volume models, focusing on fiber direction.
- To enable quantification of belief in tractography results by estimating global connectivity.
Main Methods:
- Developed a probabilistic framework for local parameter estimation.
- Applied the framework to diffusion tensor and partial volume models.
- Introduced a technique to estimate global connectivity from local density functions.
Main Results:
- Successfully estimated local probability density functions for diffusion model parameters.
- Enabled estimation of global connectivity, quantifying belief in tractography.
- Applied to human thalamus cortical connectivity, showing good correspondence with primate tracer data.
Conclusions:
- The probabilistic framework provides a robust method for diffusion MRI analysis.
- This approach enhances tractography by quantifying connectivity and belief.
- Results support the validity of the method for in vivo human brain connectivity studies.