Clustering White Matter Fiber Bundles Using Diffusion Maps
Author Information
Author(s): Demian Wassermann, Maxime Descoteaux, Rachid Deriche
Primary Institution: INRIA/ENPC/ENS INRIA, Sophia-Antipolis, France
Hypothesis
Can diffusion maps clustering improve the segmentation of complex white matter fiber bundles in magnetic resonance imaging?
Conclusion
The diffusion maps clustering method successfully segments complex white matter fiber bundles, outperforming traditional methods.
Supporting Evidence
- The diffusion maps clustering method requires fewer assumptions from the input data.
- It reduces artifacts in segmentation compared to classical methods.
- The method successfully segments fiber bundles in real-brain datasets.
Takeaway
This study shows a new way to look at brain images to better understand how different parts of the brain connect, especially where the fibers cross each other.
Methodology
The study used diffusion maps clustering applied to diffusion MRI data to segment white matter fiber bundles, focusing on overcoming limitations of traditional methods.
Limitations
The method may still face challenges in accurately segmenting fibers in regions with complex crossings and varying sampling densities.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website