New Method for Segmenting the Corpus Callosum in Brain Images
Author Information
Author(s): He Qing, Duan Ye, Miles Judith, Takahashi Nicole
Primary Institution: University of Missouri-Columbia
Hypothesis
Can a context-sensitive active contour improve the segmentation of the corpus callosum in MR images?
Conclusion
The proposed method is fast, accurate, and easy to operate for segmenting the corpus callosum.
Supporting Evidence
- The method demonstrated high accuracy with low false negative and false positive fractions.
- Segmentation results were validated against manual expert segmentation.
- The algorithm completed segmentation in 2-3 seconds per MR image.
Takeaway
This study created a new way to help computers find and outline a part of the brain called the corpus callosum in pictures, making it easier and faster for doctors.
Methodology
The method involves user-initialized seed contours that evolve based on prior knowledge and specific deformation rules.
Limitations
The method requires user interaction for initialization, which may vary in effectiveness based on user skill.
Participant Demographics
16 subjects with varying brain MR images.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website