This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Segmentation is one of the essential tasks in medical image analysis Many sophisticated automatic segmentation algorithms exist … … which might fail in some cases (low contrast, noise, biological
variability)
What to do? Manual segmentation? Takes too long Different algorithm? Might fail as well Locally correct the error!
Why do we need efficient segmentation editing tools?Solution Results Outlook Conclusion
Requirements: Intuitive interaction in 2D – estimate the user’s intention in 3D Local modifications Real-time feedback Provide a general tool (for different objects and modalities) Be independent of the preceding automatic algorithm
The user expects the tool to allow him or her to correct all errors With only a few steps!
The segmentation problems are typically hard (noise, low contrast, …) Do not use the image!
What makes segmentation editing a difficult problem?Solution Results Outlook Conclusion
Use methods known from object reconstruction Contour-based representation Can be treated as a point cloud Reconstruct a smooth surface using variational interpolation
Segmentation Formulated as an Object Reconstruction Problem
Segmentation editing is an indispensable step in the segmentation process
Efficient editing in 3D is challenging
Sketching provides an intuitive interface for segmentation editing in 2D
We have proposed a general, efficient editing tool 2D corrections are extrapolated to 3D using object reconstruction Can be used for any 3D modality and any compact object