“School of Cognitive Sciences”
Back to Papers HomeBack to Papers of School of Cognitive Sciences
| Paper IPM / Cognitive Sciences / 8482 |
|
||||||
| Abstract: | |||||||
|
In this paper, we introduce a novel interactive method based on symmetry and distance constraints for the segmentation of medical images. Our new symmetry interaction and distance constraint are integrated with the Herbulot�s entropy minimization and Chen�s shape-prior segmentation methods. This incorporates knowledge-based constraints that increase the accuracy and reduce the initialization dependency. We applied our algorithm to segment ventricle and caudate in magnetic resonance images (MRI) of the brain. Comparative results show the effects of the proposed constraints. More accurate results and less dependency to initialization are obtained when using the proposed method.
Download TeX format |
|||||||
| back to top | |||||||
