Proceedings of the 18th annual conference on Computer graphics and interactive techniques
Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Zoom-invariant vision of figural shape: the mathematics of cores
Computer Vision and Image Understanding
Region-Based 2D Deformable Generalized Cylinder for Narrow Structures Segmentation
ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part II
Hi-index | 0.00 |
This paper presents a novel robust automatic method for the segmentation of the Inferior Vena Cava (IVC) in the proximity of the liver. In clinical diagnosis and surgery planning, IVC segmentation is essential since it strongly impacts both liver volumetry accuracy and vascularity analysis. Given the anatomical variability, the lack of clear boundaries and complexity of the surrounding structures along the IVC, its segmentation remains a difficult and open problem. To cope with such challenging conditions, we developed an implicit representation of a generalized cylinder and optimized a local region-based criterion under dedicated anatomical constraints. Our method was tested on a dataset of 20 contrast-enhanced CT scans, achieving 80% success rate in fully automatic mode. The remaining cases needed minimal user input (one point) to reach 95% success under radiology expert criteria.