Fast automatic liver segmentation combining learned shape priors with observed shape deviation
Authors:
- 1 October 2010
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10637125,p. 249-254
Abstract
We present a novel statistical shape model approach for fully automatic CT liver segmentation. Unlike previous techniques, our method combines learned local shape priors with constraints that are directly derived from the current curvature of the model in order to restrict adaptation to regions where large deformations are expected and observed. Our approach is based on a multi-tiered framework that is more robust against model initialization errors than existing methods, because the model's degrees of freedom are step-wise increased. We evaluated our method on a large data base of 86 CT liver scans from different vendors, protocols, varying resolution and contrast enhancement. For comparison, 50 of the scans were taken from 2 public data bases, one of it being the MICCAI'07 liver segmentation challenge data base. Evaluation shows state of the art results with an average mean surface distance between 1.3 mm and 1.85 mm compared to ground truth depending on the image resolution. With an average segmentation time of 45 seconds our approach outperforms other automatic methods.Keywords
This publication has 7 references indexed in Scilit:
- Computer aided segmentation of kidneys using locally shape constrained deformable models on CT images2010
- Statistical shape models for 3D medical image segmentation: A reviewMedical Image Analysis, 2009
- Comparison and Evaluation of Methods for Liver Segmentation From CT DatasetsIEEE Transactions on Medical Imaging, 2009
- Statistical Location Model for Abdominal Organ Localization2009
- Hierarchical, learning-based automatic liver segmentation2008
- Automatic Model-Based Segmentation of the Heart in CT ImagesIEEE Transactions on Medical Imaging, 2008
- Active Shape Models-Their Training and ApplicationComputer Vision and Image Understanding, 1995