Accurate colon residue detection algorithm with partial volume segmentation

Colon cancer is the second leading cause of cancer-related death in the United States. Earlier detection and removal of polyps can dramatically reduce the chance of developing malignant tumor. Due to some limitations of optical colonoscopy used in clinic, many researchers have developed virtual colonoscopy as an alternative technique, in which accurate colon segmentation is crucial. However, partial volume effect and existence of residue make it very challenging. The electronic colon cleaning technique proposed by Chen et al is a very attractive method, which is also kind of hard segmentation method. As mentioned in their paper, some artifacts were produced, which might affect the accurate colon reconstruction. In our paper, instead of labeling each voxel with a unique label or tissue type, the percentage of different tissues within each voxel, which we call a mixture, was considered in establishing a maximum a posterior probability (MAP) image-segmentation framework. A Markov random field (MRF) model was developed to reflect the spatial information for the tissue mixtures. The spatial information based on hard segmentation was used to determine which tissue types are in the specific voxel. Parameters of each tissue class were estimated by the expectation-maximization (EM) algorithm during the MAP tissue-mixture segmentation. Real CT experimental results demonstrated that the partial volume effects between four tissue types have been precisely detected. Meanwhile, the residue has been electronically removed and very smooth and clean interface along the colon wall has been obtained.

[1]  Koenraad Van Leemput,et al.  A unifying framework for partial volume segmentation of brain MR images , 2003, IEEE Transactions on Medical Imaging.

[2]  D R Haynor,et al.  Partial volume tissue classification of multichannel magnetic resonance images-a mixel model. , 1991, IEEE transactions on medical imaging.

[3]  Zhengrong Liang,et al.  Partial volume segmentation of medical images , 2003, 2003 IEEE Nuclear Science Symposium. Conference Record (IEEE Cat. No.03CH37515).

[4]  Stephen M. Smith,et al.  Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm , 2001, IEEE Transactions on Medical Imaging.

[5]  Zhengrong Liang,et al.  Parameter estimation and tissue segmentation from multispectral MR images , 1994, IEEE Trans. Medical Imaging.

[6]  Bin Li,et al.  A novel approach to extract colon lumen from CT images for virtual colonoscopy , 2000, IEEE Transactions on Medical Imaging.

[7]  A. M. Youssef,et al.  Automated polyp detection at CT colonography: feasibility assessment in a human population. , 2001, Radiology.

[8]  J. Liang.  Virtual Colonoscopy : An Alternative Approach to Examination of the Entire Colon , 2001 .

[9]  W. Eric L. Grimson,et al.  Adaptive Segmentation of MRI Data , 1995, CVRMed.

[10]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[11]  Y. Masutani,et al.  Computerized detection of colonic polyps at CT colonography on the basis of volumetric features: pilot study. , 2002, Radiology.

[12]  Zhengrong Liang,et al.  A Self-adaptive Vector Quantization Algorithm for MR Image Segmentation , 1999 .

[13]  Hiroyuki Yoshida,et al.  Three-dimensional computer-aided diagnosis scheme for detection of colonic polyps , 2001, IEEE Transactions on Medical Imaging.

[14]  Koenraad Van Leemput,et al.  Automated model-based tissue classification of MR images of the brain , 1999, IEEE Transactions on Medical Imaging.

[15]  R. Jaszczak,et al.  Parameter estimation of finite mixtures using the EM algorithm and information criteria with application to medical image processing , 1992 .