期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2014
卷号：60
期号：3
出版社：Journal of Theoretical and Applied
摘要：Liver segmentation from CT volumes has been a challenging problem due to the high inter-organs intensity similarity, the intra-liver intensity variability, and the partial volume effect. In this paper, we perform an extensive review of the liver segmentation literature from CT and MRI. Furthermore, we propose a Bayesian model for a robust and reproducible semi-automatic technique for liver segmentation from CT volumes. We train our model and validate it using 44 clinical volumes for patients with various types of liver abnormality including tumor. Our segmentation results show a robust and clinically acceptable liver volume for all the 44 clinical cases we have with average area overlap accuracy over 87%. Our method is superior to all state of the art methods that has only been validated on less number of subjects as we show during the literature survey.
