Automatic Segmentation of Colon in 3D CT Images and Removal of Opacified Fluid Using Cascade Feed Forward Neural Network

Table 1

Comparative study of the method proposed with its limitation.

Author	Method	Limitation
Bert et al. (2009) [5]	Automatic segmentation of colon using 3D seeded region growing algorithm	It is obvious that the most serious problem of region growing is the power and time consuming
Losnegård et al. (2010) [7]	Semiautomatic segmentation	Disadvantage is that it consumes more time and lesser accuracy
Lu and Zhao (2011) [8]	Noncolonic attachment classification algorithm and a heuristic connection algorithm	This method could achieve 92.86% coverage of human-generated colons, which is of 13.68% higher than the conventional method
Chowdhury and Whelan (2011) [9]	Automatic colon segmentation from CT data using colon geometrical features	This approach performs better and provides efficient results in colon segmentation
Kilic et al. (2009) [12]	Automatic three-dimensional computer-aided detection system	Average coverage is about 87.5% of the entire colon
Taimouri et al. (2011) [10]	Constrained least-squares filtering (CLSF)	Applicable only for specific cases, not converged early