Fuzzy entropy based optimization of clusters for the segmentation of lungs in CT scanned images |
| |
Authors: | M Arfan Jaffar Ayyaz Hussain Anwar Majid Mirza |
| |
Affiliation: | 1. Department of Computer Science, FAST National University of Computer and Emerging Sciences, A.K. Brohi Road, H-11/4, Islamabad, Pakistan
|
| |
Abstract: | In this paper, we have proposed a method for segmentation of lungs from Computed Tomography (CT)-scanned images using spatial
Fuzzy C-Mean and morphological techniques known as Fuzzy Entropy and Morphology based Segmentation. To determine dynamic and
adaptive optimal threshold, we have incorporated Fuzzy Entropy. We have proposed a novel histogram-based background removal
operator. The proposed system is capable to perform fully automatic segmentation of CT Scan Lung images, based solely on information
contained by the image itself. We have used different cluster validity functions to find out optimal number of clusters. The
proposed system can be used as a basic building block for Computer-Aided Diagnosis. The technique was tested against the 25
datasets of different patients received from Aga Khan Medical University, Pakistan. The results confirm the validity of technique
as well as enhanced performance. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|