Fuzzy entropy based optimization of clusters for the segmentation of lungs in CT scanned images

  • Authors:

  • M. Arfan Jaffar;Ayyaz Hussain;Anwar Majid Mirza

  • Affiliations:

  • FAST National University of Computer and Emerging Sciences, Department of Computer Science, A.K. Brohi Road, H-11/4, Islamabad, Pakistan;FAST National University of Computer and Emerging Sciences, Department of Computer Science, A.K. Brohi Road, H-11/4, Islamabad, Pakistan;FAST National University of Computer and Emerging Sciences, Department of Computer Science, A.K. Brohi Road, H-11/4, Islamabad, Pakistan

  • Venue:
  • Knowledge and Information Systems
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

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.