A New Approach to Lung Image Segmentation using Fuzzy Possibilistic C-Means Algorithm

TL;DR

This paper introduces a modified Fuzzy C-Means and Fuzzy Possibilistic C-Means algorithms to improve lung image segmentation accuracy, especially in noisy medical images, outperforming standard methods.

Contribution

The paper proposes a generalized, noise-robust segmentation approach by modifying the FCM algorithm and integrating FPCM for better medical image analysis.

Findings

Modified FCM reduces noise effects in segmentation

FPCM improves accuracy over standard FCM

Proposed methods outperform traditional algorithms in experiments

Abstract

Image segmentation is a vital part of image processing. Segmentation has its application widespread in the field of medical images in order to diagnose curious diseases. The same medical images can be segmented manually. But the accuracy of image segmentation using the segmentation algorithms is more when compared with the manual segmentation. In the field of medical diagnosis an extensive diversity of imaging techniques is presently available, such as radiography, computed tomography (CT) and magnetic resonance imaging (MRI). Medical image segmentation is an essential step for most consequent image analysis tasks. Although the original FCM algorithm yields good results for segmenting noise free images, it fails to segment images corrupted by noise, outliers and other imaging artifact. This paper presents an image segmentation approach using Modified Fuzzy C-Means (FCM) algorithm and Fuzzy Possibilistic c-means algorithm (FPCM). This approach is a generalized version of standard Fuzzy CMeans Clustering (FCM) algorithm. The limitation of the conventional FCM technique is eliminated in modifying the standard technique. The Modified FCM algorithm is formulated by modifying the distance measurement of the standard FCM algorithm to permit the labeling of a pixel to be influenced by other pixels and to restrain the noise effect during segmentation. Instead of having one term in the objective function, a second term is included, forcing the membership to be as high as possible without a maximum limit constraint of one. Experiments are conducted on real images to investigate the performance of the proposed modified FCM technique in segmenting the medical images. Standard FCM, Modified FCM, Fuzzy Possibilistic CMeans algorithm (FPCM) are compared to explore the accuracy of our proposed approach.