Performance evaluation on segmentation methods for medical images

Authors:
Manish Khare;Rajneesh Kumar Srivastava;Ashish Khare
Affiliations:
University of Allahabad, India;University of Allahabad, India;University of Allahabad, India
Venue:
ACAI '11 Proceedings of the International Conference on Advances in Computing and Artificial Intelligence
Year:
2011

Citing 9
Cited 0

Geodesic Active Contours

International Journal of Computer Vision
Level set methods: an overview and some recent results

Journal of Computational Physics
Level Set Evolution without Re-Initialization: A New Variational Formulation

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01
Image Processing, Analysis, and Machine Vision

Image Processing, Analysis, and Machine Vision
Fast Global Minimization of the Active Contour/Snake Model

Journal of Mathematical Imaging and Vision
The Split Bregman Method for L1-Regularized Problems

SIAM Journal on Imaging Sciences
Quantitative performance evaluation on segmentation methods for SAR ship images

Proceedings of the Third Annual ACM Bangalore Conference
Geometric Applications of the Split Bregman Method: Segmentation and Surface Reconstruction

Journal of Scientific Computing
Active contours without edges

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents the performance evaluation of different segmentation algorithms for medical images. Accuracy and clarity are very important issues for medical imaging and same in the case with segmentation. In this paper, we have proposed a level set method without reinitialization with some specific shapes for segmentation and compared our proposed approach for segmentation with three other approaches where the first approach is based on region splitting based region growing category, second is based on region merging based region growing category and third is based on level set. This comparison is based on six different performance parameters namely Mean Square Error (MSE), Peak Signal to Noise Ratio (PSNR), Maximum Difference (MD), Normalized Cross Correlation (NCC), Normalized Absolute Error (NAE) and Structural Content (SC). We have compared the proposed approach with three above mentioned approaches for several images, out of which results for four images are provided in this paper, and we find that the approach is better than the other one.