Cancer tissues recognition system using box counting method and artificial neural network

Author

George, Loay E. ; Mohammed, Esraa Z.

Author_Institution

Dept. of Comput. Sci., Univ. of Baghdad, Baghdad, Iraq

fYear

2011

fDate

14-16 Oct. 2011

Firstpage

5

Lastpage

9

Abstract

The research presented in this paper was aimed to develop a recognition system for microscopic images of breast tissues samples. The system should classify breast tissues as malignant or not, or identifying their malignancy types. In this paper, multi-scale fractal dimension concept was used to extract a set of textural features in order to perform texture analysis for breast tissues samples. The box counting method was used to estimate the multi fractal dimensions. A feed forward neural network was used to classify different types of breast tissues according to the extracted fractal dimension vectors. For ANN training purpose the back-propagation training algorithm was used. Evaluation tests were carried on 368 breast tissues images. The test results indicated that the best attained success rate was around 97%.

Keywords

backpropagation; cancer; feature extraction; feedforward neural nets; image classification; image texture; medical image processing; artificial neural network; backpropagation training algorithm; box counting method; breast tissue classification; breast tissues; cancer tissue recognition system; feedforward neural network; fractal dimension concept; fractal dimension vector; microscopic image; textural feature extraction; texture analysis; Artificial neural networks; Breast tissue; Cancer; Feature extraction; Fractals; Training; Vectors; Box counting; Breast cancer; fractal dimension; image classification; medical diagnosis; neural network;

fLanguage

English

Publisher

ieee

Conference_Titel

Soft Computing and Pattern Recognition (SoCPaR), 2011 International Conference of

Conference_Location

Dalian

Print_ISBN

978-1-4577-1195-4

Type

conf

DOI

10.1109/SoCPaR.2011.6089105

Filename

6089105