Classification of the medical images by the Kohonen network SOM and LVQ

被引：12

作者：

Chalabi, Z. ^{[1
]}

Berrached, N. ^{[1
]}

Kharchouche, N. ^{[1
]}

Ghellemallah, Y. ^{[1
]}

Mansour, M. ^{[2
]}

Mouhadjer, H. ^{[2
]}

机构：

[1] Research Laboratory in Intelligent Systems, Department of Electronics, University of Sciences and Technology, Mohamed Boudiaf, USTOran 31000

[2] Laboratory of LSI-REC, University of Sciences and Technology, Mohamed Boudiaf, USTOran 31000

来源：

Journal of Applied Sciences | 2008年 / 8卷 / 07期

关键词：

Classification; Corpus; Detection; MR image; SOM and LVQ; Tumour;

D O I：

10.3923/jas.2008.1149.1158

中图分类号：

学科分类号：

摘要：

This study fits within the framework of the diagnosis assistance and deals with the MR brain image types. To highlight the possibility of cerebral pathology such as tumours, one of the newest techniques of pattern recognition which exploits SOM (Self Organization Map) and LVQ (Learning Vector Quantization) algorithms of Kohonen is proposed. A short outline on these algorithms is brought back. Pre-processing adopted method is presented describing the training basis construction. Three classification approaches are carried out, comparative studies are conducted. The algorithm's proprieties are verified according to thėiteration number and the maps size. The classification quality is expressed via two parameters: the quantization error (QE%) and the good classification rate (CR%). Five pathological images and a healthy one arrested. The obtained results are in accordance with those of the results presented in the referred bibliographic. © 2008 Asian Network for Scientific Information.

引用

页码：1149 / 1158

页数：9

共 14 条

[1]

Aleksandra M., Myodrag V.P., at al, Wavelet image extension for analysis and classification of infarcted myocardial tissue, IEEE Trans. Biomed. Eng, 44, 9, pp. 856-866, (1997)

[2]

Andre Q., Emanuel R., Comparison of methods of classification: Application to the identification of underwater ferromagnetic objects. Electronic and Data-Processing Department, Military Technical Academy, Buccarest, Romania, Processing of the Signal, 15, 3, pp. 185-195, (1998)

[3]

Axel W., Anke M.B., Olivier L., Et al., Model-free functional MRI analysis based on unsupervised clustering, J. Biomed. Inform, 37, 1, pp. 10-18, (2004)

[4]

Cedric A., Torsten B., Vlad P., Michel V., Jean-Philippe T., Supervised nonparametric theoretic classification, International ICPR 2004 Proceedings 17th Conference one Pattern Recognition

[5]

Cambridge, 23-26, pp. 414-417, (2004)

[6]

Jan L., Arend H., Johan A.K., A combined MRI and MRSI based multiclass system for brain tumour recognition using LS-SVMs with class probabilities and feature selection Preprint Submitted to Artificial, Intel. Med, 40, 2, pp. 87-102, (2007)

[7]

Kohonen T., The Map Coil-Organizing, (1997)

[8]

Lory P., Gietl D., Neural Networks for Two-groups Classification Problems with Monoticity Hints Wütchaftsinformatik, (2000)

[9]

Lu D., Weng Q., A survey of image classification methods and techniques for improving classification performance, Int J. Remote Sensing, 28, 5, pp. 823-870, (2007)

[10]

Lukas L., Devos A., Suykens J.A.K., Vanhamme L., Et al., Brain Tumour Classification based on long echo time proton MRS signals, Artificial Intel. Med, 31, 1, pp. 73-89, (2004)

← 1 2 →