Open Access Open Access  Restricted Access Subscription Access

Classification of Chronic Kidney Diseases with Statistical Analysis of Textural Parameters: A Data Mining Technique

Hafeez Ullah Janjua, Anser Jahangir, Ghulam Gilanie

Abstract


A computer-aided system is planned for automatic classification of ultrasound kidney diseases in this paper. Two types of images were considered, normal and chronic. By using data mining technique, we will be able to differentiate between normal and abnormal ultrasonic kidney images, by extracting statistical feature. Feature extraction using data mining technique is a computer vision application. By using feature extraction software (FES) a set statistical features were extracted from the region of interest of each image at different frame rates. The datasets which were obtained using (FES) at different frame rates were then classified by using Weka. These extracted features results were classified by using Weka and 96.5% correct classification rate is obtained. The difference between the values of these features differentiates between normal and abnormal image.

Full Text:

PDF

References


R. Chellappa. Digital Image Processing. IEEE Computer Society Press; 1992.

D. Balasubramanian, P. Srinivasan, R. Gurupatham. Automatic classification of focal lesions in ultrasound liver images using principal component analysis and neural networ. ks, Engineering in Medicine and Biology Society, 2007. EMBS 2007; In: 29th Annual International Conference of the IEEE. Lyon, France, August 2007, 2134–7p.

J. Blahuta, T. Soukup, P. Čermák. The image recognition of brain-stem ultrasound images with using a neural network based on PCA, recent researches in communications, electrical and computer engineering, EMEH. 2011, 11: 134–42p.

E.S. Wan Mahani Hafizah. Automatic region of interest generation for kidney ultrasound images ACS11, In: Proceedings of the 11th World Scientific and Engineering Academy and Society (WSEAS) International Conference on Applied computer science Stevens Point. Wisconsin, USA, 2011, 70–5p.

http://www.cs.waikato.ac.nz/ml/weka/.

http://www.cs.waikato.ac.nz/ml/weka/arff.html.

D. Gupta, D. Kumari, Raina. Hybrid feature based natural scene classification using neural network, Int J Comput Appl. 2012; 41(16): 48–52p.


Refbacks

  • There are currently no refbacks.