Web Based Diagnostic Classifier System for Kidney Lesions using Image Texture Parameters

Medical Informatics comprises the theoretical and practical aspects of information processing and communication, based on knowledge and experience derived from processes in medicine and health care. Telemedicine can be defined as the use of telecommunications technologies to facilitate healthcare delivery. Ultrasound imaging is a powerful tool for characterizing the state of soft tissues; however, in some cases, where only subtle differences in images are seen as in certain kidney lesions, existing B-scan methods are inadequate. With this idea, an attempt has been made in this paper for in-vivo differentiation of normal and kidney lesions using image texture parameters of ultrasound B-scan image and classification using Web based classifier. A set of 12 Image texture parameters were computed for ultrasonic scanned images of kidney. Artificial Neural Network with an input layer representing the Image texture parameters and an output layer representing the classification results were used. The 12 Image texture parameters were given as inputs to Neural Network and output layer with 3 target values were used. The back propagation algorithm trains a given feed forward multilayer neural network for a given set of input patterns with known classifications. The MATLAB´s web server toolbox and Neural Network toolbox were used for analyzing and classification. In the web based medical information system developed, the remote user can submit the ultrasound B-scan image using the Web Browser and the remote user will receive the diagnosis result in the desk using the Internet which is explained in the paper.