A Medical Image Re
A Medical Image Retrieval Framework using Genetically Optimized Elman Neural Network
1T. Baranidharan and 2D. K. Ghosh
Department of EIE, K.S.Rangasamy College of Technology, Tiruchengode- 637 215, India
E-mail: [email protected]
Department of Mathematics, V.S.B Engineering College, Karur, Tamilnadu state, INDIA
Abstract: In this paper, a Content Based Image Retrieval (CBIR) framework for medical images has been presented. An algorithm based on energy information obtained from Hilbert Transform for extracting features from medical images based on imaging modalities. The features are selected on the basis of the correlation among the extracted vectors depending upon the class label. An enhanced Genetic Algorithm Optimized based Elman (GAOE) Neural Network for image retrieval is presented. The performance of proposed method has been evaluated using a dataset consisting of 180 medical images. The experimental results demonstrated that the efficiency of the GAOE neural network is high compared to methods using low-level features.
[T. Baranidharan and D. K. Ghosh. A Medical Image Retrieval Framework using Genetically Optimized Elman Neural Network. Life Sci J 2013;10(4s):542-548] (ISSN: 1097-8135). http://www.lifesciencesite.com. 82
Keywords: Content Based Image Retrieval (CBIR), Medical images, Multilayer Perceptron, Neural Network, Genetic Optimization.