An Improved CBMIR Architecture, Based On Modified Classifiers & Feedback Method for Tumor Image Retrieval from MRI Images

Abstract

Abstract-MRI images are commonly used in medical field for various patients' analysis such as brain tumor detection. It is quite difficult and challenging task to efficiently retrieve accurate tumor various MRI image sets. Various researchers have been proposed several methods, but still improvements are required. Content based image retrieval methods are one of them with widely used in retrieving medical images from MRI data sets. In this research paper we are presenting improved CBMIR architecture, based on modified classifiers & feedback method for tumor image retrieval from MRI images. Proposed ICBMIR method uses new concepts of texture and shape properties of the images by using fuzzy c-means clustering and modified SVM classifier. Proposed ICBMIR methods also use subjective feedback method for efficient and accurate results. Experimental results clearly influence that proposed ICBMIR methods perform outstanding over existing CBMIR method in terms of accuracy, detection rate and time.