A Mathematical Approach for Feature Selection & Image Retrieval of Ultra Sound Kidney Image Databases

This paper aims to focus on the feature extraction, selection and database creation of ultra sound kidney images for image retrieval which will aid for computer assisted diagnosis. The impact of content-based access to medical images is frequently reported but existing systems are designed for only a particular context of diagnosis. But, our concept of image retrieval in medical applications aims at a general structure for semantic content analysis that is suitable for numerous applications in case-based reasoning. Around fifty ultrasound kidney image has been collected from the clinical laboratory and nearly fifteen features are extracted. By using these features, the database created for comparison. The created database contains the normal and abnormal kidney images. The statistical tool (Ttest) is used to measure the similarity between the stored database image and the query image. The image which is more similar to the query image is retrieved as the resultant image. If the query image does not match with the stored database image, it will be considered as the new image to the database system. This decision making system is developed in windows platform matlab7.