Automated Feature Extraction and Retrieval of Ultra Sound Kidney Images using Maxi-Min Approach

A general purpose medical image retrieval framework has been proposed with two subsystems namely enrollment and the query subsystem. As an attempt to design a new content based image retrieval methodology following the above framework, MAXI-MIN approach is implemented for the ultra sound kidney images for the retrieval process. Around hundred ultrasound kidney images have been collected from the clinical laboratory and fourteen features have been extracted from the existing literature for database creation. The difference between the feature of query image and features of each image in the database has been calculated. The image which is more similar to the query image has been retrieved as the resultant image based on the maximum number of occurrences of features for the minimum difference. If the query image does not match with the stored database image, the query image is added as a new image in the database. The process is highly automated and the system is capable of working effectively across different issues without human interference.