An Image Mining Technique for Identifying Tuberculosis Meningitis of the Brain

Abstract

The main focus of image mining in the proposed system is concerned with the identification of meningitis in the membrane of brain using CT scan brain image. dentifying the type of tuberculosis affecting the meninges of brain is a crucial step in computer assisted Meningitis TB detection. The system proposes a method based on modified K mean clustering to enhance the diagnosis of medical images like CT scan brain image. The system analyzes medical images and automatically generates suggestions of diagnosis employing modified K mean clustering and Hu Moment Invariant method. The proposed method uses two important algorithm of image mining. The first method extracts features present in the CT scan of brain image and the second method cluster the type of meningitis present in the image. In the existing system it classifies the presence of bacteria through the sputum analysis and identifies the TB affecting the lung. The proposed system identifies Meningitis TB affecting the membranes of the brain. The method has been applied on several real datasets, and the results shows high accuracy to claim that the use of modified K mean clustering is a powerful means to assist in the diagnosing task.