Brain Anomaly Prediction Using Modified Clustering(MC)

Clustering methods are used generally for the image segmentation purpose. Since the edge of a brain image is a complex pattern, it is very difficult to accurately detect three different tissue segments: white matter, gray matter and cerebrospinal fluid as well as different diseases that is affected in the brain. In this paper we have made a modified clustering to put forward a new method, which is Modified clustering. We have used this method for brain-image disease detection like-the tumor and gray matter related anomaly detection. In this paper, we have decided to give color to the detected anomaly part. All the colored clusters are merged to get the final colored image. By the color pixels the percentage of white matter, the percentage of grey matter and the percentage of cerebrospinal fluid can be calculated. Then depending on the percentage of color, the brain anomaly is predicted according to a dataset; which contains the name of the brain disease and the percentage of gray matter and white matter of the affected person.