Detection Of Brain Disorders using Clustering based Techniques- Expectation Maximization

Brain is a central as well as an important part of a human body. The Brain activities are very complicated and difficult to understand. Many psychiatric disorders related to the brain are difficult to detect or identify. Functional magnetic resonance imaging or fMRI is a technique which is responsible for measuring a brain activity. The basic signal of fMRI depends upon the blood-oxygen-level-dependent (BOLD) effect which helps to study human brain functions. The purpose of Clustering Technique is to understand the complex interaction patterns among brain regions as well as identify brain disorders. To detect clusters of objects with similar interaction patterns we proposed a partitioning clustering algorithm that is Expectation Maximization (EM) algorithm. The Expectation Maximization (EM) algorithm is Gaussian Mixtures which begins with an initial guess to the cluster centers, and iteratively refines them an efficient algorithm for partitioning clustering.