Data Mining and Knowledge Engineering

V. Saravanan ¹, S. Chitra ²

Affiliations
1 Dept. of CSE, Manakula Vinayagar Institute of Technology, IN
2 Dept of IT, Manakula Vinayagar Institute of Technology, IN

Abstract

Functional magnetic resonance imaging or functional MRI (fMRI) is a efficient neuroimaging procedure using MRI tools that dealings brain movement by detecting related changes in blood flow. The goal of fMRI data investigation is to detect correlations among brain activation and a task the subject performs during the scan. It also aims to determine correlations with the specific cognitive states, such as memory and recognition, induced in the subject. In this system, we propose a novel framework for clustering the essential fMRI signals based on their interactions and also correlation which is generated in a multivariate time series. To formalize this framework we cluster only Important Interactions based on the patient's medical records with the help of Essential Clustering Algorithm. The Essential clusters (EC) are then clustered again based on their dependencies on various brain regions. These EC's are grouped under specific models. The changes detected are mined based on the type of cluster grouped under a certain model. Our method shows that certainly increases the efficiency of the system along with increases in the effectiveness with minimal resource utilization.

Keywords

Clustering, Dependencies, Brain Region, Efficiency.

Full Text