Title :
Functional MRI activation detection using genetic K-means clustering
Author :
Shi, Lin ; Heng, Pheng Ann ; Wong, Tien-Tsin
Author_Institution :
Dept. of Comput. Sci. & Eng., Chinese Univ. of Hong Kong, China
Abstract :
We propose a novel clustering approach to fMRI activation detection using a genetic K-means algorithm, which is more likely to find a global optimal solution to the K-means clustering, and is independent of the initial assignments of the cluster centroids. The experiments show that the proposed method solves fMRI activation detection problem with higher accuracy than ordinary K-means clustering.
Keywords :
biomedical MRI; genetic algorithms; pattern clustering; K-means algorithm; cluster centroids; fMRI activation detection problem; functional magnetic resonance imaging; genetic K-means clustering; Clustering algorithms; Clustering methods; Data preprocessing; Genetics; Independent component analysis; Iterative algorithms; Magnetic resonance imaging; Partitioning algorithms; Principal component analysis; Signal to noise ratio; activation detection; fMRI; genetic K-means clustering;
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
DOI :
10.1109/ICMLC.2005.1527215
