Machine Learning
Journal of Cognitive Neuroscience
Fast and robust fixed-point algorithms for independent component analysis
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
This paper proposes a novel method of automatic classification of magnetic resonance images based on independent component analysis (ICA). The ICA-based method is composed of three steps. First, all magnetic resonance imaging (MRI) scans are aligned and normalized by statistical parametric mapping. Then FastICA is applied to the preprocessed images for extracting specific neuroimaging components as potential classifying feature. Finally, the separated independent coefficients are fed into a classifying machine that discriminates among Alzheimer's patients, and mild cognitive impairment, and control subjects. In this study, the MRI data is selected from the Alzheimer's Disease Neuroimaging Initiative databases. The experimental results show that our method can successfully differentiate subjects with Alzheimer's disease and mild cognitive impairment from normal controls.