Combined SVM and PCA to Recognize the Brain Function from fMRI Images
Rong Chun Guo, Songyun Xie, Xina Cheng, Haitao Zhao · 2009
In this paper, SVM and PCA are incorporated to classify brain fMRI images. This method well overcomes the difficulty of classifying high-dimensional data. PCA is utilized to extract the most representative features. SVM classifier based on selected features is trained to decode brain states. Experimental results show that the proposed method yields good performance. The correct classification rate of our bi-class recognition problems reaches as high as 97%.