Evaluating SVM and MLDA in the extraction of discriminant regions for mental state prediction

• Published in: NeuroImage Vol. 46; no. 1; pp. 105 - 114
• Main Authors: Sato, João Ricardo, Fujita, André, Thomaz, Carlos Eduardo, Martin, Maria da Graça Morais, Mourão-Miranda, Janaina, Brammer, Michael John, Junior, Edson Amaro
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 15.05.2009; Elsevier BV; Elsevier Limited
• Subjects: Acoustic Stimulation; Adolescent; Adult; Brain; Brain - physiology; Data compression; Female; Humans; Image Interpretation, Computer-Assisted; Image Interpretation, Computer-Assisted - methods; Magnetic Resonance Imaging; Male; Medical imaging; Middle Aged; Pattern recognition; Pattern Recognition, Automated; Pattern Recognition, Automated - methods; Photic Stimulation; Principal Component Analysis; Principal components analysis; Psychomotor Performance; Psychomotor Performance - physiology; Studies; Young Adult
• DOI: 10.1016/j.neuroimage.2009.01.032
• ISSN: 1053-8119; 1095-9572

Pattern recognition methods have been successfully applied in several functional neuroimaging studies. These methods can be used to infer cognitive states, so-called brain decoding. Using such approaches, it is possible to predict the mental state of a subject or a stimulus class by analyzing the spatial distribution of neural responses. In addition it is possible to identify the regions of the brain containing the information that underlies the classification. The Support Vector Machine (SVM) is one of the most popular methods used to carry out this type of analysis. The aim of the current study is the evaluation of SVM and Maximum uncertainty Linear Discrimination Analysis (MLDA) in extracting the voxels containing discriminative information for the prediction of mental states. The comparison has been carried out using fMRI data from 41 healthy control subjects who participated in two experiments, one involving visual–auditory stimulation and the other based on bi-manual fingertapping sequences. The results suggest that MLDA uses significantly more voxels containing discriminative information (related to different experimental conditions) to classify the data. On the other hand, SVM is more parsimonious and uses less voxels to achieve similar classification accuracies. In conclusion, MLDA is mostly focused on extracting all discriminative information available, while SVM extracts the information which is sufficient for classification.