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

Autor(es)

Sato, Joao Ricardo ^[1] ; Fujita, Andre ^[2] ; Thomaz, Carlos Eduardo ^[3] ; Morais Martin, Maria da Graca ; Mourao-Miranda, Janaina ^[4] ; Brammer, Michael John ^[4] ; Amaro Junior, Edson

Afiliação do(s) autor(es)

^[1] Univ Sao Paulo, Sch Med, Hosp Clin, Inst Radiol, NIF LIM44, BR-05403001 Sao Paulo - Brazil

^[2] Univ Tokyo, Inst Med Sci, Tokyo 1138654 - Japan

^[3] Ctr Univ FEI, Sao Paulo - Brazil

^[4] Kings Coll London, Inst Psychiat, Brain Image Anal Unit, London - England

• Tipo de documento: Artigo Científico
• Fonte: NeuroImage; v. 46, n. 1, p. 105-114, MAY 15 2009.
• Citações Web of Science: 33

Resumo

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 bimanual 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. (C) 2009 Elsevier Inc. All rights reserved. (AU)

• Processo FAPESP: 05/02899-4 - Imagem, estatística e data mining: métodos computacionais para análise do cérebro humano
• Beneficiário: Carlos Eduardo Thomaz
• Linha de fomento: Auxílio à Pesquisa - Apoio a Jovens Pesquisadores