Functional magnetic resonance imaging (fMRI) ???brain reading???: detecting and classifying distributed patterns of fMRI activity in human visual cortex, NeuroImage, vol.19, issue.2, pp.261-270, 2003. ,
DOI : 10.1016/S1053-8119(03)00049-1
Decoding mental states from brain activity in humans, Nature Reviews Neuroscience, vol.16, issue.7, pp.523-534, 2006. ,
DOI : 10.1038/nrn1931
Estimating mutual information, Physical Review E, vol.69, issue.6, 2004. ,
DOI : 10.1103/PhysRevE.69.066138
Mutual information for the selection of relevant variables Fig. 5. Representation of the voxels selected by the MI algorithm (green) and the Anova method (values of the F-score between black and yellow), in the case of discrimination between the 3 different sizes, for the subject 6, with ? = 0.05 and k = 20. The main activity found by MI is at the localization (42,-75,-5) mm, in the lateral occipital cortex. The localization of the activity is far more defined with the MI selection than the reference method, and allows a better interpretation of the results, Chemometrics and intelligent laboratory systems, pp.215-226, 2006. ,
Resampling methods for parameter-free and robust feature selection with mutual information, Neurocomputing, vol.70, issue.7-9, pp.1276-1288, 2007. ,
DOI : 10.1016/j.neucom.2006.11.019
Support-vector networks, Machine Learning, pp.273-297, 1995. ,
DOI : 10.1007/BF00994018
The relevance vector machine, Advances in Neural Information Processing Systems, 2000. ,
Multivariate adaptive regression splines (with discus-sion), Ann. Stat, vol.9, 1991. ,
DOI : 10.1214/aos/1176347963
URL : http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.382.970
Functional magnetic resonance imaging activity patterns in human lateral occipital complex carry information about object exemplars within category, Journal of Cognitive Neuroscience, vol.20, issue.2, 2007. ,