Portal de Periódicos da CAPES

Discrimination of Resting-State fMRI for Schizophrenia Patients with Lattice Computing Based Features

2013; Springer Science+Business Media; Linguagem: Inglês

10.1007/978-3-642-40846-5_48

1611-3349

Darya Chyzhyk, Manuel Graña,

EEG and Brain-Computer Interfaces

Resting state fMRI data can be used to find biomarkers of specific neurological conditions, such as schizophrenia. In this paper we report results on the discrimination between schizophrenia patients and healthy control, as well as the discrimination of subpopulations of schizophrenia patients with and without auditory hallucinations. Data features for classification are obtained as follows: a Multivariate reduced ordering based on a h-function constructed from Lattice Autoassociative Memories recall. The Pearson correlation coefficient between the h-function values and the categorical variable at each voxel site allows to identify the most informative voxel sites. Feature vectors are constructed as the h-function values at these sites. Results on a database of healthy controls and schizophrenia patients with and without auditory hallucinations show that the approach can provide accurate discrimination between these populations.