Bayesian common spatial patterns for multi-subject EEG classification

• Published in: Neural networks 2014-09, Vol.57, p.39-50
• Main Authors: Kang, Hyohyeong, Choi, Seungjin
• Format: Article
• Language: English
• Subjects: Algorithms; Applied sciences; Bayes Theorem; Biological and medical sciences; Brain Waves; Brain–computer interface; Central nervous system; Common spatial patterns; Computer science; control theory; systems; Data Interpretation, Statistical; Data processing. List processing. Character string processing; EEG classification; Electrodiagnosis. Electric activity recording; Electroencephalography - classification; Electroencephalography - methods; Electrophysiology; Exact sciences and technology; Fundamental and applied biological sciences. Psychology; Humans; Indian Buffet processes; Investigative techniques, diagnostic techniques (general aspects); Linear inference, regression; Mathematics; Medical sciences; Memory organisation. Data processing; Models, Neurological; Nervous system; Nonparametric Bayesian methods; Probability and statistics; Sciences and techniques of general use; Software; Statistics; Vertebrates: nervous system and sense organs
• ISSN: 0893-6080; 1879-2782
• DOI: 10.1016/j.neunet.2014.05.012

Multi-subject electroencephalography (EEG) classification involves algorithm development for automatically categorizing brain waves measured from multiple subjects who undergo the same mental task. Common spatial patterns (CSP) or its probabilistic counterpart, PCSP, is a popular discriminative feature extraction method for EEG classification. Models in CSP or PCSP are trained on a subject-by-subject basis so that inter-subject information is neglected. In the case of multi-subject EEG classification, however, it is desirable to capture inter-subject relatedness in learning a model. In this paper we present a nonparametric Bayesian model for a multi-subject extension of PCSP where subject relatedness is captured by assuming that spatial patterns across subjects share a latent subspace. Spatial patterns and the shared latent subspace are jointly learned by variational inference. We use an infinite latent feature model to automatically infer the dimension of the shared latent subspace, placing Indian Buffet process (IBP) priors on our model. Numerical experiments on BCI competition III IVa and IV 2a dataset demonstrate the high performance of our method, compared to PCSP and existing Bayesian multi-task CSP models.