SeminarsClassification on Riemannian manifolds - Application to Brain-Computer Interfacing
94
reads
Neurological disorders can disrupt the pathways through which the brain controls its external environment. Patients may lose all voluntary muscle control and may be completely locked into their bodies. A potential solution is to provide the brain with a new communication and control channel, a brain-computer interface (BCI) for conveying messages and commands to the external world. A portable, inexpensive and non-invasive BCI technology is based on electroencephalogram readouts, typically effected through 20-30 sensor channels. Such a technology produces a big deal of data to be pre-processed and classified. The acquired data may summarized via covariance matrices, which belong to the manifold of symmetric, positive-definite matrices. Covariance matrices of EEG signals contain the essential information about the correlation between groups of neurons occurring during the execution of mental tasks. The talk reviews classification methods on Riemannian manifolds are discusses experimental results.
reads
Simone Fiori
2014-12-19
13:30:00 - 14:30:00
308 , Mathematics Research Center Building (ori. New Math. Bldg.)
Neurological disorders can disrupt the pathways through which the brain controls its external environment. Patients may lose all voluntary muscle control and may be completely locked into their bodies. A potential solution is to provide the brain with a new communication and control channel, a brain-computer interface (BCI) for conveying messages and commands to the external world. A portable, inexpensive and non-invasive BCI technology is based on electroencephalogram readouts, typically effected through 20-30 sensor channels. Such a technology produces a big deal of data to be pre-processed and classified. The acquired data may summarized via covariance matrices, which belong to the manifold of symmetric, positive-definite matrices. Covariance matrices of EEG signals contain the essential information about the correlation between groups of neurons occurring during the execution of mental tasks. The talk reviews classification methods on Riemannian manifolds are discusses experimental results.