Semi-blind source separation in EM brain signal processing

• Published in: 3rd IEE International Seminar on Medical Applications of Signal Processing pp. 87 - 92
• Main Authors: James, C.J, Hesse, C.W
• Format: Conference Proceeding
• Language: English
• Published: Stevenage IEE 2005
• Subjects: Bioelectric signals; Biomagnetic signals; Biomagnetism; Electric and magnetic fields (medical uses); Electrical activity in neurophysiological processes; Electrodiagnostics; Other topics in statistics; Probability theory, stochastic processes, and statistics; Signal processing and detection; electroencephalography; frequency distribution; EM brain signal processing; ICA; brain-computer interface; semiblind source separation; diseases; epileptic seizure; spatio-temporal distribution; brain; source distribution; spatiotemporal phenomena; MEG; medical signal processing; independent component analysis; neurophysiological expectation; neurophysiology; brain activity; EM brain signal; blind source separation; magnetoencephalography; EEG recording
• ISBN: 9780863415708; 0863415709
• DOI: 10.1049/ic:20050337

The processing of electromagnetic (EM) brain signals can be viewed as the identification and separation of a series of overlapping sources of brain activity with slowly varying source distribution and/or levels of activity. blind source separation (BSS) techniques such as independent component analysis (ICA) lend themselves well to the analysis of such problems. The problem, however, still remains largely ill- posed even through the use of powerful assumptions such as those posed in ICA and other such techniques. It is generally the case in EM brain signals that a certain level of a priori knowledge is available on the spatio-temporal and/or frequency distribution of the activities of interest, based on neurophysiological expectations. Here we describe a number of techniques we employ in posing constraints on the ICA of different EM brain signal problems. The problems we consider include the analysis of spontaneous EEG recordings, to detecting and tracking epileptic seizures, and to detecting and enhancing evoked potentials for brain-computer interfacing paradigms.