330 likes | 545 Views
M.Tech. (CS), Semester III, Course B50. Functional Brain Signal Processing: EEG & fMRI Lesson 1. Kaushik Majumdar Indian Statistical Institute Bangalore Center kmajumdar@isibang.ac.in. Brain Signal Type. Functional (EEG, MEG, PET, fMRI) Structural (CT, MRI). Motivation to Study EEG.
E N D
M.Tech. (CS), Semester III, Course B50 Functional Brain Signal Processing: EEG & fMRILesson 1 Kaushik Majumdar Indian Statistical Institute Bangalore Center kmajumdar@isibang.ac.in
Brain Signal Type • Functional (EEG, MEG, PET, fMRI) • Structural (CT, MRI)
Motivation to Study EEG • Monitoring state of consciousness. • Brain Computer Interface. • Understanding cognitive processes. • Monitoring pathological brain conditions.
Human Brain or Neocortex https://eprints.usq.edu.au/19846/2/Bashar_2011_whole.pdf http://en.wikipedia.org/wiki/Human_brain
Functional Organization: Brodmann’s Areas http://spot.colorado.edu/~dubin/talks/brodmann/brodmann.html http://en.wikipedia.org/wiki/Human_brain
Electroencephalogram (EEG) Acquisition: 10 – 20 System http://www.ese.wustl.edu/~nehorai/research/eegmeg/EMEG-Overview.html http://en.wikipedia.org/wiki/10-20_system_%28EEG%29
Majumdar, IEEE Trans. Biomed. Eng., vol. 56(4), p. 1228 – 1235, 2009. Channels by Number
Multi-Channel EEG Signals http://maxim.ece.illinois.edu/teaching/fall08/
Mountcastle, Brain, 120:701-722, 1997. Six Layer Cortex
Cortical Basis of Scalp EEG Baillet et al., IEEE Sig. Proc. Mag., Nov 2001, p. 16.
Forward Problem : Schematic Head Model EEG Channels Source Scalp Skull Brain
Dipole Source Model (parametric model) Distributed Source Model (nonparametric model) Source Models
Majumdar, IEEE Trans. Biomed. Eng., vol. 56(4), p. 1228 – 1235, 2009. Distributed Source Model
Kybic et al., Phys. Med. Biol., vo. 51, p. 1333 – 1346, 2006 Forward Calculation • Poisson’s equation in the head
Hallez et al., J. NeuroEng. Rehab., 2007, open access. http://www.jneuroengrehab.com/content/4/1/46 Published Conductivity Values
Hallez et al., J. NeuroEng. Rehab., 2007, open access. http://www.jneuroengrehab.com/content/4/1/46 6 Parameter Dipole Geometry
Potential at any Single Scalp Electrode Due to All Dipoles r is the position vector of the scalp electrode rdip - i is the position vector of the ith dipole di is the dipole moment of the ith dipole
Generalization V = GD + n G is gain matrix, n is additive noise.
EEG Gain Matrix Calculation For detail of potential calculations see Geselowitz, Biophysical J., 7, 1967, 1-11.
Hallez et al., J. NeuroEng. Rehab., 2007, open access. http://www.jneuroengrehab.com/content/4/1/46 Nested Head Tissues SKULL BRAIN SCALP
Hallez et al., J. NeuroEng. Rehab., 2007, open access. http://www.jneuroengrehab.com/content/4/1/46 Finite Elements Method
Home Work: Loading EEG Data Files in Computer RAM and Familiarization • EEGLAB (needs MATLAB) available at http://sccn.ucsd.edu/eeglab/ • Sample EEG Data: http://sccn.ucsd.edu/eeglab/data/headit.html
Must Reading • Baillet, Mosher & Leahy, “Electromagnetic brain mapping,” IEEE Sig. Proc. Mag., p. 14 – 30, Nov 2001. • Hallez et al., “Review on solving the forward problem in EEG source analysis,” J. Neuroeng. Rehab., open access, available at http://www.jneuroengrehab.com/content/4/1/46
THANK YOUThis presentation is available at http://www.isibang.ac.in/~kaushik