Ingo @ CISCP   Dr Ingo Bojak, PhD & MSc in Th. Physics (Dortmund)
  ibojak@swin.edu.au
  Postdoctoral Research Fellow in Brain Dynamics
  Sessional Lecturer: HET419 Physiological Modelling &
  HET408 Biomedical Imaging and Emerging Technologies    Postal Address:
     CISCP, LSS H31
     Swinburne University of Technology
     P.O. Box 218
     Hawthorn, VIC 3122, Australia

The movies can be played with QuickTime (Mac / Win) or with the QuickTime Alternative for Windows' MediaPlayer. In Linux use Xine.
It's my pleasure to thank Dr. Paul Bourke for his excellent Surface Animator.
Please also visit the homepage of Dr D.T.J. Liley.

Self-organized 40 Hz synchronization in a physiological theory of EEG
I. Bojak and D.T.J. Liley
submitted to Neurocomputing (CNS '06, accepted for oral presentation)

0.23 MB (local)

Movie: Hilbert phase freeze out
[ spikephase.mov , 8.4 MB, 400x300, plays for 15 sec, shows 0.77 sec of brain time]
This shows the rapid emergence of large scale (Hilbert instantaneous) phase correlations in the gamma frequency range for a cortex with marginal instability. Regions of similar Hilbert phase - relative to mean phase and unwrapped - share the same color in this full cortex (51.2 cm x 51.2 cm) simulation. Compare with Fig. 2 in the paper. The instability is caused by changing just one physiological parameter, namely by a 12.5% reduction of the inhibitory post-synpatic potential (IPSP) amplitude.

Movie: rapid change of dominant frequency
[ radpow2.mov , 5.6 MB, 400x300, plays for 17 sec, shows 1.0 sec of brain time]
This shows the rapid change of the dominant frequency in the (radial) power spectrum from low beta to gamma during roughly the same simulation time as in the Hilbert phase movie above. Note that the introduction of the marginal instability has previously shifted the dominant frequency from regular alpha to low beta and removed delta / theta contributions, compare Fig. 1 Right in the paper.

Understanding the transition to seizure by modelling the epileptiform activity of general anaesthetic agents
D.T.J. Liley and I. Bojak
Journal of Clinical Neurophysiology 22 (2005) 300-313, invited contribution

1.8 MB (external) [also available here (external)]

Movie: ictogenesis in EEG electrode resolution
[ spike32s.mov , 9.4 MB, 400x300, plays for 39 sec, shows 2.0 sec of brain time]
This shows part of a simulation of an epileptic seizure. Epileptiform activity spreads rapidly from a focal point. The electrical activity of underlying cortex is summed over 1.6 cm x 1.6 cm patches to approximate the effects of measuring with a high density EEG. (The movie shows the picture one would obtain with 32 x 32 electrodes, in practice only a subset of these could be measured.) The movie is based on a parameter set different from the one used for Fig. 12 in the paper.

Movie: ictogenesis in cortical macrocolumn resolution
[ spike512s.mov , 40.2 MB, 400x300, plays for 40 sec, shows 2.0 sec of brain time]
This shows the same simulation as above, but now at full simulation resolution. We see that the EEG electrode summation hides much of the underlying structure of the electrocortical activity.

Modelling the effects of anaesthesia on the electroencephalogram
I. Bojak and D.T.J. Liley
Phys. Rev. E 71 (2005) 041902, 1-22
selected for the Virtual Journal of Biological Physics Research, Volume 9, Issue 8

1.1 MB (external) [also available here (external)]

Movie: anaesthetic induction with "physiological" parameter set
[ a1537.mov , 30.6 MB, 400x300, plays for 2:27 min, shows 37 sec of brain time]
This parameter set yields a physiological power spectrum, appropriate mean firing rates, and a stable "biphasic" power surge under anaesthesia. Here an induction with the general anaesthetic agent isoflurane is simulated. First, for 3 s brain / 12 s movie time isoflurane concentration is zero. Next over 31 s brain / 123 s movie time isoflurane concentration is linearly raised to 0.81 mM (3.33 MAC equivalent). Finally, for 3 s brain / 12 s movie time the cortex is kept at highest concentration. Compare with Tab. 3 and Figs. 4,5,7,9 in the paper. [At every time step deviations from the current grid average have been enhanced by a factor 50 to make them clearly visible.]

Movie: (radial) power spectrum of the above
[ rad1537.mov , 15.7 MB, 225x882, plays for 37 sec, shows 37 sec of brain time]
This shows the (radial) power spectrum of the simulation run above. The y-axis gives frequency from 0 to 30 Hz, the x-axis gives 1/wavelength (radial) from 0 to 0.45/cm. The large letters on top show the isoflurane concentration in MAC. Compare with Fig. 4 in the paper.

Electrorhythmogenesis and anaesthesia in a physiological mean field theory
I. Bojak, D.T.J. Liley, P.J. Cadusch, and K. Cheng.
Neurocomputing 58-60 (2004) 1197-1202

0.41 MB (local)

Movie: standard parameter set with v = 7 m/s
[ EEG_v7_0.mov , 6.71 MB, 400x300, plays for 1:31 min, shows 9 sec of brain time]
This is a typical example of a run that leads to a stable alpha rhythm with "fast" mean cortico-cortical conduction velocity. Under colored noise input, large scale activation patterns self-organize slowly from the initial state. In the initial state the mean membrane potential is artificially set to the resting potential everwhere. The final alpha rhythm (in the last 8.192 s of the 15 s run, the first 9 s are shown here) has a frequency of 10.74 Hz and a wavelength of 17.1 cm.

Movie: standard parameter set with v = 3 m/s
[ EEG_v3_0.mov , 4.93 MB, 400x300, plays for 0:43 min, shows 4 sec of brain time]
This is a typical example of a run that leads to a stable alpha rhythm with "slow" mean cortico-cortical conduction velocity. Under colored noise input, small scale activation patterns self-organize quickly from the initial state. In the initial state the mean membrane potential is artificially set to the resting potential everwhere. The final alpha rhythm (in the last 8.192 s of the 15 s run, the first 4 s are shown here) has a frequency of 11.6 Hz and a wavelength of 12.8 cm.