“Freiburg Evoked Potentials”(EP2000) fully implements stimulation, recording and analysis of visual evoked potentials, be it from the retina (ERG, PERG) or the cortex (VEP). It runs on any standard recent Macintosh computer in demo mode, and requires just a second graphics card with monitor and the data acquisition card “PCI- MIO-16XE-50” from National Instruments. The system consists of two modules, a program for data acquisition (implemented in C++) and an analysis package running under IGOR Pro by Wavemetrics (the latter comes with full source code). [March 2004: an excellent solution now is to use the PowerMac G4 (1.25 GHz) with the video card “ATI Radeon 9000 Pro”. This model supports Mac OS 9 boot, and the video card can drive two monitors. An “ADC adaptor” to connect the second monitor is necessary. I have just tried this in London and it worked “out of the box”.]
Since Jan 2005 it is no longer possible to buy a suitable Mac from Apple – they don't offer any more OS9-bootable machines. All the more pressing it is to work on the OS X version. We’ll see, I need it myself.
In Jan 2005 EP2000 is routinely in used in 5 labs: Michelle McKerral, Montréal, Canada; Vaegan, Visiontest, Sydney, Australia; Jaegle, Tübingen, Germany; Potter, Boston, USA; and my own. EP2000 has already provided data for a number of publications.
Future roadmap: Analysis tools for grand means (done). Conversion to OS X, EEG acquisition via sound inputs, and plug-in architecture for custom stimuli still under way – I started by porting my ERG & EOG recording systems to OS X, was finished in 2007. Multifocal paradigm planned. Analysis tools are constantly updated in the lab. Should you want to use the system, don't hesitate to request the latest versions, the present page will not be regularly updated.
Jan 2005: fixed latency error in the on-line display, added Laplacian analysis
April 2004: Stimulation paradigm “PERGLA” after Porciatty (2004)
March 2004: The dual-video card Radeon 9000 works perfectly and obviates the need of a 2nd video card
Feb 2004: Now EP2000 runs in Montréal, Sydney, Tübingen & Boston, in addition to, yes, Freiburg
Jun 2003: more stimulus pictures, manual peak selection, grand mean facility
Nov 2002: Alternating hemifield checkerboards. Added “ep2000-setup.rtf”
Oct 2002: Sinusoidal gratings for Vaegan
Mar 2002: Stimuli exported as Quicktime movies
Jan 2002: Motion-based P300 paradigm, added 10-20 locations
Jul 2001: Read display setting from preferences
Jun 2001: Texture seg stimuli (motion + orientation)
May 2001: On-line Fourier analysis
Feb 2001: Double pulse experiments
Nov 2000: Motion stimuli
Sep 2000: Balloon help, mirror oscilloscope
Aug 2000: Abort with save, resume with “prevEP”
Jul 2000: Recording reliably
Two screens: one with the stimulus, viewed by patient,
the other screen for the operator (see picture below) sporting
the current stimulus pattern (in reduced size, with live update)
an on-line osccilloscope with signals from eye and scalp
patient data + status info
averaged traces, arranged by stimulus, on-line Fourier analysi for steady-state
Wide range of stimuli including:
motion onset (dart board & blobs)
alternating hemifield reversal
texture segregation (orientation & motion)
sinusoidal gratings with contrast & spatial frequency sweep
pulsed patterns with variable ISI
long-term potentiation & depression paradigms
Sequences of stimuli (e.g. different check sizes) are interleaved for optimal comparison
Stimuli are rendered with 800 x 600 pixel resolution at 75 Hz
The stimulus monitor is calibrated and linearized (“gamma correction”)
Stimulus with a fixation target of adjustable size, plus random digit to be reported by patient for fixation control
Stimulus sequences can be exported as a movie
On-line Fourier analysis for steady-state to allow later statistical tCirc evaluation
Amplification can be automatically calibrated (if your preamp has a build-in calibration signal)
Analysis is fully automated with the following features
full or selective plot of session data
1 or 2 column display (right eye / left eye)
phase-free low-pass filtered trace with optionally superimposed raw data, trend removal
automatic peak finding, automatically taking stimulus type into account
optional manual peak identification
automatic calculation of the Fourier spectrum, where appropriate
graphical indication of the stimulus properties (type & shape, check size, contrast)
automatic acuity estimate where appropriate
easy transfer of analysis results to a spreadsheet program
export of selected traces
separate module for grand mean across subjects or conditions
Acquisition module / das Messprogramm
A screenshot of acquisition module just after a 3-checksize sequence has been acquired.
This is a newer version of the control dialog (bottom left in the figure above) which may help you to appraise some of the options (balloon help is on).
Analysis module / das Plotprogramm
Two printouts from the analysis package [NB: I choose a pathological VEP here on purpose]:
The 3 VEP traces on the right were record to 3 different check sizes as shown on the left. The small dots represent the original, unfiltered trace, digitally low-pass filtering at 45 Hz results in the continuous line. All peaks were detected automatically.
The peak latencies of 125 and 122 ms are clearly in the pathologic range, the amplitudes are lower than the normal mean but still in the normal range. There is significant noise intrusion, so (as always, not depicted here) a second recording was performed to assess the reliability of these peaks.
4 PERGs: two eyes, two check sizes, rapid stimulus (steady state) with corresponding Fourier spectra on the right.
EP2000 has been used in the following publications:
Bach M, Maurer JP, Wolf ME (2008) Visual evoked potential-based acuity assessment in normal vision, artificially degraded vision, and in patients. Brit J Ophthalmol 92:396–403