Detection of Epileptogenic Sequences in Video Material: The Role of Color.
Abstract number :
1.140
Submission category :
Year :
2001
Submission ID :
324
Source :
www.aesnet.org
Presentation date :
12/1/2001 12:00:00 AM
Published date :
Dec 1, 2001, 06:00 AM
Authors :
J. Parra, MD, Clinical Neurophysiology, Dutch Epilepsy Clinics Foundation, Meer & Bosch, Heemstede, Netherlands; S.N. Kalitzin, PhD, Medical Physics, Dutch Epilepsy Clinics Foundation, Meer & Bosch, Heemstede, Netherlands; E. Dekker, REEGT, Clinical Neuro
RATIONALE: Photosensitive epilepsy is the most common reflex epilepsy, affecting up to 10% of epileptic children. The overall presence of visual stimuli in our modern environment facilitates the manifestation of this trait. This study is aimed at validating a new method to detect epileptogenic sequences using the popular cartoon (Pokémon).
METHODS: We selected 12 potentially epileptogenic video scenes of equal length that were presented to 22 photosensitive patients (mean age 20 years, range 8-51 years) on a 100 Hz TV at 50 cm distance, both in color and black & white (BW) using an A-B-B-A paradigm. A human-eye sensor was used to correlate the variances in luminance with the EEG changes. We digitized the different scenes at 480 Hz and analyzed the three spectral components (red, green and blue) with a set of scaled Gabor filters (frequency range 1-40 Hz) Time and frequency localized amplitudes and modulation depths for the three colors were measured from the filter coefficients. Photoparoxysmal responses(PPRs)were classified according to Waltz et al. (Electroencephalogr Clin Neurophysiol 1992;83:138-145).
RESULTS: Two scenes with an intense red modulation triggered PPRs more consistently than the rest, eliciting at least one PPR in 17 of the 22 (77%) patients. All of them had PPRs with the color scenes, whereas only 3 patients developed PPRs with the BW ones as well. PPRs elicited by these color scenes were reproducible in 13 of 17 (76%) patients, whereas those triggered by the BW ones were reproducible only in one patient. Color scenes provoked more generalized PPRs (type 4), than those triggered by BW scenes which were more restricted (type 2-3), except in one patient. The onset of the PPRs was well correlated with the appearance of this red modulation, independently of the changes in luminance.
CONCLUSIONS: Changes in red modulation appear to be a powerful epileptogenic factor, independent of the changes in luminance.This new digital method opens the possibility of on-line automatic detection and suppression of epileptogenic sequences in video material. Further studies are necessary to assess the role of different color combinations.