Martinerie et al. reply
Until 1998, neuroscientists thought that epileptic seizures began abruptly, just a few seconds before clinical onset. It was during that year that two independent studies4,8 showed that the non-linear time series analysis of EEG data could reveal dynamical changes several minutes before seizure onset. The usefulness of non-linear measures for the detection of pre-ictal changes has since been confirmed9. This new approach has opened a new field of seizure anticipation and defined a framework for better understanding of seizure generation mechanisms.
McSharry et al. have re-analyzed our 1998 database and have shown that the non-linear index is sensitive to amplitude variance fluctuation. We have been aware of this limitation for some time now. We developed, in 1999, a new method10 that did not involve the reconstruction of the dynamics from the amplitude of the signal, presenting a number of practical advantages over our previous method. The new method measures similarity to quantify the extent to which the EEG dynamics, reconstructed from the phase information, differ between periods taken at distant moments in time. The phase is defined as the time between two successive zero-crossing intervals. This relative measure reveals the spatial distribution of pre-ictal dynamic changes (both linear and non-linear) that involve the epileptogenic area but do not seem to be confined to the restricted ictal onset region. Furthermore, it is very robust against noise and artifacts, and fast enough to be carried out in real time.
The surrogate data that we had selected for the 1998 study to test the presence of deterministic structure in the time series4 had been built for each block of data (20 s; this may not have been not clear in the paper) and were designed to reject a null hypothesis of a non-linear transformation of linearly filtered noise. Thus, the variances of the raw data and surrogate data were the same. We found a statistical difference between the values of C(r0) calculated from the raw data and those calculated from the surrogate data, which led us to reject this null hypothesis. We know that one should be extremely careful with the use of surrogate procedures (which can be very sensitive to the presence of spikes in the data and detect spurious non-linearity11, for example). It is advisable to obtain consistent results with more than one type of surrogate, to get an indication of non-linear deterministic structure. New strategies and algorithms are now available12.
In conclusion, our recent results13,14 using the similarity method support the idea that pre-ictal dynamic changes (either linear, non-linear or both) have a higher probability of occurring before epileptic seizures. As previously reported15, McSharry et al. suggest that some linear methods can detect pre-ictal changes in a manner similar to non-linear methods. Both analyses probably constitute different ways of viewing the same thing and some combination of them will be a good method for reliable seizure anticipation. Real progress may require collaboration between research groups, which has already begun in an international program (special interest group session on engineering and epilepsy, 56th Annual Meeting of the American Epilepsy Society, Seattle, Washington, December 6–10, 2002; and the First International Conference on Seizure Anticipation, Bonn, Germany, April 24–27, 2002).
See "Prediction of epileptic seizures: are nonlinear methods relevant?" by McSharry et al.
References
Schreiber, T. Is nonlinearity evident in time series of brain electrical activity? in Chaos in Brain? (eds. Lehnertz, K., Elger, C.E., Arnhold, J. & Grassberger, P.) 13–22 (World Scientific, Singapore, 2000).
Theiler, J. On the evidence for low-dimensional chaos in an epileptic electroencephalogram. Phys. Lett. A 196, 335–341 (1995).
Grassberger, P. & Procaccia, I. Characterization of strange attractors. Phys. Rev. Lett. 50, 346–349 (1983).
Martinerie, J. et al. Epileptic seizures can be anticipated by non-linear analysis. Nat. Med. 4, 1173–1176 (1998).
Theiler, J., Eubank, S., Longtin, A., Galdrikan, B. & Farmer, J.D. Testing for nonlinearity in time series: the method of surrogate data. Physica D 58, 77–94 (1992).
Smith, L.A. Identification and prediction of low-dimensional dynamics. Physica D 58, 50–76 (1992)
Allen, M.R. & Smith, L.A. Monte Carlo SSA: detecting irregular oscillations in the presence of coloured noise. J. Clim. 9, 3373–3404 (1996).
Lehnertz, K. & Elger, C.E. Can epileptic seizures be predicted? Evidence from nonlinear time series analysis of brain electrical activity. Phys. Rev. Lett. 80, 5019–5022 (1998).
Sackellares, J. et al. Chaos in the Brain? (eds. Lehnertz, K. et al.) 112–133 (World Scientific, Singapore, 2000).
Le Van Quyen, M. et al. Anticipating epileptic seizure in real time by a nonlinear analysis of similarity between EEG recordings. NeuroReport 10, 2149–2155 (1999).
Schreiber, T. & Schmitz, A. Surrogate time series. Physica D 142, 346–382 (2000).
Rapp, P.E. et al. Surrogate data pathologies and the false-positive rejection of the null hypothesis. Int. J. Bifurcation Chaos 11, 983–997 (2001).
Le Van Quyen, M. et al. Anticipation of epileptic seizures from standard EEG recordings. Lancet 357, 183–188 (2001).
Navarro, V. et al. Seizure anticipation in human neocortical partial epilepsy. Brain 125, 640–655 (2002).
Litt, B. et al. Epileptic seizures may begin hours in advance of clinical onset: a report of five patients. Neuron 30, 51–64 (2001).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Rights and permissions
About this article
Cite this article
Martinerie, J., Le Van Quyen, M., Baulac, M. et al. Reply to "Prediction of epileptic seizures: are nonlinear methods relevant?". Nat Med 9, 242 (2003). https://doi.org/10.1038/nm0303-242
Issue Date:
DOI: https://doi.org/10.1038/nm0303-242
This article is cited by
-
Critical slowing down as a biomarker for seizure susceptibility
Nature Communications (2020)
