Summary: Purpose:We studied heart‐rate (HR) changes at the transition from the preictal to the ictal state in patients with focal epilepsies to gain some insight into the mechanisms involved in the neuronal regulation of cardiovascular function.Methods:We assessed ECG changes during 145 seizures recorded with scalp EEG in 58 patients who underwent video‐EEG monitoring. Consecutive RR intervals were analyzed with a newly developed mathematical method for a total of 90 s.Results:Ictal‐onset tachycardia occurred in 86.9% of all seizures, whereas bradycardia was documented only in 1.4%. The incidence as well as the amount of ictal HR increase was significantly more pronounced in patients with mesial temporal lobe epilepsy (TLE) as compared with those with non‐lesional TLE or extratemporal epilepsy. Moreover, right hemispheric seizures were associated with ictal‐onset tachycardia. On average, ictal HR increase preceded EEG seizure onset by 13.7 s in TLE patients and 8.2 s in patients with extratemporal epilepsy. This difference was significant. Ictal HR changes could be classified according to their temporal evolution into two different patterns. These two patterns differed significantly between the temporal lobe and the extratemporal epilepsy patient group.Conclusions:Epileptic discharges directly influence areas of the central autonomic network, thus regulating HR and rhythm. Such changes occur before ictal discharges appear on surface electrodes. Our newly developed method may be of potential use for clinical applications such as automatic seizure‐detection systems. Moreover, our method might help to clarify further the basic mechanisms of interactions between heart and brain.
23 January 2007 in Methodology and Computing in Applied Probability
Wang and Pötzelberger (1997) derived an explicit formula for the probability that a Brownian motion crosses a one-sided piecewise linear boundary and used this formula to approximate the boundary crossing probability for general nonlinear boundaries. The present paper gives a sharper asymptotic upper bound of the approximation error for the formula, and generalizes the results to two-sided boundaries. Numerical computations are easily carried out using the Monte Carlo simulation method. A rule is proposed for choosing optimal nodes for the approximating piecewise linear boundaries, so that the corresponding approximation errors of boundary crossing probabilities converge to zero at a rate ofO(1/n2).
26 November 2001 in Mathematical Proceedings of the Cambridge Philosophical Society
We show that the asymptotic behaviour of the quantization error allows the definition of dimensions for probability distributions, the upper and the lower quantization dimension. These concepts fit into standard geometric measure theory, as the upper quantization dimension is always between the packing and the upper box-counting dimension, whereas the lower quantization dimension is between the Hausdorff and the lower box-counting dimension.
2 November 2004 in Mathematical Proceedings of the Cambridge Philosophical Society
We prove results on the asymptotic behavior of the quantization error for self-similar distributions $P$ when $n$, the number of prototypes, goes to infinity. In particular, we show that the suitably scaled quantization error converges, if $P$ is nonarithmetic and that it is asymptotically periodic in the arithmetic case. The results hold also for distributions that are absolutely continuous with respect to a self-similar distribution.