Generalized epileptic seizure prediction based on HRV analysis and its mechanism consideration

Abstract

The present study aims to predict generalized seizures based on heart rate variability (HRV) analysis and multivariate statistical process control (MSPC). We applied the existing anomaly monitoring algorithm, i.e. HRV-based MSPC developed by Fujiwara et al. to 17 pre-ictal episodes and 74 inter-ictal episodes whose total length were about 63 hours. Consequently, the sensitivity and the false positive rate were 76.5% and 1.39 times per hour respectively, and the proportion of duration under false alarms was 5.96%. These results suggest that generalized seizures may be predicted by analyzing HRV. Based on the previous research, we hypothesize that the change in the autonomic nervous activity induce generalized epileptic seizures.