RESUMEN
OBJECTIVE: Epilepsy monitoring unit (EMU) admissions are critical for presurgical evaluation of drug-resistant epilepsy but may be nondiagnostic if an insufficient number of seizures are recorded. Seizure forecasting algorithms have shown promise for estimating the likelihood of seizures as a binary event in individual patients, but methods to predict how many seizures will occur remain elusive. Such methods could increase the diagnostic yield of EMU admissions and help patients mitigate seizure-related morbidity. Here, we evaluated the performance of a state-space method that uses prior seizure count data to predict future counts. METHODS: A Bayesian negative-binomial dynamic linear model (DLM) was developed to forecast daily electrographic seizure counts in 19 patients implanted with a responsive neurostimulation (RNS) device. Holdout validation was used to evaluate performance in predicting the number of electrographic seizures for forecast horizons ranging 1-7 days ahead. RESULTS: One-day-ahead prediction of the number of electrographic seizures using a negative-binomial DLM resulted in improvement over chance in 73.1% of time segments compared to a random chance forecaster and remained >50% for forecast horizons of up to 7 days. Superior performance (mean error = .99) was obtained in predicting the number of electrographic seizures in the next day compared to three traditional methods for count forecasting (integer-valued generalized autoregressive conditional heteroskedasticity model or INGARCH, 1.10; Croston, 1.06; generalized linear autoregressive moving average model or GLARMA, 2.00). Number of electrographic seizures in the preceding day and laterality of electrographic pattern detections had highest predictive value, with greater number of electrographic seizures and RNS magnet swipes in the preceding day associated with a higher number of electrographic seizures the next day. SIGNIFICANCE: This study demonstrates that DLMs can predict the number of electrographic seizures a patient will experience days in advance with above chance accuracy. This study represents an important step toward the translation of seizure forecasting methods into the optimization of EMU admissions.
Asunto(s)
Epilepsia , Humanos , Teorema de Bayes , Epilepsia/diagnóstico , Convulsiones/diagnóstico , Técnicas y Procedimientos DiagnósticosRESUMEN
PURPOSE: The responsive neurostimulation system (RNS) aims to improve seizures by delivering electrical stimulation in response to epileptiform patterns detected by electrocorticograms. Seizure-onset patterns (SOPs) correspond to outcomes in intracranial EEG (IC-EEG), although whether this is true for RNS is unknown. This study characterizes common RNS SOPs and correlates them with seizure outcomes. METHODS: Among 40 patients with RNS implants, long-episode electrocorticogram characteristics of each patient's seizures were classified by visual analysis as one of the eight patterns previously described in IC-EEG. Correlation between each type of SOP and eventual seizure outcome was analyzed, with ≥50% improvement in a number of patient-reported seizure counts defined as a favorable outcome. RESULTS: Across 263 LEs analyzed, the most common SOP observed was low-voltage fast activity. There was no difference between the distribution of RNS SOPs and that of IC-EEG SOPs described in the literature (Kolmogorov-Smirnov test, P = 0.98). Additionally, there was no correlation between any particular SOP and favorable outcomes (Fisher's omnibus test, P = 0.997). CONCLUSION: This initial description of RNS SOPs finds them to be similar to previously described IC-EEG SOPs, which suggests similar prognostic/therapeutic potential. However, we found that RNS efficacy is independent of patient SOP, suggesting that RNS is likely an equally effective treatment for all SOPs. Future research on stimulation parameters for particular RNS SOPs and correlation with IC-EEG SOPs in the same patients would be instrumental in guiding personalized neurostimulation.