Christian Meisel, MD, PhD: Expanding Preliminary Data on Seizure Forecasting

"If we combined data and increased the data sets from, potentially, other labs and build longer recording methods that collect longer periods of time, then we can improve this algorithm even further. At least, that’s what our data suggest.”

A study presented at the American Epilepsy Society (AES) Virtual Meeting, December 4–8, 2020, found easy-to-use, non-invasive devices in combination with deep learning can provide statistically significant and clinically meaningful seizure forecasting. Lead author Christian Meisel, MD, PhD, and colleagues used multi-modal wristband sensor data in combination with the Empactica E4 device on 69 persons with epilepsy (PWE) in their clinical trial.

The results proved to be positive, with a mean prediction horizon of 1896 (±101) seconds, a time that may be long enough to afford reasonable warning of seizures in advance. The researchers of the study concluded that the initial results may provide the basis for future re-evaluation, algorithm improvement and benchmarking as a step towards patient empowerment and objective epilepsy diagnostics for broad application.

Meisel, a neurologist from Universitätsmedizin Berlin and Berlin Institute of Health, sat down with NeurologyLive to discuss the specific ways of how to build from these results and what else needs to be done in the seizure forecasting field.

For more coverage of AES 2020, click here.

REFERENCE

1. Meisel C, El Atrache R, Jackson M, et al. Machine learning from wristband sensor data for wearable, non-invasive seizure forecasting. Presented at AES 2020 Annual Meeting; December 4–8, 2020; Abstract 241