Wearable Biosensors Inform Circadian and Infradian Cycles in Epilepsy
Abstract number :
3.09
Submission category :
2. Translational Research / 2B. Devices, Technologies, Stem Cells
Year :
2021
Submission ID :
1826374
Source :
www.aesnet.org
Presentation date :
12/6/2021 12:00:00 PM
Published date :
Nov 22, 2021, 06:53 AM
Authors :
Boney Joseph, MBBS - Mayo Clinic; Nicholas Gregg, MD - Bioelectronics Neurology and Engineering Laboratory, Department of Neurology - Mayo Clinic; Tal Pal Attia - Bioelectronics Neurology and Engineering Laboratory, Department of Neurology - Mayo Clinic; Mona Nasseri, PhD - School of Engineering - University of North Florida; Ewan Nurse, PhD - Seer Medical Pty Ltd; Pedro Viana, MD - Institute of Psychiatry, Psychology and Neuroscience - King’s College London; Rachel Stirling - Department of Biomedical Engineering - The University of Melbourne; Philippa Karoly, PhD - Department of Biomedical Engineering - The University of Melbourne; Andreas Schulze-Bonhage, MD, PhD - Epilepsy Center, Department for Neurosurgery - University Medical Center Freiburg; Gregory Worrell, MD, PhD - Bioelectronics Neurology and Engineering Laboratory, Department of Neurology - Mayo Clinic; Dean Freestone, PhD - Seer Medical Pty Ltd; Mark Richardson, BMBCh, PhD, FRCP - Institute of Psychiatry, Psychology and Neuroscience - King’s College London; Benjamin Brinkmann, PhD - Bioelectronics Neurology and Engineering Laboratory, Department of Neurology - Mayo Clinic
Rationale: Historical data from epileptic colonies and chronic brain recordings from humans and animals have established that seizure timing is not random for many individuals; rather, seizure risk is modulated over circadian and infradian (multiday) cycles. Physiological cycles are ubiquitous, transcending species, organ system, and diseases. Relationships between physiological signals amenable to non-invasive monitoring (cardiac, autonomic, behavioral) and brain excitability may enable seizure risk forecasting by wearable biosensors. Here, we use concurrent recordings from a brain implant and a wrist worn biosensor to assess if noninvasively measured physiology informs circadian and infradian rhythms in epilepsy.
Methods: Nine patients with epilepsy had chronic recordings with the NeuroPace RNS® System and the Empatica E4 wristband. Median recording duration was 131 days (std. dev. 75). The RNS® device provided measures of interictal epileptiform activity (IEA) and electrographic seizures (epileptologist visually confirmed). The E4 provided measures of cardiac (inter-beat interval (IBI), blood pressure (BP), heart rate variability (HRV)) and autonomic activity (tonic and phasic electrodermal activity (tEDA, pEDA)), temperature (TEMP), and 3-axis accelerometry (ACC). The continuous wavelet transform characterized circadian and infradian cycles from hourly averaged IEA and wristband biosignals. Cycles of interest were defined as peaks in the IEA amplitude spectral density (ASD) above the 95th percentile of white noise. Magnitude squared coherence assessed phase synchrony between IEA and wristband recordings. Seizure timing was evaluated relative to the phase of significant circadian and infradian cycles. Phase was assessed by 3rd order least-squares non-causal finite impulse response filters. Circular statistics (Rayleigh test) determined the phase locking value (PLV) of seizures to biosignal cycles (P< 0.05 significance level).
Translational Research