Seizure Forecasting with Non-invasive and Minimally Invasive Mobile Devices: The Epilepsy Foundation’s My Seizure Gauge Study - European Medical Journal

Continue

Seizure Forecasting with Non-invasive and Minimally Invasive Mobile Devices: The Epilepsy Foundation’s My Seizure Gauge Study

2 Mins
Neurology
Authors:
†Pedro F. Viana,1,2 †Ewan S. Nurse,3,4 †Mona Nasseri,5 Phillippa Karoly,3,4 Tal Pal Attia,5 Nicholas Gregg,5 Boney Joseph,5 Caitlin Grzeskowiak,6 Matthias Dümpelmann,7 Mark Cook,3,4 Gregory A. Worrell,5 Andreas Schulze-Bonhage,7 Dean R. Freestone,3 Mark P. Richardson,1 *Benjamin H. Brinkmann5

1. Institute of Psychiatry, Psychology & Neuroscience, King’s College London, UK
2. Faculty of Medicine, University of Lisbon, Portugal
3. Seer Medical, Melbourne, Australia
4. University of Melbourne, Departments of Biomedical Engineering and Medicine, Australia
5. Department of Neurology, Mayo Foundation, Rochester, Minnesota, USA
6. Epilepsy Foundation of America, Bowie, Maryland, USA
7.Department of Neurosurgery, Epilepsy Center, University Medical Center, Faculty of Medicine, University of Freiburg, Germany
*Correspondence to [email protected]
†Equal author contribution

Disclosure:

Viana has received a grant from the Epilepsy Foundation of America for this manuscript, which was paid to his institution; payment or honoraria from Eisai Pharmaceuticals; and travel support from UNEEG Medical. Nurse has received a grant from the Epilepsy Foundation of America for this manuscript, which was paid to his institution; the BMTH3 grant from MTPConnect and CRC-P grant, both of which were paid to his institution; and has stock in Seer Medical. Nasseri received a grant (NSF CBET-2138378), which was paid to her institution. Karoly has received support from the Australian Government National Health and Medical Research Council for this manuscript; grants from the Australian Government National Health and Medical Research Council, Australian Research Countil, Australian Government Medical Research Future Fund, and L’Oréal-UNESCO For Women in Science Fellowship; and has two patents (US Patent 11,298,085 and US Patent 17,484,979). Attia has declared no conflicts of interest. Joseph has received the My Seizure Gauge grant from the Epilepsy Foundation of America for this manuscript, which was paid to his institution. Gregg is a recipient of the American Epilepsy Society Research and Training Fellowship for Clinicians; an investigator for the Medtronic Deep Brain Stimulation Therapy for Epilepsy Post-approval Study. Grzeskowiak has received salary support from the Epilepsy Foundation and is the Senior Director of Research and Innovation at the Epilepsy Foundation of America. Dümpelmann has declared no conflicts of interest. Cook has two patents (US Patent 10,568,574 and US Patent 11,298,085); is in a leadership or fiduciary role at the Epilepsy Foundation of Victoria, O’Brien Foundation, and St. Vincent’s Hospital Foundation; and has stock in Seer Medical and Epiminder. Worrell has received support from the Epilepsy Foundation of America (EI2 My Seizure Gauge) and the National Institutes of Health (NS123066 and R01NS 92882) for this manuscript; research support from UNEEG Medical; royalties or licenses from Cadence Neuroscience and NeuroOne; consulting fees from Applied Aerosol Technologies; travel expenses from Ictal 2022, which was paid to his institution; patents for various electroencephalogram and device analysis; is on the Data Safety Monitoring Board for Brain-Computer Interface Implant for People with Severe Communication Disability (FDA IDE Q191197), as part of the National Institutes of Health advisory panel; holds stock in NeuroOne; and received devices from Medtronic. Schulze-Bonhage has received funding from the Epilepsy Foundation (My Seizure Gauge) and a UNEEG grant. Freestone has received the My Seizure Gauge grant from the Epilepsy Foundation of America for this manuscript, which was paid to his organisation; the BMTH3 grant from MTPConnect and CRC-P grant, both of which were paid to his institution; and has stock in Seer Medical. Richardson has received a grant from the Epilepsy Foundation of America for this manuscript, which was paid to his institution. Brinkmann has received a grant from the Epilepsy Foundation of America (EI2 My Seizure Gauge) and the National Institutes of Health (NS123066) for this manuscript; research support from UNEEG Medical and Seer Medical; royalties or licenses from Cadence Neuroscience. and Seer Medical; consulting fees from Otsuka Pharmaceuticals; payment or honoraria from Ology Debates, which was paid to his institution; travel support from the Southern Epilepsy and EEG and International Federation for Clinical Neurophysiology, both of which were paid to his institution; holds various patents on electroencephalogram and wearable device analysis; is on the Data Safety Monitoring Board for Epitel; and received devices from Medtronic.

Citation:
EMJ Neurol. ;10[1]:35-38. DOI/10.33590/emjneurol/10101245. https://doi.org/10.33590/emjneurol/10101245.
Keywords:
Epilepsy, mobile devices, mobile health, seizure detection, seizure forecasting, wearable technology.

Each article is made available under the terms of the Creative Commons Attribution-Non Commercial 4.0 License.

BACKGROUND AND AIMS

Seizure unpredictability is consistently reported as one of the most disabling aspects of living with chronic epilepsy. Reliable forecasting systems could have far-reaching consequences for patients, from allowing protection at times of high seizure risk to improving confidence during everyday life at times of low seizure risk.1,2 The feasibility of forecasting has been demonstrated with chronic intracerebral electroencephalographic (EEG) recordings;3 however, invasive devices carry significant risks and are not appropriate for all patients.

The Epilepsy Foundation’s My Seizure Gauge Project is an ongoing multicentre study aiming to determine the feasibility of seizure forecasting with non-invasive and minimally invasive mobile devices (Figure 1).

Figure 1 The overarching goal of the Epilepsy Foundation’s My Seizure Gauge project is to establish the feasibility of seizure forecast

Figure 1: The overarching goal of the Epilepsy Foundation’s My Seizure Gauge project is to establish the feasibility of seizure forecasting using non-invasive and minimally invasive mobile devices.
Multimodal data can be incorporated into patient-specific forecasts of seizure risk.

MATERIALS AND METHODS

Adult patients with active (>10 seizures/6 months) drug-resistant epilepsy were enrolled for ultra-long-term (>8 months) monitoring. Patients were asked to self-report their seizures on an electronic diary (Seer app [Seer, Melbourne, Australia]) while using a wearable device (Empatica E4 [Empatica, Boston, Massachusetts, USA], Fitbit Charge 4/HR [Fitbit, San Francisco, Calirfornia, USA], or Fitbit Inspire) and simultaneous chronic ambulatory EEG monitoring (RNS [NeuroPace, Mountain View, California, USA], 24/7 EEG™ SubQ [UNEEG, Allerød, Denmark] or Epi-Minder sub-scalp [Bionics Institute, Melbourne, Australia] systems). Recorded data from these multiple modalities was analysed via traditional machine learning or deep learning methods, together with the identification of patient-specific circadian and multiday seizure risk cycles to forecast seizures.

RESULTS

To date, 40 enrolled subjects have recorded over 11,400 days (>31 years) of ambulatory data, and over 1,700 seizures have been annotated. Nine patients left the study prematurely due to device malfunctions, complications, poor adherence, poor data quality, or unanticipated seizure freedom. However, 20 patients are continuing to record data and 11 have completed the study.

Selected results from analysis of this cohort have been presented. Circadian and multiday seizure cycles were detected with ultra-long-term subcutaneous EEG.4,5 Heart rate circadian and multiday cycles were also detected in a high proportion of patients using a commercial fitness tracker (Fitbit), and were found to correlate with self-reported seizure likelihood.6 Using data from a multimodal wearable device (Empatica E4), electrodermal activity, heart-rate, and accelerometery data were significantly correlated with electrographic seizures detected by the RNS device in 11 patients. In another group of 11 subjects, patient-specific seizure forecasts trained on Fitibit data were found to be significantly better than chance in 91–100% of subjects.7 Seizure forecasting was also better than chance in five out of six patients using the Empatica E4 data, validated on RNS-detected seizures.8 Finally, seizure forecasting using ultra-long-term subcutaneous EEG data was significantly better than chance in at least 50% of patients in a six-patient cohort.9,10

CONCLUSION

This project has established the feasibility of forecasting seizures using seizure cycles, wearable devices, and subcutaneous EEG. Seizure cycles are common in patients with epilepsy. They are measurable across a range of non-invasive wearables, minimally invasive EEG, and intracranial EEG devices, and they are strong forecasts of seizure likelihood. The next steps in this project are to establish a freely available data science competition for forecasting using wearable data, and to trial a prospective seizure forecasting smartphone app (Seer), which is now available to the public.

References
Dumanis SB et al. Seizure forecasting from idea to reality. Outcomes of the My Seizure Gauge Epilepsy Innovation Institute workshop. eNeuro. 2017;4(6):ENEURO.0349-17.2017. Grzeskowiak CL, Dumanis SB. Seizure forecasting: patient and caregiver perspectives. Front Neurol. 2021;12:717428. Cook MJ et al. Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study. Lancet Neurol. 2013;12(6):563-71. Viana PF et al. 230 days of ultra long-term subcutaneous EEG: seizure cycle analysis and comparison to patient diary. Ann Clin Transl Neurol. 2021;8(1):288-93. Stirling RE et al. Seizure forecasting using a novel sub-scalp ultra-long term EEG monitoring system. Front Neurol. 2021;12:713794. Karoly PJ et al. Multiday cycles of heart rate are associated with seizure likelihood: an observational cohort study. EBioMedicine. 2021;72:103619. Stirling RE et al. Forecasting seizure likelihood with wearable technology. Front Neurol. 2021;12:704060. Nasseri M et al. Ambulatory seizure forecasting with a wrist-worn device using long-short term memory deep learning. Sci Rep. 2021;11:21935. Viana PF et al. Seizure forecasting using minimally invasive, ultra-long-term subcutaneous electroencephalography: individualized intrapatient models. Epilepsia. 2022;DOI:10.1111/epi.17252. Pal Attia T et al. Seizure forecasting using minimally invasive, ultra-long-term subcutaneous EEG: generalizable cross-patient models. Epilepsia. 2022;DOI:10.1111/epi.17265.

Please rate the quality of this content

As you found this content interesting...

Follow us on social media!

We are sorry that this content was not interesting for you!

Let us improve this content!

Tell us how we can improve this content?