Forecasting Seizure Likelihood with Wearable Technology
Abstract number :
3.078
Submission category :
2. Translational Research / 2A. Human Studies
Year :
2021
Submission ID :
1825681
Source :
www.aesnet.org
Presentation date :
12/6/2021 12:00:00 PM
Published date :
Nov 22, 2021, 06:44 AM
Authors :
Rachel Stirling, ME - The University of Melbourne; David grayden, PhD – The University of Melbourne; Wendyl D'Souza, MBBS, MPH, PhD – St Vincent's Hospital, The University of Melbourne; Mark Cook, MBBS, MD – St Vincents Hospital, The University of Melbourne; Ewan Nurse, PhD – Seer Medical; Daniel Payne, ME – Seer Medical; Tal Pal Attia, PhD – Mayo Clinics; Pedro Viana, PhD – Kings College London; Mark Richardson, BMBCh, PhD, FRCP – Kings College London; Benjamin brinkmann, PhD – Mayo Clinics; Dean Freestone, PhD – Seer Medical; Philippa Karoly, PhD – The University of Melbourne
Rationale: The unpredictability of epileptic seizures exposes people with epilepsy to potential physical harm, restricts day-to-day activities, and impacts mental well-being. Accurate seizure forecasters would reduce the uncertainty associated with seizures but need to be feasible and accessible in the long-term. Wearable devices are perfect candidates to develop non-invasive, accessible forecasts but are yet to be investigated in long-term studies. We hypothesized that machine learning models could utilize heart rate as a biomarker for well-established cycles of seizures and epileptic activity, in addition to other wearable signals, to forecast high and low risk seizure periods.
Methods: This feasibility study tracked participants’ (n = 11) heart rates, sleep, and step counts using wearable smartwatches and seizure occurrence using mobile seizure diaries for at least 6 months (mean = 14.6 months, SD = 3.8 months). Eligible participants had a diagnosis of refractory epilepsy and reported at least 20 seizures (mean = 135, SD = 123) during the recording period. An ensembled machine learning and neural network model estimated seizure risk either daily or hourly, with retraining occurring on a weekly basis as additional data was collected. Performance was evaluated retrospectively against a rate-matched random forecast using the area under the receiver operating curve. A pseudo-prospective evaluation was also conducted on a held-out dataset.
Results: Of the 11 participants, seizures were predicted above chance in all (100%) participants using an hourly forecast and in ten (91%) participants using a daily forecast (Figure 1). The average time spent in high risk (prediction time) before a seizure occurred was 37 minutes in the hourly forecast and 3 days in the daily forecast. Cyclic features added the most predictive value to the forecasts, particularly circadian and multiday heart rate cycles (Figure 2).
Conclusions: Wearable devices can be used to produce patient-specific seizure forecasts, particularly when biomarkers of seizure and epileptic activity cycles are utilized.
Funding: Please list any funding that was received in support of this abstract.: This study was funded by the Australian government's National Health and Medical Research Council BioMedTech Horizons (BMTH) grant program.
Translational Research