[Abstract + References] Evaluation of an Activity Tracker to Detect Seizures Using Machine Learning

Abstract

Currently, the tracking of seizures is highly subjective, dependent on qualitative information provided by the patient and family instead of quantifiable seizure data. Usage of a seizure detection device to potentially detect seizure events in a population of epilepsy patients has been previously done. Therefore, we chose the Fitbit Charge 2 smart watch to determine if it could detect seizure events in patients when compared to continuous electroencephalographic (EEG) monitoring for those admitted to an epilepsy monitoring unit. A total of 40 patients were enrolled in the study that met the criteria between 2015 and 2016. All seizure types were recorded. Twelve patients had a total of 53 epileptic seizures. The patient-aggregated receiver operating characteristic curve had an area under the curve of 0.58 [0.56, 0.60], indicating that the neural network models were generally able to detect seizure events at an above-chance level. However, the overall low specificity implied a false alarm rate that would likely make the model unsuitable in practice. Overall, the use of the Fitbit Charge 2 activity tracker does not appear well suited in its current form to detect epileptic seizures in patients with seizure activity when compared to data recorded from the continuous EEG.

References

1.	Sander, JW . The epidemiology of epilepsy revisited. Curr Opin Neurol. 2003;16(2):165–170. Google Scholar | Crossref | Medline
2.	Jory, C, Shankar, R, Coker, D, McLean, B, Hanna, J, Newman, C. Safe and sound? A systematic literature review of seizure detection methods for personal use. Seizure. 2016;36:4–15. Google Scholar | Crossref | Medline
3.	Beniczky, S, Jeppesen, J. Non-electroencephalography-based seizure detection. Curr Opin Neurol. 2019;32(2):198–204. Google Scholar | Crossref | Medline
4.	Ryvlin, P, Ciumas, C, Wisniewski, I, Beniczky, S. Wearable devices for sudden unexpected death in epilepsy prevention. Epilepsia. 2018;59(suppl 1):61–66. Google Scholar | Crossref | Medline
5.	Poh, MZ, Loddenkemper, T, Reinsberger, C, et al. Convulsive seizure detection using a wrist-worn electrodermal activity and accelerometry biosensor. Epilepsia. 2012;53(5):e93–e97. Google Scholar | Crossref | Medline
6.	Feehan, LM, Geldman, J, Sayre, EC, et al. Accuracy of Fitbit devices: systematic review and narrative syntheses of quantitative data. JMIR mHealth uHealth. 2018;6(8):e10527. Google Scholar | Crossref | Medline
7.	Fisher, RS, Cross, JH, French, JA, et al. Operational classification of seizure types by the International League Against Epilepsy: position paper of the ILAE Commission for Classification and Terminology. Epilepsia. 2017;58(4):522–530. Google Scholar | Crossref | Medline
8.	Fisher, RS, Cross, JH, D’Souza, C, et al. Instruction manual for the ILAE 2017 operational classification of seizure types. Epilepsia. 2017;58(4):531–542. Google Scholar | Crossref | Medline
9.	Scheffer, IE, Berkovic, S, Capovilla, G, et al. ILAE classification of the epilepsies: position paper of the ILAE Commission for Classification and Terminology. Epilepsia. 2017;58(4):512–521. Google Scholar | Crossref | Medline
10.	Lee, J, Finkelstein, J. Consumer sleep tracking devices: a critical review. Stud Health Technol Inform. 2015;210:458–460. Google Scholar | Medline
11.	Montgomery-Downs, HE, Insana, SP, Bond, JA. Movement toward a novel activity monitoring device. Sleep Breath. 2012;16(3):913–917. Google Scholar | Crossref | Medline | ISI
12.	Moreno-Pino, F, Porras-Segovia, A, Lopez-Esteban, P, Artes, A, Baca-Garcia, E. Validation of Fitbit charge 2 and Fitbit alta HR against polysomnography for assessing sleep in adults with obstructive sleep apnea. J Clin Sleep Med. 2019;15(11):1645–1653. Google Scholar | Crossref | Medline
13.	Regalia, G, Onorati, F, Lai, M, Caborni, C, Picard, RW. Multimodal wrist-worn devices for seizure detection and advancing research: focus on the Empatica wristbands. Epilepsy Res. 2019;153:79–82. Google Scholar | Crossref | Medline
14.	Van Ness, PC . Are seizure detection devices ready for prime time? Epilepsy Curr. 2019;19(1):36–37. Google Scholar | SAGE Journals
15.	Halford, JJ, Sperling, MR, Nair, DR, et al. Detection of generalized tonic-clonic seizures using surface electromyographic monitoring. Epilepsia. 2017;58(11):1861–1869. Google Scholar | Crossref | Medline
16.	Patterson, AL, Mudigoudar, B, Fulton, S, et al. SmartWatch by SmartMonitor: assessment of seizure detection efficacy for various seizure types in children, a large prospective single-center study. Pediatr Neurol. 2015;53(4):309–311. Google Scholar | Crossref | Medline
17.	van Andel, J, Leijten, F, van Delden, H, van Thiel, G. What makes a good home-based nocturnal seizure detector? A value sensitive design. PLoS One. 2015;10(4):e0121446. Google Scholar | Crossref | Medline
18.	Bruno, E, Simblett, S, Lang, A, et al. Wearable technology in epilepsy: the views of patients, caregivers, and healthcare professionals. Epilepsy Behav. 2018;85:141–149. Google Scholar | Crossref | Medline
19.	Patel, AD, Moss, R, Rust, SW, et al. Patient-centered design criteria for wearable seizure detection devices. Epilepsy Behav. 2016;64(pt A):116–121. Google Scholar | Crossref | Medline
20.	Kurada, AV, Srinivasan, T, Hammond, S, Ulate-Campos, A, Bidwell, J. Seizure detection devices for use in antiseizure medication clinical trials: a systematic review. Seizure. 2019;66:61–69. Google Scholar | Crossref | Medline
21.	Benedetto, S, Caldato, C, Bazzan, E, Greenwood, DC, Pensabene, V, Actis, P. Assessment of the Fitbit charge 2 for monitoring heart rate. PLoS One. 2018;13(2):e0192691. Google Scholar | Crossref | Medline
22.	Gutierrez, EG, Crone, NE, Kang, JY, Carmenate, YI, Krauss, GL. Strategies for non-EEG seizure detection and timing for alerting and interventions with tonic-clonic seizures. Epilepsia. 2018;59(suppl 1):36–41. Google Scholar | Crossref | Medline
23.	Beniczky, S, Ryvlin, P. Standards for testing and clinical validation of seizure detection devices. Epilepsia. 2018;59(suppl 1):9–13. Google Scholar | Crossref | Medline