Seizure detection using wearable sensors and machine learning: Setting a benchmark

AbstractObjectiveTracking seizures is crucial for epilepsy monitoring and treatment evaluation. Current epilepsy care relies on caretaker seizure diaries, but clinical seizure monitoring may miss seizures. Wearable devices may be better tolerated and more suitable for long‐term ambulatory monitoring. This study evaluates the seizure detection performance of custom‐developed machine learning (ML) algorithms across a broad spectrum of epileptic seizures utilizing wrist‐ and ankle‐worn multisignal biosensors.MethodsWe enrolled patients admitted to the epilepsy monitoring unit and asked them to wear a wearable sensor on either their wrists or ankles. The sensor recorded body temperature, electrodermal activity, accelerometry (ACC), and photoplethysmography, which provides blood volume pulse (BVP). We used electroencephalographic seizure onset and offset as determined by a board‐certified epileptologist as a standard comparison. We trained and validated ML for two different algorithms: Algorithm 1, ML methods for developing seizure type‐specific detection models for nine individual seizure types; and Algorithm 2, ML methods for building general seizure type‐agnostic detection, lumping together all seizure types.ResultsWe included 94 patients (57.4% female, median age = 9.9 years) and 548 epileptic seizures (11 066 h of sensor data) for a total of 930 seizures and nine seizure types. Algorithm 1 detected eight of nine seizure types better than chance (area under the receiver operating characteristic curve [AUC‐ROC] = .648–.976). Algorithm 2 detected all nine seizure types better than chance (AUC‐ROC = .642–.995); a fusion of ACC and BVP modalities achieved the best AUC‐ROC (.752) when combining all seizure types together.SignificanceAutomatic seizure detection using ML from multimodal wearable sensor data is feasible across a broad spectrum of epileptic seizures. Preliminary results show better than chance seizure detection. The next steps include validation of our results in larger datasets, evaluation of the detection utility tool for additional clinical seizure types, and integration of additional clinical information.

Topics

No keywords indexed for this article. Browse by subject →

References

49

[1]

The Epidemiology of Epilepsy

Ettore Beghi

Neuroepidemiology 10.1159/000503831

[2]

10.1097/00019052-200304000-00008

[3]

10.1016/s1474-4422(18)30038-3

[4]

10.1001/archneur.64.11.1595

[5]

Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study

Mark J Cook, Terence J O'Brien, Samuel F Berkovic et al.

The Lancet Neurology 10.1016/s1474-4422(13)70075-9

[6]

10.1016/j.yebeh.2016.10.036

[7]

10.1111/epi.13938

[8]

Automated seizure detection systems and their effectiveness for each type of seizure

A. Ulate-Campos, F. Coughlin, M. Gaínza-Lein et al.

Seizure 10.1016/j.seizure.2016.06.008

[9]

10.1007/s13760-020-01417-z

[10]

Multimodal seizure detection: A review

Frans S. S. Leijten

Epilepsia 10.1111/epi.14047

[11]

10.1016/j.yebeh.2014.06.023

[12]

GarbarinoM LaiM BenderD PicardRW TognettiS.Empatica E3—A wearable wireless multi‐sensor device for real‐time computerized biofeedback and data acquisition. In 2014 4th International Conference on Wireless Mobile Communication and Healthcare‐Transforming Healthcare Through Innovations in Mobile and Wireless Technologies (MOBIHEALTH).IEEE.2014; p.39–42. 10.4108/icst.mobihealth.2014.257418

[13]

10.1109/tbme.2009.2038487

[14]

10.1111/epi.12120

[15]

10.1016/j.yebeh.2011.01.019

[16]

10.7309/jmtm.19

[17]

10.1212/wnl.0000000000004893

[18]

10.1109/tbme.2007.895114

[19]

10.1016/j.yebeh.2013.04.017

[20]

10.1016/j.yebeh.2018.05.044

[21]

Automated seizure detection systems and their effectiveness for each type of seizure

A. Ulate-Campos, F. Coughlin, M. Gaínza-Lein et al.

Seizure 10.1016/j.seizure.2016.06.008

[22]

10.1016/j.yebeh.2019.106717

[23]

10.1111/epi.16555

[24]

10.1038/s41746-020-0264-0

[25]

AsifU RoyS TangJ HarrerS.SeizureNet: Multi‐spectral deep feature learning for seizure type classification. In:Machine learning in clinical neuroimaging and radiogenomics in neuro‐oncology.Cham:Springer;2020; p.77‐87. 10.1007/978-3-030-66843-3_8

[26]

10.1111/epi.13899

[27]

10.1111/epi.13671

[28]

10.1097/wnp.0000000000000316

[29]

LinM ChenQ YanS.Network in network. arXiv preprint. 2013; arXiv:1312.4400.

[30]

10.1111/j.1528-1167.2012.03444.x

[31]

Heldberg BE "Using wearable sensors for semiology‐independent seizure detection—towards ambulatory monitoring of epilepsy" Conf Proc IEEE Eng Med Biol Soc (2015)

[32]

Epileptic Seizure Prediction Using Big Data and Deep Learning: Toward a Mobile System

Isabell Kiral-Kornek, Subhrajit Roy, Ewan Nurse et al.

EBioMedicine 10.1016/j.ebiom.2017.11.032

[33]

LoeckxD BuckinxT.In‐field validation of epihunter for the detection of absence seizures using wearable EEG. Paper presented at 2nd International Congress on Mobile Devices and Seizure Detection in Epilepsy. Lausanne Switzerland; September 6–7 2019.

[34]

10.3390/s17102338

[35]

Voelker R "Smart watch detects seizures" JAMA (2018) 10.1001/jama.2018.1809

[36]

"Drug and device news" P T (2019)

[37]

Artificial Intelligence for Clinical Trial Design

Stefan Harrer, Pratik Shah, Bhavna Antony et al.

Trends in Pharmacological Sciences 10.1016/j.tips.2019.05.005

[38]

Photoplethysmography and its application in clinical physiological measurement

John Allen

Physiological Measurement 10.1088/0967-3334/28/3/r01

[39]

10.1111/epi.16621

[40]

10.1111/epi.16590

[41]

10.1212/wnl.0b013e318258f7f1

[42]

10.1111/j.1535-7597.2004.42001.x

[43]

10.1016/j.yebeh.2019.02.018

[44]

10.1038/s41598-020-68434-z

[45]

10.1016/j.seizure.2016.07.012

[46]

10.1007/s10286-018-0568-1

[47]

10.1016/j.seizure.2018.09.013

[48]

10.1016/j.yebeh.2019.07.028

[49]

10.1111/epi.14046

Cited By

129

Improving wearable-based seizure prediction by feature fusion using an explainable growing network

Tanuj Hasija, Maurice Kuschel · 2025

Artificial Intelligence in Medicine

Innovating pediatric epilepsy: transforming diagnosis and treatment with AI

Giovanni Battista Dell’Isola, Antonella Fattorusso · 2025

World Journal of Pediatrics

Artificial intelligence‐enhanced epileptic seizure detection by wearables

Shuang Yu, Rima El Atrache · 2023

Epilepsia

Explainable AI for wearable seizure logging: Impact of data quality, patient age, and antiseizure medication on performance

Mustafa Halimeh, Michele Jackson · 2023

Seizure

Skin-interfaced electronics: A promising and intelligent paradigm for personalized healthcare

Yangzhi Zhu, Jinghang Li · 2023

Biomaterials

Seizure-related differences in biosignal 24-h modulation patterns

Solveig Vieluf, Rima El Atrache · 2022

Scientific Reports

Metrics

129

Citations

49

References

Details

Published: Jul 15, 2021
Vol/Issue: 62(8)
Pages: 1807-1819
License: View

Authors

J

Jianbin Tang

IBM Research Australia Melbourne Victoria Australia

R

Rima El Atrache

Boston Children's Hospital and Harvard Medical School Boston Massachusetts USA

S

Shuang Yu

IBM Research Australia Melbourne Victoria Australia

U

Umar Asif

IBM Research Australia Melbourne Victoria Australia

M

Michele Jackson

Boston Children's Hospital and Harvard Medical School Boston Massachusetts USA

S

Subhrajit Roy

IBM Research Australia Melbourne Victoria Australia; Google Brain London UK

M

Mahtab Mirmomeni

IBM Research Australia Melbourne Victoria Australia

S

Sarah Cantley

Boston Children's Hospital and Harvard Medical School Boston Massachusetts USA

T

Theodore Sheehan

Boston Children's Hospital and Harvard Medical School Boston Massachusetts USA

S

Sarah Schubach

Boston Children's Hospital and Harvard Medical School Boston Massachusetts USA

C

Claire Ufongene

Boston Children's Hospital and Harvard Medical School Boston Massachusetts USA

S

Solveig Vieluf

Boston Children's Hospital and Harvard Medical School Boston Massachusetts USA

C

Christian Meisel

Department of Neurology Charité‐Universitätsmedizin Berlin Berlin Germany; Berlin Institute of Health Berlin Germany

S

Stefan Harrer

IBM Research Australia Melbourne Victoria Australia; Digital Health Cooperative Research Centre Melbourne Victoria Australia

T

Tobias Loddenkemper

Boston Children's Hospital and Harvard Medical School Boston Massachusetts USA

Cite This Article

Jianbin Tang, Rima El Atrache, Shuang Yu, et al. (2021). Seizure detection using wearable sensors and machine learning: Setting a benchmark. Epilepsia, 62(8), 1807-1819. https://doi.org/10.1111/epi.16967

Seizure detection using wearable sensors and machine learning: Setting a benchmark

You May Also Like