Cited by 146Epilepsy research and treatmentEEG and Brain-Computer InterfacesContext-Aware Activity Recognition Systems

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

Open access
Tang, Jianbin, El Atrache, Rima, Yu, Shuang, Asif, Umar, Jackson, Michele, Roy, Subhrajit, Mirmomeni, Mahtab, Cantley, Sarah +7 more·2021·Epilepsia

Abstract

Objective: Tracking 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. Methods: We 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. Results: We 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. Significance: Automatic 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.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1111/epi.16967

Cites

48 of 48 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Any longer-term data sets from wearable devices, especially if recorded in ambulatory settings, will contain periods where the device was recording, but not actually on the subject's body in the correct manner.

    “Any longer-term data sets from wearable devices, especially if recorded in ambulatory settings, will contain periods where the device was recording, but not actually on the subject’s body in the correct manner, as also noted in previous studies21,30.”

    From Data quality evaluation in wearable monitoring (2022) · cites this paper as [21]

  • The device memory mode is very reliable and data loss only occurs whenever the device is switched off to swap it or when using the shower.

    “The device memory mode is very reliable and data loss only occurs whenever the device is switched off to swap it or when using the shower, as seen in previous studies20,21,54.”

    From Data quality evaluation in wearable monitoring (2022) · cites this paper as [21]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

1 of the 49 references couldn’t be matched to a known paper.

Research works printed but not linked: 1

  1. [21]Ulate‐Campos A, Coughlin F, Gainza‐Lein M, Fernández IS, Pearl Pl, Loddenkemper T. Automated seizure detection systems and their effectiveness for each type of seizure. Seizure. 2016; 40: 88– 101. 27376911

Cited by

Sorted by citations

Authors
Tang, J., El Atrache, R., Yu, S., Asif, U., Jackson, M., Roy, S., Mirmomeni, M., Cantley, S., Sheehan, T., Schubach, S., Ufongene, C., Vieluf, S., Meisel, C., Harrer, S., & Loddenkemper, T.
Affiliations
  • IBM Research - Australia (AU)
  • Boston Children's Hospital and Harvard Medical School Boston Massachusetts USA
  • Boston Children's Hospital (US)
  • Google (United Kingdom) (GB)
  • Berlin Institute of Health at Charité - Universitätsmedizin Berlin (DE)
Corresponding author
Tang, Jianbin
Venue
Epilepsia
Publisher
Wiley
Published
2021-07-15
Volume
62
Issue
8
Pages
1807-1819
ISSN
  • 0013-9580 (ISSN-L)
  • 1528-1167
OpenAlex
W3177992706
Cited by
146
References
49
Citations by year
2026: 24 · 2025: 33 · 2024: 35 · 2023: 24 · 2022: 24 · 2021: 6
FWCI
15.95
Type
journal article
Language
en
Open access
Locations
  • Epilepsia (publishedVersion, cc-by-nc)
  • Europe PMC (PubMed Central) (publishedVersion, other-oa)
  • PubMed Central (publishedVersion, cc-by-nc)
  • PubMed (publishedVersion)
Topics
  • Health Sciences › Medicine › Psychiatry and Mental health › Epilepsy research and treatment
  • Life Sciences › Neuroscience › Cognitive Neuroscience › EEG and Brain-Computer Interfaces
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
Keywords
seizure detection, epilepsy, wearable sensors, electrodermal activity, accelerometry, photoplethysmography, body temperature, seizure type classification, seizure monitoring, epilepsy monitoring unit, machine learning, multimodal sensor fusion
MeSH terms
Machine Learning, Wearable Electronic Devices, Child, Electroencephalography, Epilepsy, Female, Humans, Male, Seizures, Benchmarking

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review