Ultra-long-term subcutaneous EEG (sqEEG) records months of continuous brain activity at home through a minimally invasive subscalp implant, promising objective seizure counts to replace unreliable patient diaries. The foundational real-life study (Weisdorf et al., 2019: 490 days of recordings from nine patients with temporal lobe epilepsy, 338 expert-annotated electrographic seizures) left a six-part open agenda: validation of electrographic seizure detection against an independent gold standard, automated detection algorithms, controlled clinical-outcome trials, artifact reduction, generalizability beyond temporal lobe epilepsy, and per-event validation with time-stamped diaries. Two of the agenda's components are covered by the cited sources: seizure-detection algorithms validated specifically on sqEEG (86-94% sensitivity at roughly 1-6 false detections per day) and cohort generalization to generalized and drug-resistant epilepsies; by contrast intracranial-EEG validation exists only for a different single-channel subgaleal device (98% sensitivity, 99% specificity), no completed controlled outcome trial exists (a single-arm UK interventional trial is registered but unpublished), and no dedicated electromyographic (EMG, muscle-generated electrical) artifact-rejection method for two-channel sqEEG has been validated. The open core is the detection-validity, clinical-benefit, and artifact-bias evidence required before sqEEG can become a standard clinical seizure-counting tool.

Background

Subcutaneous EEG (sqEEG, also called subscalp EEG) is a minimally invasive epilepsy monitoring modality in which a small electrode array is implanted beneath the scalp, outside the skull, and records continuous brain electrical activity for weeks to months while the patient lives at home. The most-studied system, the 24/7 EEG SubQ (UNEEG Medical), uses a subcutaneous implant with one reference and two active leads yielding two bipolar EEG channels (each channel the differential signal between one active lead and the reference) — in the original studies placed over the temporal region, sampled at 207 Hz — plus an external body-worn logging device for power and data transfer. Because the electrodes sit under the skin, the signal is far less affected by electrode detachment and scalp motion than ordinary scalp EEG, and the system avoids the surgical risk of intracranial EEG (electrodes placed inside the skull, such as stereo-EEG depth electrodes), which is the most sensitive but most invasive way to record seizures. Epilepsy care and clinical trials depend on seizure counts, traditionally taken from patient diaries; these are known to be unreliable — many focal seizures with impaired awareness go unreported (underreporting), while non-epileptic events, including psychogenic nonepileptic seizures (PNES, episodes that resemble epileptic seizures but occur without epileptic brain discharges), may be misreported (overreporting). Weisdorf and colleagues (2019) reported the first real-life ultra-long-term test of sqEEG: nine patients with temporal lobe epilepsy wore the SubQ device at home, yielding 490 days of recordings (73% of possible time on average) and 338 expert-annotated electrographic seizures (seizures identified in the EEG signal itself), with no serious device-related adverse events. Objective electrographic counts frequently diverged from diary counts, revealed circadian (about 24-hour) seizure patterns invisible in diaries, and uncovered responses to antiseizure medication that the diaries missed. The authors, however, identified decisive unresolved issues before such monitoring can enter routine practice. First, electrographic detections had never been validated against an independent gold standard — ideally intracranial EEG — leaving open how many seizures are missed because they involve only a small cortical area, have a discreet electrographic presentation, or do not propagate to cortex beneath the implant (small or non-propagating seizures can be undetectable from any extracranial recording, and participants with apparently overreported diary events could not be classified without ambiguity). Second, expert visual review of roughly 12,000 hours of spectrograms (time–frequency displays of the EEG signal) had taken months, and at that time no automated seizure-detection algorithm had been properly validated on sqEEG. Third, no controlled trial had tested whether care guided by objective ultra-long-term EEG improves clinical outcomes, and standards for clinical trials of seizure-detection devices had only just been proposed. Fourth, daytime electromyographic (EMG, muscle-generated electrical) artifact from the nearby temporal and mastoid muscles lowers the signal-to-noise ratio and could bias seizure detection toward the night. Fifth, the cohort was small and restricted to temporal lobe epilepsy, limiting generalizability. Sixth, validation against patient experience was limited because diaries lacked precise time stamps, and PNES — a form of overreporting with, in the authors' words, a great need for new diagnostic tools — cannot be excluded by EEG alone.

Problem Statement

The problem, as posed by the source study, is to complete the evidence base required before ultra-long-term subcutaneous EEG can serve as a validated clinical tool for objective ambulatory seizure counting in epilepsy. Three linked questions remain open. (1) Detection validity against an independent gold standard: what do the single-channel subgaleal validation and the sqEEG algorithm results bound the expected sensitivity and specificity of electrographic seizure detection from the two-channel temporal subcutaneous implant to, and what paired-recording cohort and measurement design would be required to establish those values directly against simultaneous intracranial EEG, in particular for seizures that involve only a small cortical area, have a discreet electrographic presentation, or do not propagate to cortex beneath the implant, which the source study could neither detect nor exclude? (2) Clinical benefit under controlled conditions: no completed controlled trial of sqEEG-guided management exists — what standards and designs should such device trials follow, and how should the registered but unpublished single-arm interventional trial be evaluated for evidence that management informed by objective ultra-long-term sqEEG seizure counts improves clinical outcomes (treatment decisions, seizure-related morbidity, quality of life) relative to standard diary-based care? (3) Artifact control and detection bias: which technical improvements and postprocessing methods reduce or reject artifacts, notably daytime electromyographic contamination from the temporal and mastoid muscles, in two-channel sqEEG, and by what mechanism and in which direction residual artifacts could bias seizure counts and inferred circadian patterns? Two further components of the source agenda — per-event validation using precisely time-stamped digital seizure diaries or multimodal sensors, and disentangling diary underreporting from electrographically undetectable events such as psychogenic nonepileptic seizures — remain partially open and belong to the same objective. Resolution requires addressing all three numbered questions; each is resolved by the strongest account the cited evidence supports — an attainable bound, mechanism, evaluation, or required study design — where a directly measured value or controlled comparison is not yet obtainable; the two further components are in-scope extensions that are not required for a complete resolution. By contrast, the source study's call for validated automated detection algorithms specifically for sqEEG has been substantially met, and generalization beyond temporal/frontotemporal epilepsy has been demonstrated for drug-resistant focal and generalized epilepsies, so these are treated as prior progress rather than open targets; generalization to developmental and epileptic encephalopathies remains open, with the evidence so far at the protocol stage.

References

  1. Weisdorf S, Duun-Henriksen J, Kjeldsen MJ, Poulsen FR, Gangstad SW, Kjær TW. Ultra-long-term subcutaneous home monitoring of epilepsy—490 days of EEG from nine patients. Epilepsia. 2019;60(11):2204-2214. doi:10.1111/epi.16360. https://doi.org/10.1111/epi.16360
  2. Remvig LS, Duun-Henriksen J, Fürbass F, Hartmann M, Viana PF, Kappel Overby AM, Weisdorf S, Richardson MP, Beniczky S, Kjær TW. Detecting temporal lobe seizures in ultra long-term subcutaneous EEG using algorithm-based data reduction. Clinical Neurophysiology. 2022;142:86-93. doi:10.1016/j.clinph.2022.07.504. https://doi.org/10.1016/j.clinph.2022.07.504
  3. Park S, Clark JS, Viana PF, Cui J, Duun-Henriksen J, Mandrekar J, Gregg N, Kremen V, Worrell GA, Richardson MP, Brinkmann BH. Seizure detection using ultra-long-term subcutaneous electroencephalography: A deep learning CNN-BiLSTM approach. Epilepsia. 2026;67(1):437-447. doi:10.1111/epi.18652. https://doi.org/10.1111/epi.18652
  4. Viana PF, Duun-Henriksen J, Biondi A, Winston JS, Freestone DR, Schulze-Bonhage A, Brinkmann BH, Richardson MP. Real-world epilepsy monitoring with ultra-long-term subcutaneous electroencephalography: A 15-month prospective study. Epilepsia. 2025;66(11):4476-4489. doi:10.1111/epi.18566. https://doi.org/10.1111/epi.18566
  5. Rubboli G, Bø MH, Alfstad K, Armand Larsen S, Jacobsen MDH, Vlachou M, Weisdorf S, Rasmussen R, Egge A, Henning O, Lossius M, Beniczky S. Clinical utility of ultra long-term subcutaneous electroencephalographic monitoring in drug-resistant epilepsies: a "real world" pilot study. Epilepsia. 2024;65(11):3265-3278. doi:10.1111/epi.18121. https://doi.org/10.1111/epi.18121
  6. Munteanu T, Daly B, O'Donnell C, Breen A, Choekyi T, Staunton-Grufferty B, O'Brien D, Sweeney K, Kilbride R, Widdess-Walsh P, El-Naggar H, Delanty N. Unilateral ultra long-term subcutaneous EEG monitoring in drug-refractory idiopathic generalized epilepsy. Epilepsia. 2026;67(1):396-407. doi:10.1111/epi.18644. https://doi.org/10.1111/epi.18644
  7. McWilliam M, Viana PF, Dursun E, Davies J, Winston JS, Ågren J, Hasegawa H, Duun-Henriksen J, Richardson MP. Real world testing and cost-effectiveness analysis of subcutaneous EEG (REAL-ASE): Protocol for a prospective multicentre interventional trial. Epilepsia Open. 2025 (online ahead of print). doi:10.1002/epi4.70064. https://doi.org/10.1002/epi4.70064
  8. Viana PF, McWilliam M, Koren J, Duarte C, Lisy T, Baumgartner C, Rössler K, Tomschik M, Celdrán de Castro A, Rocamora R, Vilella L, Rubboli G, Weisdorf S, Friedrichs-Maeder C, Kjær TW, Bhatia S, McAleer D, Hamandi K, Spencer D, Mohanraj R, Richardson MP. Monitoring Antiseizure Medication Change Using Ultra Long-Term Electroencephalogram: A Multicenter Study. Annals of Neurology. 2026;100(3):600-612. doi:10.1002/ana.78305. https://doi.org/10.1002/ana.78305
  9. Pacia SV, Doyle WK, Friedman D, Bacher DH, Kuzniecky RI. Intracranial EEG validation of single-channel subgaleal EEG for seizure identification. Journal of Clinical Neurophysiology. 2022;39(4):283-288. doi:10.1097/WNP.0000000000000774. https://doi.org/10.1097/WNP.0000000000000774
  10. Bacher D, Amini A, Friedman D, Doyle W, Pacia S, Kuzniecky R. Validation of an EEG seizure detection paradigm optimized for clinical use in a chronically implanted subcutaneous device. Journal of Neuroscience Methods. 2021;358:109220. doi:10.1016/j.jneumeth.2021.109220. https://doi.org/10.1016/j.jneumeth.2021.109220
  11. Viana PF, Remvig LS, Duun-Henriksen J, Glasstetter M, Dümpelmann M, Nurse ES, Martins IP, Schulze-Bonhage A, Freestone DR, Brinkmann BH, Kjær TW, Richardson MP. Signal quality and power spectrum analysis of remote ultra long-term subcutaneous EEG. Epilepsia. 2021;62(8):1820-1828. doi:10.1111/epi.16969. https://doi.org/10.1111/epi.16969
  12. Viana PF, Duun-Henriksen J, Glasstëtter M, Dümpelmann M, Nurse ES, Martins IP, Dumanis SB, Schulze-Bonhage A, Freestone DR, Brinkmann BH, Richardson MP. 230 days of ultra long-term subcutaneous EEG: seizure cycle analysis and comparison to patient diary. Annals of Clinical and Translational Neurology. 2021;8(1):288-293. doi:10.1002/acn3.51261. https://doi.org/10.1002/acn3.51261
  13. Dan J, Pale U, Amirshahi A, Cappelletti W, Ingolfsson TM, Wang X, Cossettini A, Bernini A, Benini L, Beniczky S, Atienza D, Ryvlin P. SzCORE: Seizure Community Open-Source Research Evaluation framework for the validation of electroencephalography-based automated seizure detection algorithms. Epilepsia. 2025;66(S3):14-24. doi:10.1111/epi.18113. https://doi.org/10.1111/epi.18113
  14. van Norden SJ, Kho KH, Meppelink AM, Ardesch JJ, Tjepkema-Cloostermans MC, van Putten MJAM. Ultra-long-term subcutaneous EEG recordings in ten epilepsy patients: Experiences and circadian rhythms in epileptiform discharges. Epilepsy & Behavior Reports. 2025;31:100781. doi:10.1016/j.ebr.2025.100781. https://doi.org/10.1016/j.ebr.2025.100781
  15. Schulze-Bonhage A, Richardson MP, Brandt A, Zabler N, Dümpelmann M, San Antonio-Arce V. Cyclical underreporting of seizures in patient-based seizure documentation. Annals of Clinical and Translational Neurology. 2023;10(10):1863-1872. doi:10.1002/acn3.51880. https://doi.org/10.1002/acn3.51880
  16. Biondi A, Zabler N, Kalousios S, Simblett S, Laiou P, Viana PF, Dumpelmann M, Schulze-Bonhage A, Richardson MP. The value of self-reported variables in epilepsy monitoring and management. A systematic scoping review. Seizure. 2024;122:119-143. doi:10.1016/j.seizure.2024.10.004. https://doi.org/10.1016/j.seizure.2024.10.004
  17. Halliday AJ, Gillinder L, Lai A, Seneviratne U, Fontenot H, Cameron T, McLean K, Niemiec A, Raghupathi R, Ganguly TM, Ellis C, Conrad EC, Briggs R, Bulluss K, Kwan P, Perucca P, O'Brien TJ, McGonigal A, Gutman M, Papacostas J, Fong MWK, Lee A, Crompton DE, Laing J, Wijayath M, Morokoff AP, Murphy M, D'Souza WJ, Cook MJ. The UMPIRE study: A first-in-human multicenter trial of bilateral subscalp monitoring for epileptic seizure detection. Epilepsia. 2025;66(9):3426-3439. doi:10.1111/epi.18458. https://doi.org/10.1111/epi.18458
  18. van Maren E, Alnes SL, Ramos da Cruz J, Sobolewski A, Friedrichs-Maeder C, Wohler K, Barlatey SL, Feruglio S, Fuchs M, Vlachos I, Zimmermann J, Bertolote T, Z'Graggen WJ, Tzovara A, Donoghue J, Kouvas G, Schindler K, Pollo C, Baud MO. Feasibility, Safety, and Performance of Full-Head Subscalp EEG Using Minimally Invasive Electrode Implantation. Neurology. 2024;102(12):e209428. doi:10.1212/WNL.0000000000209428. https://doi.org/10.1212/WNL.0000000000209428
  19. Affronte L, Maffei S, Malerba M, Giovannini G, Manganotti P, Coppola A, Bilo L, Vaudano AE, Trivisano M, Specchio N, Meletti S. Ultra long-term EEG monitoring for developmental and epileptic encephalopathies: protocol for a prospective study using subscalp EEG. Clinical Neurophysiology Practice. 2026;11:16-23. doi:10.1016/j.cnp.2025.12.005. https://doi.org/10.1016/j.cnp.2025.12.005