Skip to content
DraconiticResearch
Search Sources API
210,049 studies31 sourcesUpdated 13 min ago
Peer-reviewedCC BY-NC-NDNature Medicine

Human–artificial intelligence collaboration for comanaging epilepsy in real time

  • Zack Goldblum1,2,3
  • Haoer Shi4,5
  • Zhongchuan Xu4,5
  • William K.S. Ojemann4,5
  • Carlos A. Aguila4,5
  • Kevin W. Long6
  • Kevin Xie7,4,8
  • Kerry C. Nix4,9
  • Katherine G. Walsh4,5
  • Ellie Chang4,5
  • Sarah B. Lavelle4
  • Brandon Bach4
5 more authors
  • Kathryn A. Davis4,8
  • Nishant Sinha4,10,8
  • Lauren H. Hammer4,5,8
  • Erin C. Conrad4,8
  • Brian Litt11,12,13,14
14 affiliations
  1. 1Center for Health, Devices and Technology, University of Pennsylvania, Philadelphia, PA, USA. [email protected]
  2. 2Center for Neuroengineering and Therapeutics, University of Pennsylvania, Philadelphia, PA, USA. [email protected]
  3. 3Department of Bioengineering, School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, PA, USA. [email protected]
  4. 4Center for Neuroengineering and Therapeutics, University of Pennsylvania, Philadelphia, PA, USA
  5. 5Department of Bioengineering, School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, PA, USA
  6. 6Penn Medicine, University of Pennsylvania, Philadelphia, PA, USA
  7. 7Center for Healthcare Transformation and Innovation, University of Pennsylvania, Philadelphia, PA, USA
  8. 8Department of Neurology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
  9. 9Department of Neuroscience, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
  10. 10Department of Biostatistics, Epidemiology and Informatics, University of Pennsylvania, Philadelphia, PA, USA
  11. 11Center for Health, Devices and Technology, University of Pennsylvania, Philadelphia, PA, USA. [email protected]
  12. 12Center for Neuroengineering and Therapeutics, University of Pennsylvania, Philadelphia, PA, USA. [email protected]
  13. 13Department of Bioengineering, School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, PA, USA. [email protected]
  14. 14Department of Neurology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. [email protected]
Open at sourcePDFDOICite

Abstract

One-third of the world's 70 million people with epilepsy have seizures that are not controlled by medication; and implantable devices are an exciting option for treatment. These devices improve seizure control and can detect impending attacks, missed medication, and impaired cognition. Unfortunately, they have no way to share this information with their hosts in real-time - a limitation common to most medical devices. This is a missed opportunity for implants and wearables to learn from patients, focus on what matters most to them, and teach them how their behavior affects their health. Here, we present a device platform that converses with patients and learns to co-manage epilepsy. The inpatient prototype links scalp and intracranial EEG (electroencephalograms) to secure large language models that communicate freely and bidirectionally with their hosts through a smartphone app. An AI agent ingests biomarkers of sleep, medication level, cognition, and seizure risk extracted from brain activity. It converses with patients to inform them of clinical events and physiological trends, records their symptoms, responses, and behaviors, and automatically retrains itself to improve performance. Both patients and the AI agent can initiate conversations to teach each other and personalize interactions. We demonstrate this platform in 13 patients undergoing inpatient video-EEG monitoring for epilepsy and validate its performance. Algorithms for detecting seizures optimized their precision over several days without expert intervention - in contrast to the months of iterative, in-person physician programming currently required. Patients responded positively to messages regarding sleep, cognition, and seizure risk while rating the system as highly usable. The platform includes several safeguards, including a system for further algorithm fine-tuning using efficient expert review, and features that ensure data security and regulate communication content. Further work will link other biosensors to measure behavior, improve performance, and optimize therapeutic stimulation. We propose this system as a scalable platform for medical devices that can rapidly adapt to patient and provider needs; one that is broadly adaptable to improving care for many medical conditions.

Related

  1. The double-edged sword of human-AI collaboration on design students’ competence development: The mediating roles of challenge and hindrance appraisals and the asymmetric moderation of teacher AI leadership
    Xiang Wang, MingXing Li, Jia Du +2
    PLOS OnePeer-reviewedCC BY 4.0Full textNeuroscienceStatisticsPLOS
  2. Toward Enactive Artificial Intelligence
    Banafsheh Rafiee, R Sutton
    arXiv (Cornell University)PreprintLicence not statedFull text6 citedNeuroscienceOpenAlex
  3. Growing up with artificial intelligence: An amplifier of inequalities and a public health issue
    Théo Mouhoud, Thierry Baubet
    L'EncéphalePeer-reviewedPublisher termsFull textGeneral biologyPsychiatry & mental healthHAL
  4. Notification: Artificial Intelligence in Autism Spectrum Disorder Diagnosis: A Scoping Review of Face, Voice, and Text Analysis Methods.
    Health science reportsPeer-reviewedCC BYFull textNeuroscienceEurope PMCPMC
  5. Early Detection of Red Palm Weevil Infestation: A Systematic Review of Sensing Technologies, Artificial Intelligence, and Field Readiness
    Lubna A. Gabralla, Amnah Alshahrani
    InsectsPeer-reviewedCC BYFull textGeneral biologyEurope PMC

Cite

BibTeXRIS
BibTeX
@article{goldblum2026humanartificial,
  title = {{Human–artificial intelligence collaboration for comanaging epilepsy in real time}},
  author = {Zack Goldblum and Haoer Shi and Zhongchuan Xu and William K.S. Ojemann and Carlos A. Aguila and Kevin W. Long and Kevin Xie and Kerry C. Nix and Katherine G. Walsh and Ellie Chang and Sarah B. Lavelle and Brandon Bach and Kathryn A. Davis and Nishant Sinha and Lauren H. Hammer and Erin C. Conrad and Brian Litt},
  journal = {Nature Medicine},
  year = {2026},
  month = oct,
  doi = {10.1038/s41591-026-04680-4},
  url = {https://doi.org/10.1038/s41591-026-04680-4},
  note = {CC BY-NC-ND},
}
RIS
TY  - JOUR
TI  - Human–artificial intelligence collaboration for comanaging epilepsy in real time
AU  - Zack Goldblum
AU  - Haoer Shi
AU  - Zhongchuan Xu
AU  - William K.S. Ojemann
AU  - Carlos A. Aguila
AU  - Kevin W. Long
AU  - Kevin Xie
AU  - Kerry C. Nix
AU  - Katherine G. Walsh
AU  - Ellie Chang
AU  - Sarah B. Lavelle
AU  - Brandon Bach
AU  - Kathryn A. Davis
AU  - Nishant Sinha
AU  - Lauren H. Hammer
AU  - Erin C. Conrad
AU  - Brian Litt
PY  - 2026
DA  - 2026/10/13
JO  - Nature Medicine
DO  - 10.1038/s41591-026-04680-4
UR  - https://doi.org/10.1038/s41591-026-04680-4
AB  - One-third of the world's 70 million people with epilepsy have seizures that are not controlled by medication; and implantable devices are an exciting option for treatment. These devices improve seizure control and can detect impending attacks, missed medication, and impaired cognition. Unfortunately, they have no way to share this information with their hosts in real-time - a limitation common to most medical devices. This is a missed opportunity for implants and wearables to learn from patients, focus on what matters most to them, and teach them how their behavior affects their health. Here, we present a device platform that converses with patients and learns to co-manage epilepsy. The inpatient prototype links scalp and intracranial EEG (electroencephalograms) to secure large language models that communicate freely and bidirectionally with their hosts through a smartphone app. An AI agent ingests biomarkers of sleep, medication level, cognition, and seizure risk extracted from brain activity. It converses with patients to inform them of clinical events and physiological trends, records their symptoms, responses, and behaviors, and automatically retrains itself to improve performance. Both patients and the AI agent can initiate conversations to teach each other and personalize interactions. We demonstrate this platform in 13 patients undergoing inpatient video-EEG monitoring for epilepsy and validate its performance. Algorithms for detecting seizures optimized their precision over several days without expert intervention - in contrast to the months of iterative, in-person physician programming currently required. Patients responded positively to messages regarding sleep, cognition, and seizure risk while rating the system as highly usable. The platform includes several safeguards, including a system for further algorithm fine-tuning using efficient expert review, and features that ensure data security and regulate communication content. Further work will link other biosensors to measure behavior, improve performance, and optimize therapeutic stimulation. We propose this system as a scalable platform for medical devices that can rapidly adapt to patient and provider needs; one that is broadly adaptable to improving care for many medical conditions.
SN  - 1078-8956
SN  - 1546-170X
SN  - 1744-7933
ER  -