Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 21 min ago
Peer-reviewedCC BYNature

An ECG biomarker for sudden cardiac death discovered with deep learning

  • Ziad Obermeyer1,2
  • Alexander Schubert3
  • James Ross4
  • Sendhil Mullainathan5,6
  • Markus Lingman7,8
8 affiliations
  1. 1Computational Precision Health, College of Computing, Data Science, and Society, University of California, Berkeley, Berkeley, CA, USA. [email protected]
  2. 2School of Public Health, University of California, Berkeley, Berkeley, CA, USA. [email protected]
  3. 3Computational Precision Health, College of Computing, Data Science, and Society, University of California, Berkeley, Berkeley, CA, USA
  4. 4Booth School of Business, University of Chicago, Chicago, IL, USA
  5. 5Department of Economics, MIT, Cambridge, MA, USA
  6. 6Department of Electrical Engineering and Computer Science, MIT, Cambridge, MA, USA
  7. 7Center for Applied Intelligent Systems Research in Health, Halmstad University & Region Halland, Halmstad, Sweden
  8. 8Institute of Medicine, Department of Molecular and Clinical Medicine, University of Gothenburg & Sahlgrenska University Hospital, Gothenburg, Sweden
Open at sourcePDFDOICite

Abstract

Sudden cardiac death is, in theory, preventable with defibrillators. But every year, many patients die without defibrillators because doctors fail to predict their risk 1 . The only predictive biomarker in wide use, cardiac left ventricular ejection fraction (LVEF), misses most sudden cardiac deaths 2 , and flags many low-risk patients for futile defibrillators that never fire 3,4 . Here we apply deep learning to a dataset linking all electrocardiograms (ECGs) in a Swedish region to death certificates. The resulting model isolates a high-risk group (2.2% of the sample) with a 7.0% annual rate of sudden cardiac death, higher than those with reduced LVEF (1.9% of the sample; 4.6% annual rate). Notably, 86.1% of the model’s high-risk patients were not flagged by LVEF. High-risk ECG patients with defibrillators implanted were 54.4% less likely to die than expected, suggesting a mortality benefit. We externally validate the model in a US health system, in which it predicts ventricular arrhythmias that cause sudden death; and a Taiwanese hospital registry, in which it specifically predicts future arrhythmic cardiac arrests. To visualize the waveform morphology ‘discovered’ by the predictive model, we pair it with a generative model of the ECG waveform. Together, they reveal a biomarker that is easily visible and robustly predicts sudden cardiac death, but has not to our knowledge been previously described. Tying the biomarker’s shape to electrophysiological first principles, we form and preliminarily test a new hypothesis on the mechanism of sudden cardiac death.

Related

  1. NKX2-5 mutation in a family with ventricular tachycardia and aborted sudden cardiac death.
    Abinash Pandey, Burt Bromberg, Dipika Menon
    HeartRhythm case reportsPeer-reviewedCC BY-NC-NDFull textCardiovascular medicineEurope PMC
  2. Research trends and hotspot analysis of sudden cardiac death in hypertrophic cardiomyopathy : A bibliometric analysis
    Zhen Chen, Fen Ai, Jinlei Li +1
    MedicinePeer-reviewedCC BYFull textGeneral medicineEurope PMC
  3. Multidimensional evaluation of generative AI chatbots for public consultation on sudden cardiac death: safety, accuracy, empathy, reliability, quality, and readability across six models
    Youyou Chen, Hui Ma, Huimin Wang +3
    BMC Public HealthPeer-reviewedCC BY-NC-ND 4.0Epidemiology & public healthCrossref
  4. Sudden Cardiac Death in Firefighters From the United States: A 16‐Year Postmortem Analysis
    Bradley J. Petek, Andrew J. Layman, Samantha L. Weller +13
    Journal of the American Heart AssociationPeer-reviewedLicence not statedCardiovascular medicineCrossref
  5. Cardiac screening in the young for the detection of cardiovascular disease and prevention of sudden cardiac death
    Daniel T Tardo, Sanjay Sharma
    PLOS MedicinePeer-reviewedCC BYFull textCardiovascular medicinePLOSEurope PMC

Cite

BibTeXRIS
BibTeX
@article{obermeyer2026biomarker,
  title = {{An ECG biomarker for sudden cardiac death discovered with deep learning}},
  author = {Ziad Obermeyer and Alexander Schubert and James Ross and Sendhil Mullainathan and Markus Lingman},
  journal = {Nature},
  year = {2026},
  month = jun,
  doi = {10.1038/s41586-026-10674-6},
  url = {https://doi.org/10.1038/s41586-026-10674-6},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - An ECG biomarker for sudden cardiac death discovered with deep learning
AU  - Ziad Obermeyer
AU  - Alexander Schubert
AU  - James Ross
AU  - Sendhil Mullainathan
AU  - Markus Lingman
PY  - 2026
DA  - 2026/06/24
JO  - Nature
DO  - 10.1038/s41586-026-10674-6
UR  - https://doi.org/10.1038/s41586-026-10674-6
AB  - Sudden cardiac death is, in theory, preventable with defibrillators. But every year, many patients die without defibrillators because doctors fail to predict their risk 1 . The only predictive biomarker in wide use, cardiac left ventricular ejection fraction (LVEF), misses most sudden cardiac deaths 2 , and flags many low-risk patients for futile defibrillators that never fire 3,4 . Here we apply deep learning to a dataset linking all electrocardiograms (ECGs) in a Swedish region to death certificates. The resulting model isolates a high-risk group (2.2% of the sample) with a 7.0% annual rate of sudden cardiac death, higher than those with reduced LVEF (1.9% of the sample; 4.6% annual rate). Notably, 86.1% of the model’s high-risk patients were not flagged by LVEF. High-risk ECG patients with defibrillators implanted were 54.4% less likely to die than expected, suggesting a mortality benefit. We externally validate the model in a US health system, in which it predicts ventricular arrhythmias that cause sudden death; and a Taiwanese hospital registry, in which it specifically predicts future arrhythmic cardiac arrests. To visualize the waveform morphology ‘discovered’ by the predictive model, we pair it with a generative model of the ECG waveform. Together, they reveal a biomarker that is easily visible and robustly predicts sudden cardiac death, but has not to our knowledge been previously described. Tying the biomarker’s shape to electrophysiological first principles, we form and preliminarily test a new hypothesis on the mechanism of sudden cardiac death.
SN  - 0028-0836
SN  - 1476-4687
ER  -