Peer-reviewedCC BYJACC. Advances
Validating ECG-Based Artificial Intelligence Models for Cardiomyopathy Detection in Obstetric Patients: A Multisite International Study.
30 affiliations
- 1Department of Cardiovascular Medicine, Mayo Clinic, Jacksonville, Florida, USA. Electronic address: [email protected].
- 2Department of Quantitative Health Sciences, Mayo Clinic, Jacksonville, Florida, USA.
- 3Department of Radiology, Emory University School of Medicine, Atlanta, Georgia, USA.
- 4Department of Cardiovascular Medicine, Mayo Clinic, Jacksonville, Florida, USA.
- 5Leviev Heart Center, Sheba Medical Center, Ramat Gan, Israel; ARC Innovation Center, Sagol AI Hub, Sheba Medical Center, Ramat Gan, Israel.
- 6Leviev Heart Center, Sheba Medical Center, Ramat Gan, Israel; ARC Innovation Center, Sagol AI Hub, Sheba Medical Center, Ramat Gan, Israel; Gray Faculty of Medical and Health Sciences, Tel Aviv University, Ramat Aviv, Tel Aviv, Israel.
- 7Leviev Heart Center, Sheba Medical Center, Ramat Gan, Israel; Department of Military Medicine, Faculty of Medicine, The Hebrew University of Jerusalem, Jerusalem, Israel.
- 8Department of Obstetrics and Gynecology, The Sheba Medical Center, Tel Hashomer, Israel; The Dina Recanati School of Medicine, Reichman University, Herzliya, Israel.
- 9ARC Innovation Center, Sagol AI Hub, Sheba Medical Center, Ramat Gan, Israel; Department of Obstetrics and Gynecology, The Sheba Medical Center, Tel Hashomer, Israel.
- 10Department of Medicine, Aminu Kanu Teaching Hospital and Bayero University, Kano, Nigeria.
- 11Department of Obstetrics and Gynecology, College of Medicine, University of Lagos, Idi-Araba, Lagos, Nigeria; Centre for Clinical Trials, Research and Implementation Science, University of Lagos College of Medicine, Idi-Araba, Lagos, Nigeria.
- 12Department of Obstetrics and Gynecology, Olabisi Onabanjo University Teaching Hospital, Sagamu, Ogun, Nigeria.
- 13Department of Medicine, Rasheed Shekoni Federal University Teaching Hospital, Dutse, Jigawa, Nigeria.
- 14Department of Obstetrics and Gynecology, University of Ilorin Teaching Hospital, Ilorin, Nigeria.
- 15Cardiology Unit, Department of Medicine, University of Ibadan, Ibadan, Oyo, Nigeria.
- 16Department of Adult Cardiology, National Institute of Cardiovascular Diseases (NICVD), Karachi, Pakistan; Department of Cardiology, Rutgers Robert Wood Johnson University Hospital, New Brunswick, New Jersey, USA.
- 17Department of Adult Cardiology, National Institute of Cardiovascular Diseases (NICVD), Karachi, Pakistan.
- 18Department of Obstetrics and Gynecology, Jinnah Postgraduate Medical Centre (JPMC), Jinnah Sindh Medical University, Karachi, Pakistan.
- 19Baptist Beaches Cardiology Center, Baptist Health, Jacksonville, Florida, USA.
- 20Department of Transplant, Mayo Clinic, Jacksonville, Florida, USA.
- 21Department of Cardiology, University of Florida College of Medicine, Jacksonville, Florida, USA.
- 22Department of Obstetrics and Gynecology, University of Florida College of Medicine, Jacksonville, Florida, USA.
- 23Department of Cardiology, Morehouse School of Medicine, Atlanta, Georgia, USA.
- 24Department of Cardiology, West Georgia Cardiology, Carrollton, Georgia, USA.
- 25Department of Obstetrics and Gynecology, Morehouse School of Medicine, Atlanta, Georgia, USA; Center for Maternal Health Equity, Department of Community Health and Preventive Medicine, Morehouse School of Medicine, Atlanta, Georgia, USA.
- 26Department of Cardiology, Essentia Health Heart and Vascular Center, Fargo, North Dakota, USA.
- 27Department of Obstetrics and Gynecology, Essentia Health, Fargo, North Dakota, USA.
- 28Anumana Inc, Cambridge, Massachusetts, USA; Nference Inc, Cambridge, Massachusetts, USA.
- 29Department of Obstetrics and Gynecology, Mayo Clinic, Rochester, Minnesota, USA.
- 30Department of Cardiology, Mayo Clinic, Rochester, Minnesota, USA.
Abstract
Background
Cardiomyopathy significantly impacts maternal health worldwide. Artificial intelligence-enabled electrocardiograms (AI-ECG) have shown effectiveness in detecting left ventricular systolic dysfunction (LVSD) in adults and limited obstetric cohorts; however, broad validation in diverse obstetric populations remains unexplored.Objectives
Our primary objective was to validate the performance of 2 versions of an AI-ECG model for detecting cardiomyopathy in an international obstetric cohort.Methods
We conducted a retrospective, multisite study of pregnant or postpartum women (≤12 months) across 11 institutions in 4 countries (Israel, Nigeria, Pakistan, and the United States). Two AI-ECG models (Food and Drug Administration-cleared v2.5.0 and refined v3.0.0) were evaluated against echocardiography. LVSD was defined as left ventricular ejection fraction ≤40% in keeping with the model's training target. Model performance was assessed by area under the receiver-operating characteristic curve (AUROC).Results
We included 1,820 women with paired ECG-echocardiogram data, of whom 7.3% had LVSD. Model 1 (v2.5.0) demonstrated strong discriminative performance with an AUROC of 0.865, sensitivity of 46.1%, and negative predictive value of 95.8%. Model 2 (v3.0.0) showed significantly improved performance, with an AUROC of 0.925, sensitivity of 64.1%, and negative predictive value of 97.2% (P < 0.001) for LVSD detection.Conclusions
AI-ECG models demonstrated strong external validity in a global obstetric cohort, with significantly improved performance in the refined model trained on expanded data. These findings support the potential role of AI-ECG for a scalable LVSD screening tool in obstetric populations, while underscoring the importance of population-specific validation prior to clinical implementation.
Related
- Artificial Intelligence-Enhanced Electrocardiography for Detection and Prediction of Hypertrophic Cardiomyopathy across Monogenic and Polygenic Susceptibility
- Validating Artificial Intelligence Guidance for Ultrasound Acquisition and Remote Interpretation
- Artificial Intelligence-Enhanced Cardiac Magnetic Resonance in Hypertrophic Cardiomyopathy: Diagnosis, Phenotypic Characterization, and Risk Prediction
- Hypertrophic cardiomyopathy phenotype is associated with postmyectomy change in diastolic function measured by artificial intelligence-enabled electrocardiogram.
- Attitudes and perceptions toward the use of artificial intelligence chatbots for peer review in medical journals: a large-scale, international cross-sectional survey.
BibTeX
@article{adedinsewo2026validating,
title = {{Validating ECG-Based Artificial Intelligence Models for Cardiomyopathy Detection in Obstetric Patients: A Multisite International Study.}},
author = {Demilade Adedinsewo and Patrick W Johnson and Carolina Morales-Lara and Maria Vitoria Coutinho and Sabrina D Phillips and Marc A Eidelhoch and Michal Cohen-Shelly and Elad Maor and Ariel Furer and Abraham Tsur and Shali Mazaki-Tovi and Kamilu M Karaye and Bosede Bukola Afolabi and Olusoji Edward Jagun and Sadiq Hassan Ringim and Kehinde Femi Ibiyemi and Okechukwu Samuel Ogah and Sabha Bhatti and Shakeela Naz and Haleema Yasmin and Maedeh Ganji and Jose Ruiz and Khadeeja Esmail and Jason Wheatley and Anekwe Onwuanyi and Eric Chang and Myah M Griffin and Oludamilola Oluleye and Stefanie Gefroh and Mohan Krishna Ranganathan and Samir Awasthi and Yvonne Butler Tobah and Francisco Lopez-Jimenez and Zachi I Attia and Paul A Friedman and Peter A Noseworthy and Rickey E Carter},
journal = {JACC. Advances},
year = {2026},
month = sep,
doi = {10.1016/j.jacadv.2026.103321},
url = {https://europepmc.org/article/MED/42863355},
note = {CC BY},
}
RIS
TY - JOUR
TI - Validating ECG-Based Artificial Intelligence Models for Cardiomyopathy Detection in Obstetric Patients: A Multisite International Study.
AU - Demilade Adedinsewo
AU - Patrick W Johnson
AU - Carolina Morales-Lara
AU - Maria Vitoria Coutinho
AU - Sabrina D Phillips
AU - Marc A Eidelhoch
AU - Michal Cohen-Shelly
AU - Elad Maor
AU - Ariel Furer
AU - Abraham Tsur
AU - Shali Mazaki-Tovi
AU - Kamilu M Karaye
AU - Bosede Bukola Afolabi
AU - Olusoji Edward Jagun
AU - Sadiq Hassan Ringim
AU - Kehinde Femi Ibiyemi
AU - Okechukwu Samuel Ogah
AU - Sabha Bhatti
AU - Shakeela Naz
AU - Haleema Yasmin
AU - Maedeh Ganji
AU - Jose Ruiz
AU - Khadeeja Esmail
AU - Jason Wheatley
AU - Anekwe Onwuanyi
AU - Eric Chang
AU - Myah M Griffin
AU - Oludamilola Oluleye
AU - Stefanie Gefroh
AU - Mohan Krishna Ranganathan
AU - Samir Awasthi
AU - Yvonne Butler Tobah
AU - Francisco Lopez-Jimenez
AU - Zachi I Attia
AU - Paul A Friedman
AU - Peter A Noseworthy
AU - Rickey E Carter
PY - 2026
DA - 2026/09/22
JO - JACC. Advances
DO - 10.1016/j.jacadv.2026.103321
UR - https://europepmc.org/article/MED/42863355
AB - Background Cardiomyopathy significantly impacts maternal health worldwide. Artificial intelligence-enabled electrocardiograms (AI-ECG) have shown effectiveness in detecting left ventricular systolic dysfunction (LVSD) in adults and limited obstetric cohorts; however, broad validation in diverse obstetric populations remains unexplored.Objectives Our primary objective was to validate the performance of 2 versions of an AI-ECG model for detecting cardiomyopathy in an international obstetric cohort.Methods We conducted a retrospective, multisite study of pregnant or postpartum women (≤12 months) across 11 institutions in 4 countries (Israel, Nigeria, Pakistan, and the United States). Two AI-ECG models (Food and Drug Administration-cleared v2.5.0 and refined v3.0.0) were evaluated against echocardiography. LVSD was defined as left ventricular ejection fraction ≤40% in keeping with the model's training target. Model performance was assessed by area under the receiver-operating characteristic curve (AUROC).Results We included 1,820 women with paired ECG-echocardiogram data, of whom 7.3% had LVSD. Model 1 (v2.5.0) demonstrated strong discriminative performance with an AUROC of 0.865, sensitivity of 46.1%, and negative predictive value of 95.8%. Model 2 (v3.0.0) showed significantly improved performance, with an AUROC of 0.925, sensitivity of 64.1%, and negative predictive value of 97.2% (P < 0.001) for LVSD detection.Conclusions AI-ECG models demonstrated strong external validity in a global obstetric cohort, with significantly improved performance in the refined model trained on expanded data. These findings support the potential role of AI-ECG for a scalable LVSD screening tool in obstetric populations, while underscoring the importance of population-specific validation prior to clinical implementation.
SN - 2772-963X
ER -