Detecting cardiomyopathies in pregnancy and the postpartum period with an electrocardiogram-based deep learning model
Problem
Cardiomyopathy is among the most common acquired cardiovascular diseases during pregnancy and the postpartum period, and needs better detection tools.
Why it matters
Cardiovascular disease is a major threat to maternal health.
What the AI did
An ECG-based deep learning model was evaluated for detecting cardiomyopathy in pregnant/postpartum women, and compared against natriuretic peptides and a clinical multivariable model.
Results
Cohort of 1807 women; 7%, 10%, and 13% had LVEF ≤35%, <45%, and <50%; AUC/accuracy/sensitivity/specificity values not stated in truncated abstract.
Impact
Peer-reviewed clinical study at Mayo Clinic evaluating AI ECG screening in 1807 pregnant/postpartum women.
Record
| Venue | European Heart Journal - Digital Health |
|---|---|
| Published | 2021 |
| Work type | article |
| Authors | Demilade A. Adedinsewo, Patrick W. Johnson, Erika J. Douglass, Itzhak Zachi Attia, Sabrina Deann Phillips, ROHAN M. GOSWAMI, Mohamad H. Yamani, Heidi M. Connolly et al. |
| Citations | 47 (OpenAlex, at import time) |
| References | 49 |
| Model or method | ECG-based deep learning model |
| Organization | Mayo Clinic |
| Primary source | https://doi.org/10.1093/ehjdh/ztab078 |
| Open access | gold |
| Keywords | ECG, deep learning, cardiomyopathy, pregnancy, postpartum, maternal health |
Importance rating
significant — thousands to millions affected
Abstract states cardiovascular disease is a major threat to maternal health and cardiomyopathy is among the most common acquired cardiovascular diseases in pregnancy/postpartum; better detection aids a significant patient group.