Echocardiography-Based Deep Learning Model to Differentiate Constrictive Pericarditis and Restrictive Cardiomyopathy
Authors' conclusion
With a standard apical 4-chamber view, our artificial intelligence model provides a platform to facilitate the detection of CP, allowing for improved workflow efficiency and prompt referral for more advanced evaluation and intervention of CP.
Study context
- Population
- Patients with confirmed constrictive pericarditis (CP) or cardiac amyloidosis (CA, representative of restrictive cardiomyopathy), Mayo Clinic Rochester, January 2003 to December 2021184 CP, 197 CA
- Participants
- 381 patients
- Comparison
- ResNet50 deep learning model on TTE apical 4-chamber view to differentiate CP vs CA
- Outcome
- AUC for differentiating CP and CA in the held-out test set
Methodological detailssingle-centre, retrospective, 60:20:20 training/validation/held-out test split
- Single-centre
- Retrospective
- 60:20:20 training/validation/held-out test split
Performance
As reported in the abstract
Differentiation of CP vs CA (test set)
| Test or model | AUC |
|---|---|
| ResNet50 | 0.97 |
ResNet50
AUC0.97
Further findings
GradCAM model interpretation
- GradCAM heatmap showed activation around the ventricular septal area
Parts of this Brief are withheld: the automatic checks could not confirm them against the abstract.
Original publication
Echocardiography-Based Deep Learning Model to Differentiate Constrictive Pericarditis and Restrictive Cardiomyopathy. JACC Cardiovasc Imaging. Published 8 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →