Peer-Reviewed Publication
Curr Cardiol Rep2026;28(1)August 31, 2026Journal Article
Echo Chambers: Bias and Representation in Cardiac Imaging Datasets for Artificial Intelligence.
Rahul Gorijavolu1,2,3,4, Nikhil Jaiswal4,5, Jayanth S Pratap1, Kaushik Madapati3,4,6, María Del Pilar Arias López7,8, Marianna Leite9,10, Denise Priscila Muwanguzi11, Nura Izath12, Konrad Samsel13,14, Ifeoluwa T Shoyombo1,3, Leo Anthony Celi15,16,17
1School of Medicine, Johns Hopkins University, Baltimore, MD, USA.
2Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, USA.
3Artificial Intelligence for Responsible, Generalizable, and Open Surgical Research Group, Baltimore, MD, USA.
4Laboratory for Computational Physiology, Massachusetts Institute of Technology, Cambridge, MA, USA.
5Faculty of Medicine and Health Sciences, McGill University, Montreal, QC, Canada.
6Berkeley College of Engineering, University of California, Berkeley, CA, USA.
7Unidad de Cuidados Intensivos Pediátricos (UCIP) del Hospital de Niños Ricardo Gutiérrez, Buenos Aires, Argentina.
8Argentine Society of Intensive Care, Programa de Calidad SATI-Q, Buenos Aires, Argentina.
9School of Medicine, Santa Marcelina College, São Paulo, Brazil.
10School of Public Health, University of São Paulo, São Paulo, Brazil.
11Department of Computing and Technology, Uganda Christian University, Mukono, Uganda.
12Department of Software and Informatics Engineering, Mbarara University of Science and Technology, Mbarara, Uganda.
13Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, ON, Canada.
14Temerty Centre for Artificial Intelligence Research and Education in Medicine, Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto, Toronto, ON, Canada.
15Laboratory for Computational Physiology, Massachusetts Institute of Technology, Cambridge, MA, USA. lceli@mit.edu.
16Division of Pulmonary, Critical Care and Sleep Medicine, Beth Israel Deaconess Medical Center, Boston, MA, USA. lceli@mit.edu.
17Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA, USA. lceli@mit.edu.
Abstract
PURPOSE OF REVIEW: We examined the landscape of publicly available cardiac imaging datasets to assess how their distribution and construction shape bias and equity in cardiovascular AI.
RECENT FINDINGS: Across 38 publicly available echocardiography, cardiac magnetic resonance imaging, and cardiac computed tomography datasets, nearly 80% originate from high-income countries, with no public cardiac…
Create a free account to keep reading
Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.
Want unlimited research access? See Pro plans