Applying artificial intelligence to cardiac MRI to diagnose congenital heart disease in low-resource settings such as Sub-Saharan Africa.
Authors
Affiliations (8)
Affiliations (8)
- School of Medicine, College of Health Sciences, Addis Ababa University, Addis Ababa, Ethiopia. [email protected].
- Krieger School of Arts and Sciences, Johns Hopkins University, Baltimore, MD, USA.
- School of Medicine, Georgetown University, Washington, DC, USA.
- MedStar Emergency Physicians, Washington Hospital Center, Washington, DC, USA.
- Department of Radiology, Hospital Clínico Universitario de Valencia, Valencia, Spain.
- Division of Cardiovascular Imaging, Department of Radiology, Basaksehir Cam and Sakura City Hospital, Istanbul, Turkey.
- Department of Radiology, Johns Hopkins University, Baltimore, MD, USA.
- Department of Radiology, Columbia University Irving Medical Center, New York, NY, USA.
Abstract
Congenital heart disease (CHD) represents a significant burden in Sub-Saharan Africa (SSA), where limited healthcare infrastructure, inadequate diagnostic facilities, and financial constraints contribute to delayed diagnosis and suboptimal care. Cardiac magnetic resonance imaging (CMR), recognized internationally for its exceptional anatomical and functional cardiac assessment capabilities, remains underutilized in SSA primarily due to inadequate infrastructure, high operational costs, lack of trained professionals, and maintenance requirements. Artificial intelligence (AI) has the potential to revolutionize the role of MRI in CHD diagnosis by reducing scan times, automating image processing, and improving diagnostic accuracy. Despite its potential for improving diagnosis, AI implementation is limited by a lack of local datasets, technological incompatibility, data privacy concerns, and lack of expertise among healthcare providers. Strategic interventions such as adopting low-field MRI technologies, enhancing public-private partnerships, and establishing dedicated cardiac imaging units at tertiary centers could significantly expand CMR access and improve diagnosis of CHD in Sub-Saharan Africa. Additionally, targeted training initiatives and locally developed AI solutions that address ethical and interoperability concerns are essential. This Review explores these strategies and emphasizes how CMR augmented by AI could substantially improve CHD diagnosis, clinical outcomes, and healthcare equity in resource-constrained African settings.