A perspective on AI implementation in medical imaging in LMICs: challenges, priorities, and strategies
European Radiology, 2026
This article explores how artificial intelligence could help expand access to medical imaging in low- and middle-income countries (LMICs), where limited infrastructure and shortages of trained professionals often delay diagnosis and treatment. This article identified key barriers to implementation and outlined practical strategies to support the responsible and scalable adoption of AI in these settings.
The authors combined a scoping review of peer-reviewed and grey literature with insights from a multidisciplinary expert workshop organized by the Johns Hopkins Science Diplomacy Hub in February 2025. The literature search covered PubMed, Scopus, and IEEE Xplore up to 31 January 2025 and focused on real-world or prospectively validated AI tools used in medical imaging in LMICs. The workshop discussions were then analyzed thematically and integrated into a consensus perspective.
Five main challenges that continue to limit AI implementation in LMICs were identified: infrastructure gaps, data limitations, workforce shortages, unclear ethical and regulatory frameworks, and insufficient long-term funding. It is also shown that progress is possible when solutions are adapted to local needs. Examples from countries such as Nigeria, Uganda, and Colombia highlight the potential of approaches such as low-field MRI, cloud-based PACS, community-engaged data collection, and public–private partnerships. Overall, the article argues that successful AI adoption in medical imaging will depend not only on technical innovation but also on investment in systems, skills, governance, and sustainable implementation models.