2025 Essay Competition Winners

Global Radiology Inequality: What is the Role of High-Income Nations in Bridging the Gap?

Ibrahim Belal

belalibrahim98@gmail.com

Royal London Hospital (Bart's Health Trust)

Introduction

Imaging is the tool used to make illnesses legible, when its available, clinicians can act sooner and more confidently. When it is lacking, uncertainty leads to delay. High income countries have accumulated CT, MRI, interventional capacity and increasingly AI-enabled tools, whereas many low-middle income countries still operate with unpredictable power, little maintenance, a shortage of human resources and limited training pipelines [1-3]. Public funds that could have been used for primary care or prevention are wasted due to delayed diagnosis [4-5]. 

This essay will examine two claims; first, the inequality is institutional and political instead of a technical one. Second, if support is incorporated into locally driven initiatives, the United Kingdom and other western partners can help close the gap. Throughout, I will discuss the key roles of technology transfer, workforce development, and the uses and misuses of artificial intelligence in limited contexts.  

What do history and theory teach about building imaging systems? 

Form the first X ray plates to multi-parametric MRI, radiology has continuously expanded the scope of medical knowledge and expertise, although the rise has been rockier than is commonly described [1]. Procurement cycles have strayed from national plans, training has been provided without approved pathways, and equipment has often been donated without service contracts [2-4]. The failure of what seems to be generous interventions can be explained through a post-colonial lens: exporting devices is simple, but exporting the institutions that support them is more challenging. Generosity decays into downtime unless collaborations create those local institutions. For instance, the Tanzania IR initiative (Road2IR and partners) seems to have produced durable capability by combining supply-chain planning, local leadership and longitudinal mentorship [5].

All along the way, radiology tends to be undermined by conflict. As seen in Iraq, senior staff depart and training pipelines stall, rapidly reducing capacity and supervision. Physical infrastructure also malfunctions; for example, in Gaza, damaged water and power supplies reduce scanner lifespan and, in certain locations, leave CT suites unusable. Supply chains seldom recover; even working machines cannot be used for their intended purpose due to restrictions on contrast agents caused by sanctions and currency volatility. While clinicians improvise with ad hoc teleradiology, such as secure messaging and social media case-sharing, which is helpful in the moment but cannot replace formal networks, care frequently shifts to military facilities. On the other hand, trauma and delayed presentations increase demand. Underutilised mammography units in parts of Gaza and the West Bank are an example of the larger trend of chronic under-service and misallocated equipment [6]. 

How uneven is access today, and what do the numbers reveal? 

According to the Lancet Commission on Diagnostics, approximately 47% of the global population lacks access to basic diagnostics, and a lack of basic tests in primary care facilities where most people seek care initially this causes conditions to manifest later and worsen, which has repercussions downstream, making the diagnostic gap a primary-care issue first [7]. Therefore, improving primary care diagnostics is essential for early detection, reduced expenses, and improved treatment results. This pattern is reflected in scanner availability. According to reports, the average MRI density in Africa is roughly 0.8 units per million, while in high-income areas it is roughly 26.5 units per million [8]. According to analysts, to close this gap, LMICs will require 11.4 more CT and 5.2 more MRI units per million population [9]. This oversimplification, however, ignores the maintenance costs, power quality, and staff needed to run these machines. For instance, Ghana has about three radiologists for every million people, while some areas are said to have none [10]. This suggests that there is a maldistribution both within and between nations. Even less capacity is available for nuclear medicine, which restricts some cardiology pathways and oncologic staging [11]. These differences imply that counting machines is not the only way to measure something. Indicators of equity that show who is truly being served, equipment uptime, workforce distribution, and reporting turnaround should all be monitored by systems.

What happens if we do nothing?

Economically speaking, delayed presentation reduces labour productivity, turns treatable diseases into costly ones, and redirects public funds towards complex care [12–14]. Operationally, even in cases where capacity is available, low staffing and unreliable equipment slow down turnaround times, undermine clinician confidence, and eventually reduce demand. Social impacts are just as significant. Unequal access disproportionately affects households in rural areas. Less wealthy patients must travel further, pay more out of pocket, and miss more follow-up appointments, which increases socioeconomic and geographic disparities in care [15-16]. Those who can afford it are eventually taken on by private providers, while public facilities handle the rest with sporadic services. The end effect is a two-tier structure that, although appearing to have nationwide coverage on paper, provides unequal access [15-16].

For HICs, the priority is not another round of gadgets but the conditions that allow technology to function. AI introduced without local governance may perform poorly and result in vendor lock-in, just as equipment that arrives without lifecycle finances tends to fail quickly, entangling hospitals in downtime and unfunded service contracts [17–20]. External support should be combined with maintenance budgets, dependable power and supply chains, equity-sensitive auditing of reporting times, transparent data stewardship, and local performance evaluation to close this gap rather than strengthen dependency. Instead of sustaining a cycle of scarcity, you have universal and sustainable coverage under those circumstances [15-17].

What can high-income nations do without doing harm? 

When investments follow careers and equipment, capacity tends to increase. Sonographers and radiographers can manage routine tasks effectively without compromising standards thanks to co-accredited residencies and fellowships, ultrasound and interventional academies, supervised procedures, and tiered service models, while radiologists focus on areas where their skills are most useful. Teleradiology can be beneficial, but only under the right conditions. While clinical ownership stays local, well-run networks offer continuous professional development, subspecialist second reading, and surge reporting [6,21]. Locals will be reduced to uploaders and remote reading will simply become an outsourced task in the absence of strong data protection, consistent connectivity, and interoperable reporting. District hospitals sending complex cases for second reads would be a good design. In a predetermined amount of time, reports and teaching notes are returned to the local PACS. To upskill the local team, monthly discrepancy meetings are planned. Lastly, the designated responsible owners of their reports continue to be the local clinicians. 

When lifecycle support is included, technology transfer is successful. An unmaintained scanner becomes a burden. Along with funded maintenance contracts, spare-parts logistics, and biomedical engineering training, more suitable pathways favour digital radiography, point-of-care ultrasound, power conditioning with solar augmentation, and dose-management tools [19]. Western partners can help systems buy transparently, buy collectively, and regulate efficiently.

In most situations, national and regional pooled procurement can reduce unit prices, stabilise supply, and prevent fragmentation. It also makes it possible to specify service contracts, spare parts, and training all in one package. The architecture is completed by regulatory strengthening: qualified authorities license facilities and operators, register devices and contrast media, establish radiation protection and quality-control standards, and conduct incident reporting and recalls. It is more likely to be available, safe, and maintainable when these features are present. 

This discussion should include artificial intelligence, but with limitations. In high volume pathways like emergency CT and chest radiographs, AI may help with triage, standardise quality control, and prioritise urgent findings [7]. But biassed results could be propagated by models trained on unrepresentative datasets, and its dependability could be jeopardised by fragile infrastructure [3]. A safer approach would be to prioritise data stewardship and local validation first, workforce training next, and finally deployment, but only with audit and feedback loops in place. Evaluations should assess whether AI narrows or widens equity gaps. 

Future Directions

To consolidate progress, three shifts are needed. Instead of episodic donations, financing should take the form of multi-year donations that cover training, maintenance, consumables, and evaluation alongside equipment, in addition to donated equipment. Institutionally, the systems to acquire and sustain technology would benefit from investments in biomedical engineering schools, regional maintenance hubs, RIS/PACS interoperability, and procurement reform. Locally led knowledge production is also needed to support appropriate imaging pathways, perform cost-effectiveness studies, and develop AI models that are trained and audited on diverse populations [19,22-23]. The benefits are mutual. LMICs will be able to conduct early diagnosis and avoid late-stage referrals, improving outbreak detection and containment, and reducing cross-border risks, and creates more representative data sets that enhance the AI and guideline cross country applicability for higher income countries.

Conclusion 

Radiology inequality is structural. High income partners must support the conditions that make the technology work through workforce, maintenance, data stewardship, and locally led partnerships. Priorities include teleradiology that teaches as it helps, lifecycle-backed procurement, and cautious, validated AI tied to equity-sensitive audit. With this approach it is plausible to turn pockets of excellence into routine practice. That is neither automatic nor guaranteed, but it is within reach if metrics, ethics and organisational design are regarded as seriously as the technology itself.

References:

1. Market Research Future, https://www. marketresearchfuture. com/ (2023) UK Radiology Services Market Size, growth report 2035, Size, Growth Report 2035. Available at: https://www.marketresearchfuture.com/reports/uk-radiology-services-market-51019 (Accessed: 10 October 2025). 

2. Re: Imaging inequality: exploring the differences in radiology between high- and low-income countries, Jackson, M.R. Clinical Radiology, Volume 79, Issue 10, e1179

3. Global radiology: Building equitable and effective partnerships with low and middle-income countries, Kwarteng, Lydia A. et al. Clinical Imaging, Volume 125, 110559

4. Abuzaid, M.M., Almuqbil, N. Charting a sustainable future in radiology: evaluating radiologists’ knowledge, attitudes, and practices toward environmental responsibility. Insights Imaging 16, 39 (2025). https://doi.org/10.1186/s13244-025-01917-7

5. Matsumoto MM Dako F. Working at the Intersection of Radiology and Global Health Equity. Appl Radiol. 2023;(1):35-37.,doi:undefined

6. Suji, T., Sullivan, R. & Bowsher, G. Radiology in conflict: scoping review. Confl Health 18, 8 (2024). https://doi.org/10.1186/s13031-023-00550-9

7. Essential diagnostics: mind the gap, The Lancet Global Health,The Lancet Global Health, Volume 9, Issue 11, e1474

8. Anazodo, U.C. et al. (2022a) A framework for advancing sustainable MRI access in Africa [Preprint]. doi:10.1101/2022.05.02.22274588. 

9. Frija G, Blažić I, Frush DP, Hierath M, Kawooya M, Donoso-Bach L, Brkljačić B. How to improve access to medical imaging in low- and middle-income countries ? EClinicalMedicine. 2021 Jul 17;38:101034. doi: 10.1016/j.eclinm.2021.101034. PMID: 34337368; PMCID: PMC8318869.

10. Sarkodie, B.D., Ohene-Botwe, B., Mensah, Y.B. et al. Density and regional distribution of radiologists in a low-income country: the Ghana situation. Chin J Acad Radiol 6, 188–195 (2023). https://doi.org/10.1007/s42058-023-00130-z

11. Grigoryan, A., Bouyoucef, S., Sathekge, M. et al. Development of nuclear medicine in Africa. Clin Transl Imaging 10, 101–111 (2022). https://doi.org/10.1007/s40336-021-00468-3

12. Gebrewold, Y. (2035) ‘Bridging the AI Gap in Clinical Imaging: Opportunities and Strategies for Low- and Middle-Income Countries’, Journal of Global Radiology [Preprint]. doi:10.7191/jgr.985. 

13. DeStigter, K., Pool, KL., Leslie, A. et al. Optimizing integrated imaging service delivery by tier in low-resource health systems. Insights Imaging 12, 129 (2021). https://doi.org/10.1186/s13244-021-01073-8

14. Kocak, B., Ponsiglione, A., Romeo, V. et al. Radiology AI and sustainability paradox: environmental, economic, and social dimensions. Insights Imaging 16, 88 (2025). https://doi.org/10.1186/s13244-025-01962-2

15. Lopez-Suarez, N. et al. (2023) ‘Practical approaches to advancing health equity in Radiology, from theajrspecial series on Dei’, American Journal of Roentgenology, 221(1), pp. 1–10. doi:10.2214/ajr.22.28783. 

16. The global future of imaging. Available at: https://www.bir.org.uk/media/408496/the_global_future_of_imaging_a4_24pp-hr.compressed.pdf (Accessed: 10 October 2025). 

17. Chen, Y., Stavropoulou, C., Narasinkan, R. et al. Professionals’ responses to the introduction of AI innovations in radiology and their implications for future adoption: a qualitative study. BMC Health Serv Res 21, 813 (2021). https://doi.org/10.1186/s12913-021-06861-y

18. Rural areas face imaging obstacles on the road to Health Care EquityRSNA. Available at: https://www.rsna.org/news/2021/june/rural-radiology-equity (Accessed: 10 October 2025). 

19. Improving access to medical imaging for more patients. ge monogram primary white RGB. Available at: https://www.gehealthcare.co.uk/insights/article/improving-access-to-medical-imaging-for-more-patients (Accessed: 10 October 2025). 

20. A paradigm shift in point-of-care imaging in low-income and middle-income countries. Frija, Guy et al. eClinicalMedicine, Volume 62, 102114

21. Mandal I, Minocha A, Yeung J, Bandula S, Rabouhans J. Interventional radiology training: a comparison of 5 English-speaking countries. Br J Radiol. 2020 Jan;93(1105):20190340. doi: 10.1259/bjr.20190340. Epub 2019 Oct 23. PMID: 31596121; PMCID: PMC6948082.

22. How to improve radiology access in underserved communities | radiology key. Available at: https://radiologykey.com/how-to-improve-radiology-access-in-underserved-communities/ (Accessed: 10 October 2025). 

23. Innovations in Imaging and Radiotherapy, Bir. Available at: https://www.bir.org.uk/media/458077/bir114.026_-_innovations_brochure_21_web_2.pdf (Accessed: 10 October 2025).