By Simpson Global Media News Desk
Nigeria and global health partners are stepping up discussions on the use of artificial intelligence and other digital technologies to strengthen malaria elimination, improve disease surveillance and make limited health resources more precisely targeted.
The renewed push emerged during a high-level meeting on the margins of the 81st United Nations General Assembly in New York, where Nigeria joined the Global Institute for Disease Elimination (GLIDE) and the RBM Partnership to End Malaria to examine how artificial intelligence could be integrated into national disease-control programmes.
The September 23 session, titled “Smarter Investment, Greater Impact: Harnessing AI to Accelerate Disease Elimination,” brought together health ministers, funders, diagnostics specialists, technology stakeholders and malaria advocates to discuss the practical conditions required for AI-enabled health interventions to move from experimentation into sustainable public-health programmes.
The discussion comes as health programmes face pressure from constrained development and humanitarian financing while countries continue to manage malaria and other preventable infectious diseases.
For Nigeria, the issue has particular significance because the Federal Government has placed malaria elimination within its 2026–2030 national health strategy, while the country continues to carry a substantial share of the global malaria burden.
Nigeria's Federal Ministry of Health and Social Welfare said earlier this year that the country's malaria prevalence had fallen from 42 per cent in 2010 to 15 per cent in 2025, but stressed that the disease remains a major public-health challenge requiring sustained intervention.
The emerging AI strategy is therefore being discussed not as a replacement for bed nets, medicines, vaccination, vector control, testing or frontline health workers, but as a possible additional layer for identifying patterns, improving surveillance, supporting decisions and directing interventions.
From Technology Demonstrations to Health-System Tools
Artificial intelligence has already entered parts of Nigeria's health innovation ecosystem.
Researchers, technology companies and health programmes have been developing systems capable of analysing medical images, patient information and disease patterns.
One example is a malaria pre-screening system developed by researchers at the University of Lagos. A July 2026 preprint describes a dual-mode AI system that combines analysis of patient biosignals with an image-based model trained on microscopy images of red blood cells.
The researchers reported that their clinician-focused model achieved 94.84 per cent accuracy in the reported evaluation, while the patient-focused model recorded 94.6 per cent accuracy under the study's specified train-test configuration. The work remains a research preprint rather than evidence that the system has become a nationally deployed diagnostic standard.
Another Nigerian innovation is being developed around retinal imaging.
AEyeCARE Nigeria's ASPIRE platform uses a portable retinal camera and artificial intelligence to analyse retinal images for indicators associated with severe or cerebral malaria. The company says its algorithm was trained on more than 100,000 retinal images and can analyse an image in less than a minute. These are company-reported capabilities and do not by themselves establish nationwide clinical effectiveness.
Such projects illustrate the range of possible applications.
AI can potentially assist with clinical screening, but its public-health role can extend much further.
Algorithms can analyse large quantities of surveillance information, identify patterns that may be difficult to detect manually, help forecast areas of increased transmission and support decisions about where limited supplies or interventions should be deployed.
That broader use of AI was at the centre of the UNGA discussion.
Why Surveillance Matters
Malaria control depends heavily on timely information.
Health authorities need to know where cases are occurring, when transmission is increasing, which populations are being affected and whether interventions are reaching the areas where they are most needed.
The quality of those decisions depends on the quality and timeliness of available data.
Nigeria has a particularly complex health-information environment because care is delivered through public facilities, private hospitals, primary healthcare centres, pharmacies, patent and proprietary medicine vendors and community-based services.
A substantial portion of healthcare activity occurs outside government hospitals.
That can create gaps between what happens at the point of care and what eventually reaches national surveillance systems.
The National Malaria Elimination Programme and Sproxil Nigeria have been working on an AI-enabled approach intended to improve malaria surveillance within the private sector.
At a strategic meeting in Lagos in August, officials from the Federal Ministry of Health, malaria researchers, development partners, the World Health Organisation, PATH and digital-health specialists examined how AI and digital tools could strengthen malaria testing, reporting and public-health communication.
The issue is important because a malaria case treated at a private medicine outlet may be clinically managed but not necessarily contribute promptly to the national picture of where transmission is occurring.
Improved digital reporting could potentially help public-health authorities see those cases more quickly.
AI could then be used to analyse the larger volume of information.
The technology, however, depends on the underlying data being accurate, representative and consistently collected.
Country Ownership at the Centre
A major message from the UNGA meeting was that countries affected by malaria should determine how AI is introduced into their own health systems.
The approach was explicitly framed around country ownership rather than technology being introduced independently of national programmes.
Nigeria's Coordinating Minister of Health and Social Welfare, Professor Muhammad Ali Pate, delivered keynote remarks at the event, with the emphasis on countries determining how emerging technologies fit into their health systems.
The RBM Partnership to End Malaria also stressed that malaria-endemic countries need to remain central to decisions about how artificial intelligence and other technologies are applied.
The position reflects an important practical consideration.
A technology that performs well in a laboratory or demonstration project may not automatically work in a rural Nigerian health facility.
Connectivity may be limited.
Electricity may be unreliable.
Health workers may have limited time for additional digital procedures.
Data may be incomplete.
Equipment may require maintenance.
A system trained predominantly on data from one population may perform differently when applied to another.
These issues mean that AI deployment must be designed around actual health-system conditions.
The New Training Push
One of the concrete announcements connected to the UNGA meeting was a new AI training effort by GLIDE for public-health leaders and national programme managers.
The initiative includes GLIDE Leadership Excellence in AI for Health, known as GLIDE_LEAH, for senior decision-makers.
It also includes AI for All, developed by the Mohamed bin Zayed University of Artificial Intelligence in collaboration with GLIDE for public-health professionals worldwide.
The objective is to build the ability of health leaders to understand and use AI rather than leaving decisions entirely to technology developers.
That distinction is increasingly important as AI systems become more common.
A health ministry does not only need people who can operate a software platform.
It needs decision-makers who can ask whether the data are appropriate, whether the algorithm is reliable, whether the results are clinically meaningful, whether privacy requirements are being met and whether the technology actually improves outcomes.
The training initiative therefore places capacity building alongside technology.
AI Is Not a Substitute for Health Workers
The discussion around artificial intelligence also raises questions about the role of doctors, nurses, laboratory scientists, community health workers and public-health officers.
The current approach being discussed by health partners does not present AI as a substitute for those workers.
Instead, AI is being considered as a decision-support and information tool.
For example, an algorithm may flag a patient or location requiring additional attention, but a trained professional would still need to interpret the result and determine the appropriate response.
This distinction is particularly important in malaria because diagnosis and treatment decisions can have serious consequences.
A false positive can lead to inappropriate treatment.
A false negative can delay care for a patient who needs it.
For surveillance systems, inaccurate data can cause health authorities to send resources to the wrong places or overlook communities experiencing increased transmission.
Human oversight therefore remains important.
The emerging global discussion on AI in health increasingly focuses on what tasks can safely be automated, what tasks require professional supervision and what safeguards are needed when algorithms are used.
Nigeria's Malaria Strategy
The AI discussion is taking place alongside Nigeria's broader malaria strategy for 2026 to 2030.
The Federal Government announced the new National Malaria Strategic Plan in April, describing it as a framework for reshaping the national response despite the decline in malaria prevalence recorded between 2010 and 2025.
The strategy comes as Nigeria continues to combine multiple malaria interventions.
These include vector control, insecticide-treated nets, seasonal malaria chemoprevention, diagnosis and treatment, surveillance, vaccination and community-based interventions.
In August, the government also announced plans to advance policy measures for large-scale introduction of Perennial Malaria Chemoprevention for children under two in areas with year-round transmission. The programme is intended to provide preventive medicines during periods when young children remain at elevated risk.
AI could potentially complement those programmes by helping determine where and when interventions are most needed.
For example, disease-surveillance models could analyse geographic and seasonal patterns to support planning.
Digital systems could help monitor whether interventions are reaching intended populations.
Data tools could assist programme managers in identifying areas where reported coverage differs from expected needs.
But those applications would need to be validated in Nigeria's operating environment.
A Financing Problem as Well as a Technology Problem
The AI discussion is also closely linked to financing.
Health programmes around the world are facing pressure as international development and humanitarian financing becomes more constrained.
The UNGA meeting explicitly examined how countries could use technology to make existing investments more efficient rather than simply increasing spending.
That does not mean AI is automatically cheaper.
Developing, testing and maintaining digital systems can require significant investment.
Governments need computing infrastructure, secure databases, connectivity, technical personnel, cybersecurity, software maintenance and training.
There may also be costs associated with integrating new systems with existing health-information platforms.
For a low- and middle-income country, those costs must be considered against other pressing health needs.
The central question is therefore not simply whether AI is innovative.
It is whether a particular application delivers enough additional public-health value to justify the resources required to develop and maintain it.
That is why the UNGA discussions focused on implementation conditions, financing and evidence alongside technology.
Connecting Capital With Viable Projects
Participants at the New York event included representatives of the Pandemic Fund, Roche Diagnostics and Malaria No More US, with discussion focused on how investment can be connected with viable AI-enabled health opportunities.
The panel examined the infrastructure, workforce and health-system requirements needed for sustained implementation.
This is important because many digital-health projects begin as pilots.
A pilot may operate for a limited period with dedicated funding and technical support.
The challenge comes when the pilot ends.
If the government cannot afford the software licence, if the equipment cannot be maintained or if trained personnel leave, the programme may stop.
The objective of the new capacity-building approach is therefore broader than funding individual experiments.
It seeks to strengthen the ability of countries to assess technologies and incorporate useful ones into established systems.
The Data Question
Data will be one of the most important foundations of AI-enabled malaria control.
Algorithms learn from data, and public-health systems depend on information that is sufficiently accurate and representative.
If rural communities are underrepresented in the data, an algorithm may not perform equally well for those populations.
If private-sector cases are missing, surveillance may underestimate disease activity.
If reporting systems contain delays or inconsistent definitions, the resulting models can inherit those problems.
For Nigeria, improving data quality is therefore part of the AI challenge.
The August partnership between the National Malaria Elimination Programme and Sproxil illustrates this issue.
The initiative is aimed partly at addressing information gaps associated with malaria activity in private-sector health settings.
Bringing more data into surveillance systems could provide a broader picture of malaria activity, but the data must also be governed appropriately.
That includes questions about who owns the data, who can access it, how it is stored and how it can be used.
Health Data Sovereignty
Data sovereignty has emerged as a significant issue in discussions about AI and health.
Health information can include highly sensitive personal details.
When AI systems are developed or hosted by external organisations, governments and health institutions need to understand where data are stored, who processes them and under what legal arrangements.
The GLIDE-related discussions have therefore emphasised country ownership and the need for countries to control their data and decisions.
The organisation has said AI must be supported by strong data systems, laboratories, health workforces, surveillance infrastructure and appropriate safeguards.
For Nigeria, that raises the importance of developing local technical capacity alongside partnerships with international technology and health organisations.
International collaboration can provide expertise and investment.
But domestic institutions need enough capacity to evaluate, regulate and operate the systems being introduced.
AI and Diagnostics
One of the most visible areas for AI in malaria is diagnosis.
Traditional malaria diagnosis relies primarily on microscopy and rapid diagnostic tests.
Microscopy requires trained laboratory personnel and appropriate equipment.
Rapid diagnostic tests can provide results quickly but also have limitations related to test performance, parasite density and other factors.
AI-based approaches could potentially assist by analysing images or other patient information.
The University of Lagos preprint provides one example of research in this direction.
Its image-based model was trained on annotated microscopy images of red blood cells, while its patient-focused component used information including heart rate, body temperature and oxygen saturation.
The researchers described the system as a pre-screening and diagnostic-support approach, particularly for settings where resources are limited.
However, research-stage performance should not be confused with established clinical deployment.
Before a diagnostic AI system can be used widely, it needs appropriate clinical validation, regulatory assessment, monitoring and evidence across different populations and settings.
That process is particularly important for high-stakes medical decisions.
Retinal Imaging and Severe Malaria
Nigeria's health-technology sector is also experimenting with less conventional diagnostic approaches.
AEyeCARE's ASPIRE platform uses retinal images as a potential source of information about severe or cerebral malaria.
The company says the system analyses microscopic changes in retinal blood vessels and can provide results rapidly through a portable device.
The approach is notable because it seeks to move some advanced screening capability closer to frontline workers.
In a country with a large and geographically dispersed population, technologies that can operate outside major hospitals could potentially expand access to specialist decision support.
But again, implementation requires evidence.
A promising technology needs to demonstrate that it works reliably across the populations and settings in which it will be used.
It also needs clear protocols defining what happens after an AI system produces a result.
If a frontline worker receives an alert indicating possible severe malaria, there must be a functioning referral pathway capable of getting the patient to appropriate treatment.
Technology cannot compensate for an absent emergency-care system.
The Last-Mile Challenge
Nigeria's malaria burden is heavily influenced by conditions at community level.
Many patients live far from major hospitals.
Primary healthcare centres may operate with limited staff and equipment.
Some communities face transportation difficulties, insecurity or seasonal isolation.
These realities make last-mile health delivery central to malaria elimination.
AI could potentially help health authorities decide where to deploy mobile services, medicines, diagnostic supplies and community interventions.
It could also support community health workers with decision-support tools.
But digital interventions need to function under the same constraints as the health system itself.
A sophisticated AI model is of limited value if the health worker cannot connect to it, the device cannot be charged or the patient cannot be referred.
This is why the UNGA discussions emphasised infrastructure and workforce capacity.
Building the Digital Foundation
Nigeria has been investing in broader digital-health infrastructure as part of its health-sector reforms.
The government and development partners are working to improve health information systems, digital identification, electronic records and data use.
The expansion of fibre infrastructure and data-centre capacity discussed elsewhere in Nigeria's technology sector could also eventually support health applications by improving connectivity and computing capacity.
But health data systems require sector-specific governance.
Medical records cannot simply be treated as ordinary commercial data.
They involve confidentiality, patient consent and professional responsibilities.
As AI becomes more widely used, Nigeria will need to strengthen the rules governing algorithmic decision-making, health-data processing and accountability for errors.
Avoiding the Pilot Trap
A recurring challenge in digital health is the gap between innovation and scale.
A technology can work successfully in one hospital without being ready for nationwide deployment.
The pilot environment may have unusually strong technical support, trained staff and dedicated financing.
Scaling introduces different conditions.
There may be thousands of facilities, different levels of connectivity and major variations in workforce skills.
The new GLIDE training programmes address part of this challenge by focusing on decision-makers and programme managers rather than technology alone.
The objective is to strengthen the capacity to determine which tools are appropriate, how they should be implemented and what safeguards are needed.
This approach also allows countries to reject technologies that do not meet their needs.
Country ownership means technology adoption is connected to national priorities rather than being driven solely by external availability.
Malaria Beyond Nigeria
The same questions apply across malaria-endemic countries.
Different countries have different transmission patterns, health-system structures, data environments and resource constraints.
An AI model developed in one country may need adaptation before it can be used elsewhere.
That is why the new GLIDE-RBM partnership is structured around country-led action.
The organisations signed a Letter of Mutual Understanding to strengthen collaboration around malaria control and elimination, connecting technical expertise, AI innovation and capacity building with national malaria programmes and affected communities.
The partnership is aligned with the World Health Organisation's Global Technical Strategy for Malaria.
The strategy places emphasis on country leadership, surveillance, prevention, diagnosis, treatment and health-system strengthening.
AI can support these areas, but it is one component of a much larger malaria-control architecture.
What AI Could Do for Programme Managers
For a national malaria programme manager, the value of AI may not necessarily come from a single dramatic diagnostic breakthrough.
It could come from hundreds of smaller improvements in decision-making.
A system could combine case reports, rainfall information, temperature, population movement, treatment data and intervention coverage to identify areas requiring additional attention.
It could help programme managers compare expected and observed disease patterns.
It could support forecasting of medicine requirements.
It could identify unusual clusters requiring investigation.
It could help analyse whether intervention campaigns are reaching intended populations.
Such systems could make public-health planning more data-driven.
But the outputs would still need human review.
Public-health decisions involve factors that may not appear in datasets, including insecurity, migration, local beliefs, health-worker availability and community behaviour.
Climate and Malaria
Climate is another area where AI could potentially support malaria control.
Temperature and rainfall influence mosquito populations and malaria transmission.
Nigeria experiences significant geographic and seasonal variation, meaning malaria risks differ across regions.
Climate-related changes can alter the timing and intensity of transmission.
Data models could potentially combine meteorological information with disease surveillance to help anticipate changes in malaria risk.
That could support earlier deployment of interventions.
However, such models also require reliable local data and appropriate validation.
A prediction is only useful if health authorities can act on it.
If an AI system identifies increased risk but medicines, diagnostic supplies or personnel cannot be moved into the affected area, the value of the prediction is limited.
Evidence Before Expansion
The increasing interest in AI creates a parallel need for stronger evidence.
Health authorities will need to know not only whether an AI model is technically accurate but whether its use improves health outcomes.
For a diagnostic system, relevant questions include whether it reduces time to diagnosis, reduces missed cases and improves treatment decisions.
For surveillance systems, questions include whether they detect outbreaks earlier or improve targeting of interventions.
For health-system tools, the question may be whether they reduce administrative workload or improve continuity of care.
These are different measures from algorithmic accuracy.
A system can have high technical accuracy but still provide little practical value if it is difficult to use or expensive to maintain.
Conversely, a modestly sophisticated system could be valuable if it solves a clearly defined operational problem.
Safeguards and Accountability
As AI becomes more prominent in healthcare, safeguards will become increasingly important.
Health authorities need clear rules for human oversight.
Patients need protection against inappropriate use of personal information.
Developers need to monitor systems for errors and performance changes.
Health workers need to understand the limitations of the tools they use.
Regulators need the capacity to evaluate AI-enabled medical products and services.
These safeguards are particularly important where AI is used to make or influence clinical decisions.
The UNGA discussions have therefore linked AI adoption with governance, workforce development, data stewardship and health equity rather than treating the technology as an isolated product.
Nigeria's Opportunity
Nigeria's scale gives it a significant potential test environment for health technology.
With a large population, extensive malaria transmission and a rapidly developing technology ecosystem, the country has both a major public-health need and a large potential user base for successful innovations.
The emergence of Nigerian AI-health companies and university research demonstrates that some of the technological capacity already exists locally.
The challenge is connecting those innovations to public-health priorities.
That requires cooperation between government agencies, universities, hospitals, technology companies, investors, international organisations and communities.
It also requires procurement systems capable of evaluating technology on evidence rather than novelty.
A Broader Health-System Transformation
The malaria discussion is part of a larger transformation taking place in healthcare.
AI is increasingly being considered for maternal health, medical imaging, disease surveillance, drug discovery, administrative work and clinical decision support.
Several UNGA81 events this week have focused on responsible AI adoption, including sessions on maternal health, public-health security and the use of AI in lower-income health systems.
For Nigeria, malaria provides a particularly important area for testing these ideas because the disease has a large public-health footprint and requires extensive surveillance and intervention.
Lessons learned from malaria could eventually inform digital approaches to other diseases.
What Happens Next
The immediate next step is likely to be continued collaboration between Nigeria, GLIDE, RBM and other health and technology partners.
The GLIDE-RBM agreement provides a framework for connecting country priorities with technical expertise, innovation and capacity building.
GLIDE's new training programmes are also expected to expand the number of public-health leaders with practical knowledge of AI.
Within Nigeria, the National Malaria Elimination Programme's work with private-sector health providers and digital-health partners provides another avenue for testing how better data can strengthen surveillance.
The country's new 2026–2030 malaria strategy provides the wider policy framework into which these technologies could fit.
The key issue will be implementation.
Nigeria will need to determine which AI applications solve clearly defined health problems, establish evidence that they work, integrate them with existing systems and ensure that frontline workers have the training and infrastructure needed to use them.
Technology With a Public-Health Purpose
The emerging discussion is therefore less about artificial intelligence as a technology in isolation and more about how it can be used to strengthen the machinery of disease control.
For malaria, that machinery includes surveillance officers, laboratory scientists, doctors, nurses, community health workers, pharmacists, researchers, supply-chain managers and public-health officials.
AI can potentially help these professionals process information more quickly and identify patterns that would otherwise be difficult to see.
But the effectiveness of those tools will depend on the systems surrounding them.
Reliable data will be needed.
Laboratories will remain necessary.
Medicines and diagnostic supplies must reach facilities.
Health workers must be available.
Patients must be able to access care.
Referral systems must function.
Communities must trust the interventions being offered.
And governments must have the capacity to sustain programmes after individual donor-funded projects end.
The UNGA discussions have placed those conditions at the centre of the AI debate.
The Larger Malaria Goal
Nigeria's malaria strategy has moved into a period in which declining prevalence provides evidence of progress while the continuing burden requires sustained action.
The Federal Government's reported reduction from 42 per cent prevalence in 2010 to 15 per cent in 2025 is significant in terms of the trajectory described by government, but it does not mean malaria has ceased to be a major public-health issue.
The continued need for prevention, diagnosis and treatment means that health authorities are looking for ways to make interventions more targeted and efficient.
Artificial intelligence is now entering that conversation.
The technology could assist with diagnosis, surveillance, forecasting, resource allocation and programme management.
Research projects in Nigeria are already exploring some of those possibilities, while partnerships such as the NMEP-Sproxil initiative are addressing the data side of malaria control.
The new international partnership involving GLIDE and RBM adds another layer by focusing on country leadership and capacity building.
From Promise to Proof
The next phase will require evidence.
AI systems will need to demonstrate that they can operate reliably in real Nigerian health settings and that they produce measurable benefits.
Research-stage accuracy figures can provide an early indication of technical performance, but national health programmes need evidence covering real patients, diverse communities, different levels of healthcare and long-term operational conditions.
That process may take time.
It will also require collaboration between technology developers and health professionals.
The developers understand the algorithms.
Health workers understand the clinical environment.
Public-health authorities understand the population-level requirements.
Communities understand the practical realities of receiving care.
Bringing those perspectives together will determine whether AI becomes a sustainable health-system tool or remains a collection of isolated projects.
For Nigeria, the opportunity is to build those technologies around national priorities, local data and existing health programmes.
For international partners, the emerging model is one in which expertise and financing support country-led decisions rather than determining them.
And for malaria programmes, the immediate objective remains straightforward: detect disease earlier, target interventions more effectively, strengthen treatment and prevent avoidable deaths.
Artificial intelligence may become part of how that work is done.
The evidence from Nigeria's research community, digital-health partnerships and the latest international discussions suggests that the technology is moving closer to the mainstream of malaria-control planning. But its eventual contribution will depend not on the novelty of the algorithms, but on whether they can be responsibly integrated into the people, institutions and systems that deliver healthcare every day.
Comments
Post a Comment