… as 70% of Nigerians pay medical bills out-of-pocket

…say AI can help Nigeria fight TB, cut healthcare costs

By Joseph Erunke, Abuja

With Nigeria battling an estimated 510,000 tuberculosis cases annually, severe shortages of healthcare professionals and a healthcare financing system that leaves more than 70 per cent of citizens paying medical bills from their pockets, health and technology experts have identified artificial intelligence, AI, as a potential game-changer in the country’s struggling health system.

The experts, however, warned that without a robust regulatory framework, locally trained AI models, reliable health data and clear patient-safety standards, the technology could compound existing challenges rather than deliver the improvements Nigerians desperately need.

They spoke in Abuja on Thursday,ahead of the country’s first Artificial Intelligence in Health Summit in Sub-Saharan Africa, where government officials, healthcare professionals, technology developers, regulators and development partners are expected to explore practical ways of deploying AI to improve diagnosis, expand access to healthcare and reduce the cost of medical services.

Founder and Chief Executive Officer of the Africa Hub for Innovation and Development, AHFID, Dr. Kunle Kakanfo, said Nigeria could no longer afford to approach AI in healthcare as a fashionable technology trend without translating its potential into measurable improvements in health outcomes.

He argued that the technology could help bridge critical gaps in the country’s overstretched healthcare workforce, accelerate disease detection, improve health insurance administration and expand access to essential medical services, particularly for underserved populations.

Kakanfo, however, stressed that AI must complement, rather than replace, human medical expertise, warning that patient safety, ethical deployment and proper validation of AI-powered medical tools must remain central to any national adoption strategy.

According to him, the proposed summit is designed to move discussions beyond technological possibilities to practical solutions, regulatory safeguards and commitments that could shape a national roadmap for AI in healthcare.

Making a case for urgent action, Kakanfo highlighted the country’s persistent disease burden, inadequate healthcare workforce and high out-of-pocket expenditure as compelling reasons to explore carefully regulated AI applications.

He cited tuberculosis as a major public health concern, noting that Nigeria records an estimated 510,000 cases annually, making it one of the countries with the highest disease burdens globally.

He also pointed to the continuing migration of Nigerian-trained doctors abroad, which further strains healthcare delivery and limits the country’s capacity to provide timely medical services.

On healthcare financing, Kakanfo noted that more than seven in every 10 Nigerians still pay for medical services directly from their pockets, leaving many households vulnerable to financial hardship whenever illness strikes.

He said the figures were not intended to portray the country as having made no progress, but to demonstrate the urgency of accelerating reforms and exploring innovations capable of improving access, efficiency and outcomes.

“The question is whether we should convene another summit just for the sake of having a summit, or whether we should convene one that can drive improved outcomes,” he said.

He explained that AI could help address some of these challenges by supporting faster diagnosis, improving the efficiency of healthcare workers and reducing administrative bottlenecks that prevent health facilities and insurance providers from serving more patients.

One of the most promising applications, according to Kakanfo, is the use of AI-powered computer-aided detection systems to analyse chest X-rays for tuberculosis screening and patient triage.

He said the World Health Organisation had recommended computer-aided detection for TB screening using chest X-rays, creating an opportunity for Nigeria to deploy the technology in a country where shortages of radiologists and other specialised healthcare professionals can delay diagnosis.

Kakanfo explained that AI-enabled imaging tools could help health facilities screen more patients, identify suspected cases more quickly and direct those requiring further investigation to appropriate medical personnel.

He also disclosed that work was already underway in hospitals within the Federal Capital Territory to collect cough sounds for the development of locally trained AI models capable of detecting tuberculosis.

The initiative, he said, reflects the need to develop solutions using data drawn from Nigerian and African populations rather than relying exclusively on technologies trained with foreign datasets.

He argued that locally developed models could improve the relevance of AI applications to the country’s disease patterns and population characteristics.

Beyond tuberculosis, he said researchers were exploring the use of sound-based AI models to identify other health conditions, including neonatal sepsis, demonstrating the potential of locally generated data to support early diagnosis.

Kakanfo also identified the use of AI-powered clinical documentation tools as a practical way to improve productivity in Nigeria’s overstretched health facilities.

He explained that automated documentation systems, sometimes called ambient scribes, could help clinicians record patient consultations and organise medical notes, reducing the time doctors spend on paperwork.

According to him, such tools could enable healthcare professionals to devote more time to patients and potentially increase the number of people treated daily.

He gave the example of primary healthcare centres, where improved efficiency could allow doctors to attend to more patients without necessarily requiring a corresponding increase in working hours.

“If a doctor needs to talk to a patient, the doctor can use an ambient scribe, which structures the consultation notes in the way a doctor would ordinarily do,” he said.

He maintained that increasing the capacity of primary healthcare facilities could help reduce the number of patients compelled to seek services at more expensive private hospitals simply because public facilities cannot attend to them promptly.

Kakanfo further identified AI-assisted processing of health insurance claims as another area where technology could reduce administrative delays and improve service delivery.

He explained that faster verification of claims could help health maintenance organisations, HMOs, recover funds more quickly and expand the number of people they cover.

By reducing operational bottlenecks and improving the flow of funds, he said, AI could help strengthen the capacity of health insurance providers to extend services to more Nigerians.

He nevertheless stressed that the technology’s contribution to reducing out-of-pocket expenditure would depend on broader improvements in access to primary healthcare, medicines and health insurance.

The AHFID chief executive also disclosed that locally developed AI technology was being deployed in cervical cancer screening, where an image of a woman’s cervix is analysed to identify possible precancerous lesions.

He said the initiative illustrated how AI trained on relevant local data could support healthcare workers in identifying conditions requiring further clinical assessment.

Kakanfo cited performance figures from the initiative suggesting that the tool could achieve about 80 per cent accuracy, compared with approximately 30 per cent for visual assessment by a trained clinician.

The figures, however, would need to be interpreted in the context of the specific study, clinical setting and evaluation method used.

He stressed that AI-assisted screening should support clinical judgement rather than serve as the final authority on diagnosis or treatment.

According to him, Nigeria must identify practical, high-impact applications that can be safely deployed and scaled instead of importing technologies without establishing whether they are suitable for local healthcare needs.

He warned that AI models trained predominantly on foreign data might not always perform optimally when applied to Nigerian patients, making local data collection, model development and rigorous evaluation critical.

Kakanfo said AI was already finding its way into clinical practice, including through doctors’ use of conversational tools such as ChatGPT to support consultations and other professional tasks.

He referred to research suggesting substantial use of conversational AI among doctors, arguing that the trend made clear rules and professional guidance increasingly urgent.

However, he warned that AI-generated responses could be inaccurate or misleading, making it dangerous to rely on such tools without appropriate clinical oversight.

Senior Technical Adviser to the Director-General of the Nigeria Centre for Disease Control and Prevention, NCDC, Dr. Abiola Oshunniyi, said Nigeria must move quickly to establish a coordinated framework for incorporating AI into healthcare.

Oshunniyi said the country had made progress in developing AI capabilities outside the health sector, but the next challenge was to translate those advances into practical solutions for the healthcare system.

He disclosed that a bill on e-health was before the National Assembly and suggested that the ongoing legislative process could provide an opportunity to incorporate relevant provisions on AI in healthcare.

He also proposed that recommendations emerging from the summit be presented to the National Council on Health for consideration as a basis for a national resolution and roadmap.

According to him, Nigeria should not wait until other countries have fully adopted AI before putting the necessary policies, standards and implementation structures in place.

He identified disease surveillance, outbreak prediction, the early positioning of medical countermeasures and improved healthcare delivery as areas where AI could contribute to national health security.

Oshunniyi also called for a framework that would enable locally developed health technology solutions to meet appropriate standards, receive government support and be deployed at scale.

He argued that Nigerian innovators should be given opportunities to develop solutions for domestic health challenges, with the potential to expand into other African markets.

The NCDC official emphasised that regulation, standards and adoption must work together to ensure that useful innovations can move beyond the development stage to deliver measurable benefits.

On her part, multiple award winning journalist and Editor, Health and Science Africa, Ojoma Akor,urged stakeholders to embrace the technology, describing the forthcoming Artificial Intelligence in Health Summit as a timely platform for exploring how AI-driven innovations could address some of the sector’s most pressing challenges.

Akor, who is a member of the summit’s Technical Advisory Committee, said the technology offered significant opportunities for Nigeria and other African countries to improve healthcare outcomes, particularly through innovations capable of making medical procedures less burdensome and more efficient.
She stressed the need for media practitioners, healthcare professionals, researchers and policymakers to engage actively with emerging AI applications and examine their potential to transform healthcare delivery.

“AI in health is one key area we should start paying attention to, especially given the spate of challenges facing the health sector. AI offers hope in a whole lot of areas,” she said.

According to her, the summit will provide an opportunity for stakeholders to examine ongoing interventions, showcase innovations and discuss how artificial intelligence can be deployed to strengthen public health systems across Nigeria and the continent.

She disclosed that representatives of major health institutions and organisations, including the World Health Organisation (WHO), Africa Centres for Disease Control and Prevention (Africa CDC) and Clinton Health Access Initiative (CHAI), are expected to share their perspectives on the application of AI to public health.

Beyond discussions, Akor said participants would also be exposed to practical innovations demonstrating how AI could improve medical procedures and enhance healthcare experiences for patients.

Using medical imaging as an example, she explained that AI-assisted technologies could help simplify procedures such as Magnetic Resonance Imaging (MRI), which many patients find physically and psychologically demanding.

She noted that innovations combining AI with medical imaging had demonstrated the technology’s potential to improve aspects of healthcare delivery, adding that such developments deserved greater attention from Nigerian health stakeholders.

“For those who have gone for an MRI, you know it can be a daunting experience. But with AI, from what some of the innovators have demonstrated, it is something that could become a lot easier,” she said.

Akor argued that the potential benefits of AI extended beyond individual medical procedures, urging stakeholders to consider what the technology could achieve for Nigeria and Africa if effectively integrated into healthcare systems.

She also drew parallels between the growing use of AI in journalism and its potential applications in medicine, noting that media practitioners were already experiencing how technology could support professional work.

Having witnessed AI’s growing relevance in journalism, she said, stakeholders should be equally interested in exploring its capacity to improve healthcare delivery, particularly in countries facing significant health system constraints.

She therefore called on journalists and other stakeholders to participate in the summit, contribute to discussions and examine the opportunities and challenges associated with deploying AI in public health.

Akor said the gathering would offer participants an opportunity to learn from institutions and innovators already working at the intersection of artificial intelligence and healthcare, while encouraging broader conversations about the technology’s future in the region.
Technology law counsel with AHFID/AI4SID and coordinator of the summit, Liss Yori, said the responsible use of AI in healthcare must begin with protecting the interests of patients.

She warned that poorly designed or inadequately regulated applications could expose sensitive health information, misrepresent patients or generate recommendations that could harm those seeking treatment.

Yori said collaboration among government agencies, technology developers, healthcare professionals and other stakeholders was necessary to ensure that patient welfare remained central to the development and deployment of AI tools.

She stressed that the ultimate measure of success should not be the number of AI applications developed, but whether the technology improves the quality, safety and accessibility of healthcare.