In a world where 1.25million children are infected with Tuberculosis (TB) and 214,000 die from it annually, one Nigerian physician is helping children breathe easier, one X-ray at a time.

Dr. Victor Anyebe, a seasoned research physician and global health innovator, recently concluded his role as a Consultant Scientist with the Foundation for Innovative New Diagnostics (FIND) in Geneva. But what he achieved in just under a year will likely shape the future of pediatric tuberculosis (TB) diagnostics for decades to come.

Between October 2022 and September 2023, Dr. Anyebe led the scientific execution of the CAPTURE Project, a global initiative designed to overcome one of medicine’s most persistent blind spots: the accurate and early diagnosis of TB in children. While TB has long been treatable, its detection, especially in young patients, remains stubbornly unreliable in many parts of the world, with nearly 50% of cases missed annually. The CAPTURE Project aimed to change that. And Dr. Anyebe was at the heart of it.

“TB in children is often underdiagnosed, overlooked, or misinterpreted due to the limitations of conventional diagnostic methods,” he said. “We set out to develop a machine learning solution that could shift this paradigm.”

The CAPTURE Project wasn’t just ambitious; it was unprecedented. It aimed to create a comprehensive repository of 8,400 pediatric chest X-ray images, encompassing a diverse spectrum of clinical cases and geographic regions. These images would then be used to train a computer-aided diagnostic (CAD) algorithm to screen for TB in children more accurately, consistently, and affordably than human eyes alone and across hard-to-reach locations.

For this to work, Dr. Anyebe had to tackle multiple scientific, ethical, and logistical challenges at once.
“We weren’t just collecting data,” he explains. “We were building a global research collaboration based on trust, quality, and precision.”

This meant designing the right protocols, curating diverse patient samples across five continents, and ensuring legal and ethical compliance in a domain where patient privacy and pediatric data carry enormous sensitivity.

What set Dr. Anyebe apart in this project wasn’t just his deep understanding of diagnostics and machine learning; it was his ability to bring institutions across 12 countries into alignment.

Acting as a global diplomat and scientific lead, he orchestrated the signing of more than 25 Data Sharing Agreements (DSAs). These weren’t just routine contracts; they were the bedrock of a data-sharing framework that balanced institutional autonomy with global collaboration.
“Every dataset is a story of a hospital, a child, a community. My job was to make sure those stories were honored, protected, and shared for the greater good.”

The partnerships stretched across Africa, Asia, Europe, North America, and Latin America, encompassing children’s hospitals, national TB programs, academic labs, and AI development firms. In each case, Dr. Anyebe ensured that legal, cultural, and clinical standards were upheld, without compromising scientific utility. Periodic newsletters were developed and distributed to keep all collaborators abreast of progress, challenges, and collective strategies to resolve them.

This global orchestration required more than logistical skill—it demanded a leadership style that was as inclusive as it was visionary.

Once the data-sharing frameworks were secured and image curation underway, Dr. Anyebe turned his attention to the intellectual scaffolding of the project: drafting the scientific protocol manuscript.

Tasked with producing a high-impact, peer-reviewed publication that would not only validate the project but serve as a methodological model for future diagnostics research, he led the writing process from hypothesis formulation to final editing of the first draft.

“We needed to write something that could outlive the project,” he said. “A resource for scientists, developers, and policymakers alike.”
The manuscript captured the nuanced technical decisions behind dataset balancing, labeling protocols, and algorithm training parameters, all while embedding the bioethical considerations essential to pediatric research. Contributions were also made by remarkable scientists on the project, including Megan Palmer, James Seddon, and Devan Jaganath from Stellenbosch University, Imperial College London, and UCSF Benioff Children’s Hospital, respectively, among others.

The result? A document that is now under submission to a leading medical journal and is already receiving widespread review among AI developers and health ministries as a potential gold standard for CAD projects. It is expected that the validation results of this project will strengthen existing evidence for its recommendation for use in children below 15 by the WHO.

TB remains one of the top 10 causes of death among children globally, with over 1 million pediatric cases annually, 50% often missed, and significant underreporting. Conventional methods, like sputum testing, are often ineffective or infeasible in children, particularly in low-resource settings.

Dr. Anyebe and his collaborators recognized this diagnostic gap as both a clinical emergency and a technological opportunity.
“Children do not present with TB the way adults do,” he notes. “You can’t solve that with guesswork. You need data. You need algorithms trained specifically on pediatric cases.”

And that’s exactly what CAPTURE was designed to do: develop an AI diagnostic tool that doesn’t just guess but learns. With its pediatric chest X-ray repository, the project now provides the largest and most diverse dataset of its kind, making it possible to train and test CAD systems with a level of reliability never achieved in this demographic.

Though the project was headquartered in Geneva, Dr. Anyebe’s leadership brought a uniquely African perspective to the effort, one rooted in equity, accessibility, and the urgent realities of healthcare systems where pediatric TB can mean a death sentence. His experience addressing TB case finding in one of the world’s high-TB-burdened countries offered a lot of reflection and perspective.

His Nigerian medical training and years in public health gave him a sensitivity to the real-world application of AI tools in clinics with limited electricity, bandwidth, or radiological expertise.

“Technology that doesn’t work in Nigeria isn’t global technology,” he often says. “Our problems are different, but solvable if designed inclusively.” Local solutions with global applicability will change the world.

As the CAPTURE project moves into its next phase, testing and refining the diagnostic model, Dr. Anyebe’s work stands as a testament to what is possible when science, ethics, and leadership converge.

His journey also signals a broader truth: The future of health innovation doesn’t just lie in Silicon Valley or Geneva boardrooms; it lives in the heart, mind, and vision of people like Victor Anyebe, who can connect global standards with local needs.

Whether it’s a child in Nairobi, a clinician in Dhaka, or a policymaker in Abuja, the impact of the CAPTURE project will be felt widely, tangibly, and enduringly.

And behind every pixel of those 8,400 X-rays eventually collected? The fingerprint of a scientist who believes that every child deserves a diagnosis, and every diagnosis deserves precision.