Dr Paul Macharia, an honorary research associate at the University of Nairobi’s Faculty of Health Sciences.
Speaking during a luncheon dubbed “Promoting Artificial Intelligence in Health and Development” in Nairobi, Dr Paul Macharia, an honorary research associate at the University of Nairobi’s Faculty of Health Sciences and lead researcher in the project, said the project aims to revolutionise healthcare delivery by improving knowledge, skills and regulatory compliance while promoting health equity.
The luncheon was organised by the Editors Guild.
“The initiative, dubbed Meditron, is a suite of open-source LLMs specifically adapted for low-resource and humanitarian settings. Developed through collaboration with medical professionals worldwide and humanitarian organisations like the International Committee of the Red Cross (ICRC), Meditron is a student-led project that has been in development for over a year, leveraging advanced computing power to process expert-designed medical literature and clinical guidelines,” he said.
He continued: “The research has already achieved significant milestones, including the successful implementation of Meditron and the MOOVE evaluation system. MOOVE provides detailed assessments of the AI’s performance, focusing not just on accuracy but also on contextual appropriateness, empathy and alignment with best medical practices.”
Dr Macharia further stated that the team has refined the AI’s ability to provide relevant and precise medical information, through extensive research involving 48 healthcare providers and over 460 clinical questions across various specialties, including HIV, reproductive health, mental health, nutrition and Type 1 diabetes.
Additionally, he said that the stakeholder engagement through three workshops has played a crucial role in the discussions on priority areas in adolescent and child health. The study protocol has been submitted to the Kenyatta National Hospital-University of Nairobi (KNH-UoN) Ethics & Research Committee, ensuring ethical oversight in the research process.
Dr Macharia emphasised the importance of ensuring the tool is accurate and ethical, stating, “Our goal is to ensure that this AI tool is not only informative but also ethically sound and aligned with real-world clinical needs. We are committed to creating a solution that healthcare providers can trust.”
He added: “We are excited about the potential of this tool to bridge knowledge gaps and support medical professionals in delivering high-quality, evidence-based care. By ensuring regulatory compliance and fostering a user-centered approach, we aim to create an AI solution that truly serves the needs of healthcare providers and patients alike.”