Professor Patrick Ndayizigamiye is an Associate Professor at Wits Business School. He holds a PhD in Information Systems and Technology and has extensive experience in higher education across 5 institutions in South Africa and abroad. His expertise in the information systems domain brings together research in artificial intelligence, data science, digital health and organisational transformation. His current research agenda centres on responsible agentic artificial intelligence and AI orchestration for healthcare and organisations, with particular attention to African and other resource-constrained environments.
At the heart of this agenda is how organisations can use increasingly autonomous AI systems effectively while retaining meaningful human oversight and accountability. His focus on agentic AI orchestration expands on how AI agents, people, data and digital tools are coordinated within organisational processes: how tasks are delegated, decisions are checked, uncertainty is escalated and failures are managed. He is particularly interested in the interplay between the autonomy granted to an AI system and an organisation’s capacity to supervise, govern and sustain its use. Through this developing research programme, Professor Ndayizigamiye aims to produce evidence and practical frameworks that help organisations assess their readiness for agentic AI, design appropriate governance structures and evaluate its contribution to performance and service delivery. His interests encompass both the technical behaviour of AI systems and the organisational conditions that shape their implementation, including skills, infrastructure, leadership, trust and institutional capacity.
His recent collaborative research includes explainable machine learning for suicide-risk assessment, artificial intelligence in emergency care, deep learning for diabetes prediction, and digital approaches to maternal health information management. His research contributions extend to business applications, including machine learning to predict investment policy lapses. His postgraduate supervision covers organisational readiness for AI, generative and agentic AI implementation, digital platforms, customer analytics, automation and ethical risks in financial services.
Professor Ndayizigamiye has authored and co-authored more than 70 scholarly publications. His research draws on quantitative, qualitative and mixed methods, including machine learning, systematic reviews, bibliometric analysis and theory-informed studies of technology adoption and implementation. This methodological breadth enables him to engage with both the development of digital solutions and the human and organisational contexts in which they operate.
At Wits Business School, Professor Ndayizigamiye aims to connect responsible AI research with the decisions that managers, healthcare leaders and policymakers face. He welcomes research collaborations, postgraduate enquiries and opportunities for industry engagement around agentic AI orchestration, AI governance and digital transformation. His broader aim is to help organisations develop the knowledge and capabilities needed to translate advances in AI into accountable, sustainable and contextually relevant practice.
Research interests
Responsible agentic AI orchestration, including agent coordination, human oversight, accountability and evaluation.
Explainable artificial intelligence, machine learning and data-driven decision support.
Organisational readiness, governance and implementation of generative and agentic AI.
Responsible digital innovation in African and resource-constrained environments.
Digital health, health information systems and healthcare system resilience.
Digital business transformation, organisational capabilities and technology adoption.
Selected publications
Matlala, L., Bokaba, T., Ndayizigamiye, P., Mhlongo, S., & Dogo, E. (2026). A comparative analysis of ensemble learning models for predicting lapses in investment policies. Journal of Management Analytics, 13(1), 109–138. https://www.tandfonline.com/doi/full/10.1080/23270012.2025.2574030
Mokheleli, T., Makaba, T., Ndayizigamiye, P., & Ndlovu, N. (2026). Artificial intelligence approach for predicting suicide-related behaviour in emergency departments. Scientific Reports, p 1-44. https://www.nature.com/articles/s41598-026-64073-y
Mokheleli, T., Makaba, T., Ndayizigamiye, P., Ndlovu, N., & Twinomurinzi, H. (2026). Explainable Machine Learning for Suicide Risk Assessment Using Social Media Data. IEEE Access, 14, 77656–77676. https://ieeexplore.ieee.org/document/11523010
Sibanda, K., Ndayizigamiye, P., & Twinomurinzi, H. (2026). A non-fungible token (NFT)-Based framework for antenatal care in resource-constrained settings. Digital Health, 12. https://journals.sagepub.com/doi/10.1177/20552076261478842
Mokheleli, T., Makaba, T., Ndayizigamiye, P., Ndlovu, N., & Twinomurinzi, H. (2026). Artificial intelligence in suicide risk assessment: a systematic literature review. Discover Artificial Intelligence, 6, Article 296. https://link.springer.com/article/10.1007/s44163-026-01031-7
Nkhwashu, J., Ofusori, L., Ndayizigamiye, P., & Makaba, T. (2026). The Use of Machine Learning in Emergency Care Units: A Systematic Review. Journal of Primary Care & Community Health, 17. https://journals.sagepub.com/doi/10.1177/21501319251414821
Patrick Ndayizigamiye