Artificial intelligence for healthcare in Africa: a scientometric analysis

Abstract

Introduction: Artificial intelligence (AI) has greatly transformed healthcare in developed countries. However, there is limited data describing the extent of AI adoption in African healthcare systems. The aim of this study was to understand the state of AI healthcare research in Africa. Methods: A scientometric analysis was conducted to visualize the state-of-the-art research of AI in healthcare in Africa. Results: Twenty-six relevant articles, published by 178 authors and affiliated with 96 organizations in 31 countries, were included. The most prolific African countries were South Africa, followed by Nigeria and Ghana. Some articles were published by authors affiliated with non-African countries. None of the contributing authors published more than 2 articles. Only 20 (11.2%) authors collaborated, forming a single cluster. The most common AI tools used in African health systems were deep learning neural networks applied in medical imaging, Adaptive Neuro-Fuzzy Inference Systems, and E-algorithms. Conclusion: Our results suggest that AI for healthcare in Africa is still in its developmental phase with limited published research. Our social network analysis highlighted a South and West African predominance in the research relational network of AI in healthcare. This discrepancy presents an opportunity for coordination and increased collaboration with healthcare institutions advanced in the use of AI within Africa and beyond.