Artificial intelligence in Nigerian oncology practice: A qualitative exploration of oncologists' perspectives

Abstract

BACKGROUND: Artificial intelligence (AI) offers potential solutions to address critical challenges in oncology practice, particularly in resource-constrained settings like Nigeria. However, successful implementation requires understanding healthcare providers' perspectives, which remain largely unexplored in the Nigerian context. AIM: To explore Nigerian oncologists' perspectives on AI applications in oncology practice, identifying knowledge levels, perceived benefits, implementation barriers, and priority areas for AI integration. METHODS: This qualitative study employed a descriptive exploratory design. Semi-structured interviews were conducted with 15 oncologists from nine major Nigerian healthcare institutions. All interviews were conducted in English. These institutions represent tertiary referral centres predominantly located in urbanised areas across different Nigerian geopolitical zones, including Southwest (OAUTH, LUTH, UCH, LASUTH, LAUTH), South-South (ISTH, UBTH), and North-Central (BSUTH, UATH). Participants represented various oncology specialties with experience ranging from 1 to 20 + years. Data were analysed using Braun and Clarke's six-phase thematic analysis approach with independent coding by multiple researchers to ensure inter-coder reliability. RESULTS: Nine key themes emerged: (1) Current Knowledge and Awareness of AI in Oncology; (2) Perceived Benefits of AI in Oncology Practice; (3) Perceived Barriers to AI Implementation; (4) AI in Oncology Research; (5) Data Management and Ethical Concerns; (6) Trust and Adoption Readiness; (7) Human-AI Interaction and Patient Dynamics; (8) Future Directions and Knowledge Requirements; and (9) Resource Allocation and Infrastructure Development. Participants demonstrated limited theoretical knowledge of AI applications, with most lacking practical implementation experience. Participants recognised AI's potential to address workforce shortages and improve diagnostic accuracy but identified significant barriers including financial constraints, infrastructure limitations, and insufficient technical expertise. CONCLUSION: Nigerian oncologists expressed cautious optimism about AI's potential to transform cancer care delivery despite substantial implementation challenges. Successful AI integration requires addressing infrastructure deficits, developing appropriate regulatory frameworks, and building technical capacity. A phased implementation approach focusing initially on diagnostic support applications is recommended, alongside sustained investment in digital infrastructure and workforce development.

Ähnliche Arbeiten