Artificial intelligence-assisted radiotherapy for pelvic and abdominal malignancies: assessing feasibility in the context of Africa-specific risks

Résumé

Cancer care in Africa remains severely limited, with pelvic and abdominal malignancies contributing substantially to the disease burden. Radiotherapy is essential but constrained by infrastructure deficits, workforce shortages, and systemic inequities. Artificial intelligence (AI) may help strengthen radiotherapy through automation and improved workflow efficiency. This narrative review summarises current evidence on AI assisted radiotherapy for pelvic and abdominal cancers in Africa, highlighting feasibility, and regional specific implementation risks. The review shows that AI tools for auto-contouring, treatment planning support, quality assurance, and workflow optimisation can improve efficiency and ease workload when implemented within appropriate clinical and governance frameworks. Their clinical impact in African radiotherapy, however, is constrained by limited digital infrastructure, workforce shortages, weak data governance, regulatory gaps, and poor model generalisability. Additional risks including data bias from non-African training datasets, and fragile IT systems underscore the need for cautious deployment. A feasibility-first, phased adoption strategy centred on hybrid AI-human workflows, regional model validation, workforce upskilling, and policy-led governance offers a safe and practical route for integrating AI into African radiotherapy. When integrated within resilient systems and guided by risk-aware strategies, AI has the potential to act as a capacity multiplier rather than a substitute, offering a more equitable access to high quality radiotherapy across Africa.

Citer ce document

Accès au document

Texte intégral en lecture en ligne, réservé aux abonnés SPHAERO et aux membres de l'institution. Se connecter

Auteur(s)

Statistiques

Consultations : 1

Téléchargements : 0