Artificial intelligence in paediatric cancer: Insights from innovation experts in the UNICA4EU project

This article presents the results of interviews with AI development and innovation experts, focusing on the potential contribution of AI to paediatric cancer treatment, its barriers and facilitators. AI-based technologies are expanding in health care provision and potentially to paediatric oncology, particular imageology. However, no AI based technology specifically developed for paediatric cancer has been identified. Interviews identified key barriers, including legal, regulatory and ethical challenges; validation and evaluation standards; integration with public healthcare systems; acceptance, explainability and trust of AI technologies; digital literacy and skills development; data management and privacy Protection; and promoting multidisciplinary collaboration. Facilitators largely coincide, including legal, regulatory and ethical considerations; data management and storage; building trust and ensuring privacy; engaging key stakeholders; promoting multidisciplinary collaboration through AI technologies; and education, training and innovation programmes. Barriers can be turned into facilitators if properly managed. These interviews were conducted under the EU funded project UNICA4EU (Towards a UNIque approach for artificial intelligence data-driven solutions to fight Childhood cAncer FOR EUrope)(1), which was implemented between 2022 and 2024. The goal of the overall project was to analyse the current landscape of Artificial Intelligence (AI) and to map applications to Childhood Cancer. This work further develops and confirms the results of a previous EU-funded pilot project, A crowdsourced ecosystem to fight childhood cancer (EU4CHILD)(1). [Display omitted] •AI applicable to paediatric cancer: Tumour screening, radio/imageology, genomics, pathology.•AI can support paediatric cancer if barriers are turned into facilitators.•Paediatric cancer Innovation roadmap should be codesigned with users, supporting integrated care.