ESTRO 2024 - Abstract Book

S2766

Interdisciplinary - Health economics & health services research

ESTRO 2024

literature review and stakeholder interviews, using the Consolidated Framework for Implementation Research (CFIR) (10,11); c) Combining CFIR determinants and Expert Recommendations for Implementation Change (ERIC) strategies (12) to collaboratively define the implementation strategy including stakeholder feedback. Based on the institutional developed implementation strategy outlined above, a hands-on multidisciplinary workshop was organized with AI teams of eight Dutch RT centres to evaluate its applicability in other centres. Participants were informed about the methods of the implementation strategy development in the specific case setting of AI implementation. Subgroups were formed to create their own strategies for implementation of AI-based auto segmentation. The workshop was assessed for acceptability, appropriateness and feasibility of the proposed implementation strategy workshop (13). First, the stakeholder analyses generated a diverse group of internal stakeholders, including physicians, physicists, radiotherapy technicians (RTTs), the chief medical information officer, patient planners, physician assistants, management of patient care, IT, finance and purchasing department and the program manager for education. Important external stakeholders were patients and patient representatives. The primary barriers and facilitators for AI implementation, revolved around AI's perceived black-box nature, privacy and security concerns, data quality and availability, legal and legislative considerations, knowledge and comprehension of AI, trust in AI systems, stakeholder involvement, AI ethics, and the importance of multidisciplinary collaboration. Table 1 shows all barriers and facilitators mentioned more than once. Additionally, Table 1, juxtaposes these factors with those already integrated into existing recommendations for AI implementation in RT, thereby providing a clear perspective on the valuable insights that have been added. These insights served as the basis for crafting both a generic implementation strategy to prepare the organization for increased AI utilization and a project-specific strategy for AI implementation projects, as summarized in Table 2. Results: