Artificial Intelligence Integration in Radiation Therapy Education: A Multi-Modal Approach

Critical workforce shortages in radiation oncology have led tertiary institutions to rapidly expand their radiation therapy (RT) student cohorts. The increase in students entering university created the challenge of scaling educational delivery while preserving the quantity and quality of learning essential for developing well-prepared, competent clinicians. Artificial Intelligence (AI) applications have been recognised as valuable tools in medical education; however, limited research has addressed their use in RT education. This study aimed to evaluate three AI educational innovations used in the RT degree at the University of Newcastle.

A cross-sectional study design was implemented to investigate the perceptions of students toward the integration of three AI educational innovations, which were embedded in the RT program curriculum across Years 1, 2 and 3. The three innovations included: (1) delivery innovation (AI video lectures), (2) assessment innovation (AI-assisted assessment feedback), and (3) content innovation (AI-simulated communication tasks). Descriptive statistics were calculated for quantitative survey responses. Open-ended responses were analysed to find recurring themes.

A total of 62 students participated in the study across three cohorts (Years 1 (n = 33), 2 (n = 13) and 3 (n = 16)), with a mean age of 20.6 years. In Year 1, n = 24 (73%) of students reported being 'satisfied' or 'very satisfied' with the AI video format. Among Year 2 students, n = 7 (54%) wanted AI feedback on future assessments, while n = 6 (46%) were unsure or opposed to future AI feedback. In Year 3, n = 12 (75%) felt 'more' or 'much more comfortable' practising with the AI patient than with peers.

This research revealed varied outcomes related to AI innovations among year groups. The integration of AI was perceived positively; however, participants favoured AI tools for formative learning over those used for assessments.