We proudly continue our Academic Express section to advance oncology nursing through scholarly exchange.This edition features groundbreaking insights from Dr. Yingchun Zeng (National University of Singapore) on merging AI into cancer nursing. Her analysis reveals nursing’s "sweet spot" where AI enhances prediction, personalization, and patient experience while addressing critical implementation challenges.Discover how nurses can lead meaningful AI integration at the care frontier.
Artificial intelligence (AI), from being a “nice-to-have” technology to being a “need-to-have” technology that can augment nursing practice across the entire continuum of cancer care from early detection through survivorship and palliative care. However, the question is not whether we are going to use this technology; rather, how are we going to integrate this technology safely and meaningfully into nursing practice.
Where AI is already making a difference
In screening and early detection, AI can enhance image interpretation and help expand access in settings where specialist capacity is limited. In treatment planning, AI can support precision (e.g., surgical support, radiation planning, toxicity prediction) and help identify patients at higher risk of complications—creating opportunities for earlier nursing interventions and tailored education.
The nursing “sweet spot”: prediction, personalization, and patient experience
Many high-impact nursing activities depend on timely recognition of risk and early supportive care. AI-enabled prediction models may help identify patients more likely to develop complications (e.g., lymphedema, venous thromboembolism, delirium), allowing nurses to intensify monitoring, education, and preventive strategies earlier. In survivorship and rehabilitation, AI can help tailor symptom management and supportive programs, while emerging approaches (e.g., analysis of large volumes of patient-generated text from forums) can broaden our understanding of patient concerns at scale—complementing, not replacing, human qualitative insights.
What stays hard: validation, bias, and trust
AI is not automatically safe because it is “data-driven.” Models require rigorous clinical validation, careful attention to privacy and consent, and ongoing monitoring for algorithmic bias and unintended harms. A particularly important concern is that some imaging AI systems may infer sensitive attributes (such as race) in ways that are not obvious to clinicians—raising equity risks that oncology nurses must help surface and mitigate.
What nurses must lead next
To ensure AI strengthens—rather than fragments—care, oncology nurses should co-design AI tools around nursing workflows, define implementation outcomes (workload, safety, equity, experience), and advocate for governance that is transparent and patient-centered. The goal is not “AI everywhere,” but AI where it improves outcomes, protects dignity, and supports nursing judgment.
Source:
Wang M, Abu-Odah H, Zeng Y. Merging artificial intelligence into cancer nursing care: Current applications, challenges, and opportunities. Asia Pac J Oncol Nurs. 2025 Dec 30;13:100849. doi: 10.1016/j.apjon.2025.100849. PMID: 41585538; PMCID: PMC12825068.
Dr Yingchun Zeng
