Application of artificial intelligence in the development of patient oriented oncology information.

Abstract

e13660 Background: *This abstract was prepared with the assistance of artificial intelligence to enhance clarity, organization, and accessibility.* Cancer diagnosis presents significant medical and emotional challenges, requiring clear communication to help patients and caregivers navigate complex information about treatments and outcomes. Effective communication improves understanding, engagement, and adherence to care, resulting in better outcomes. Artificial intelligence (AI) has shown potential in generating personalized educational materials tailored to literacy levels, cultural contexts, and emotional needs. However, AI cannot grasp emotional nuances, making the involvement of patients and caregivers essential to ensure information is empathetic, accurate, and effective. Methods: A structured approach assessed AI’s application in co-creating oncology information. The target audience included cancer patients and caregivers. Data were collected from evidence-based sources and qualitative feedback from patients and healthcare professionals to identify knowledge gaps and content preferences. An AI model utilizing natural language processing generated educational materials customized to literacy levels, cultural backgrounds, and emotional sensitivities. Patients participated in focus groups and iterative review cycles to refine content and ensure alignment with their needs. Evaluation included comprehension, user satisfaction, and practical outcomes like improved treatment adherence and reduced information gaps. Ethical considerations addressed data anonymization, transparency, and bias mitigation. Results: AI demonstrated significant impact in co-creating oncology information: • 40% improvement in comprehension, particularly about protocols and side effects. • 87% satisfaction with clarity, relevance, and empathetic tone. • 25% increase in treatment adherence. • 70% reduction in information gaps. Patients valued their involvement in the co-creation process, emphasizing how it ensured materials were accurate and supportive. These findings demonstrate AI’s potential to enhance patient education, engagement, and care adherence while reducing provider workload. Conclusions: AI can revolutionize oncology care by co-creating personalized educational materials that improve comprehension, reduce information gaps, and enhance treatment adherence. However, its inability to grasp emotional nuances reinforces the need to involve patients and caregivers in the co-creation process, ensuring materials are empathetic and supportive. By integrating AI with human insight, oncology care can become more inclusive and patient-centered. Future efforts should refine AI tools, address ethical concerns, and expand their application to diverse populations, advancing patient education and fostering compassionate communication in clinical practice.