Exploring patient and clinician opinions, perspectives and acceptance of the use of artificial intelligence in the histological diagnosis of prostate cancer

Objectives: This study aims to explore the opinions and attitudes of patients and clinicians regarding the use of artificial intelligence (AI) in the diagnosis of prostate biopsies, with a focus on acceptance and trust in the use of AI, and factors that may impact this. Subjects and methods: Surveys were sent to patient members of UK-based prostate cancer support groups and to a group of clinicians managing patients with prostate cancer (or suspected prostate cancer). Results: Of 130 patient respondents, 94% expressed acceptance of AI assistance in the diagnosis of prostate biopsies when a pathologist retains responsibility for the final diagnosis, but regard it as the responsibility of the pathologist to decide whether AI is used in this setting. Similar responses were noted among the nine clinician respondents. Regarding factors with potential impact on acceptance of AI, an understanding of how the AI was tested and its performance in comparison with a pathologist was considered to be more important than how the technology was developed, and half (51%) of patients suggested that access to additional information might impact their acceptance of AI. Conclusion: Understanding the perspectives of stakeholders is key to the successful clinical implementation of AI in the histological diagnosis of prostate biopsies. Our study shows a high level of acceptance of AI for the diagnosis of prostate biopsies among patients if a pathologist retains oversight of the diagnosis and the decision as to when AI is used. Furthermore, it suggests similar levels of acceptance among clinicians. Our study provides insight into areas for educational focus to enhance understanding of AI in this setting.