Prospective evaluation of artificial intelligence (AI) applications for use in cancer pathways following diagnosis: a systematic review

Objectives To assess the clinical readiness and deployability of artificial intelligence (AI) through evaluation of prospective studies of AI in cancer care following diagnosis.Design We undertook a systematic review to determine the types of AI involved and their respective outcomes with a PubMed a...

Full description

Saved in:
Bibliographic Details
Main Authors: Ajay Aggarwal, Ophira Ginsburg, Richard Sullivan, Andrew Hope, Peng Yun Ng, Sheba Macheka
Format: Article
Language:English
Published: BMJ Publishing Group 2024-08-01
Series:BMJ Oncology
Online Access:https://bmjoncology.bmj.com/content/3/1/e000255.full
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Objectives To assess the clinical readiness and deployability of artificial intelligence (AI) through evaluation of prospective studies of AI in cancer care following diagnosis.Design We undertook a systematic review to determine the types of AI involved and their respective outcomes with a PubMed and Web of Science search between 1 January 2013 and 1 May 2023.15 articles detailing prospective evaluation of AI in postdiagnostic cancer pathway were identified.Setting The role of AI in cancer care has evolved in the face of ageing population, workforce shortages and technological advancement. Despite recent uptake in AI research and adoption, the extent to which it improves quality, efficiency and equity of care beyond cancer diagnostics is uncertain to date.Interventions We appraised all studies using Risk of Bias Assessment of Randomised Controlled Trials (ROB­2) and Risk of Bias In Non­randomised Studies- of Interventions (ROBIN­I) quality assessment tools, as well as implementational analysis concerning time, cost and resource, to ascertain the quality of clinical evidence and real-world feasibility of AI.Results The results revealed that the majority of AI oncological research remained experimental without prospective clinical validation or deployment. Most studies failed to establish clinical validity and to translate measured AI efficacy into beneficial clinical outcomes. AI research is limited by lack of research standardisation and health system interoperability. Furthermore, implementational analysis and equity considerations of AI were largely missing.Conclusion To overcome the triad of low-level clinical evidence, efficacy-outcome gap and incompatible research ecosystem for AI, future work should focus on multi-collaborative AI implementation research designed and conducted in accordance with up-to-date research standards and local health systems.
ISSN:2752-7948