Artificial Intelligence in Nuclear Medicine: From Image Reconstruction to Clinical Decision Support

Abstract

This review aimed to answer the PICO question: In patients undergoing nuclear medicine imaging, does the use of artificial intelligence for image reconstruction and clinical decision support, compared to traditional methods, improve diagnostic accuracy and clinical workflow efficiency? Google Scholar was searched to identify the relevant papers, and the PRISMA flow process was used to screen and select 15 papers, based on certain inclusion and exclusion criteria, for this review. The review showed that AI tools help in image reconstruction, improve diagnostic accuracy, clinical workflow efficiency and clinical decision support. AI also helps to provide an inclusive and equitable personalised care in different populations. Some AI tools may be better than others; therefore, the choice of the most appropriate AI tool for a given context is important. AI tools facilitate decision-making by multi-disciplinary teams and in preventive therapies, especially when considering pandemics like COVID-19. Some limitations of this review and the scope for future research have been indicated.