AI Use in Breast Cancer Detection May Not Be Ready For Clinical Settings
“Current evidence for AI does not yet allow judgement of its accuracy in breast cancer screening programmes, and it is unclear where on the clinical pathway AI might be of most benefit,” according to a literature review published September 2, 2021, in the BMJ.
The team, led by Dr. Karoline Freeman of the University of Warwick in England, reviewed 12 studies. They determined that 34 of the 36 AI systems the studies evaluated were less accurate than a single reader, and all were less accurate than the consensus of two or more radiologists. They further found that encouraging results from smaller studies were not replicated in larger studies. Of the 12 studies, which screened among them 131, 822 women, “[n]o prospective studies measuring test accuracy of AI in screening practice were found. Studies were of poor methodological quality.” according to the team.
Freeman, et al., concluded that further studies are needed. “Prospective studies are required to measure the effect of AI in clinical practice. Such studies will require clear stopping rules to ensure that AI does not reduce programme specificity,” they wrote.
Freeman K, Geppert J, Stinton C, Todkill D, Johnson S, Clarke A et al. “Use of artificial intelligence for image analysis in breast cancer screening programmes: systematic review of test accuracy.” BMJ 2021; 374 :n1872 doi:10.1136/bmj.n1872 Published online September 2, 2021. Accessed September 15, 2021, at https://www.bmj.com/content/374/bmj.n1872