Histological proven AI performance in the UKLS CT lung cancer screening study: Potential for workload reduction

Authors
Harriet L. Lancaster, Beibei Jiang, Michael P.A. Davies, Jan Willem C. Gratama, Mario Silva, Jaeyoun Yi, Marjolein A. Heuvelmans, Geertruida H. de Bock, Anand Devaraj, John K. Field, Matthijs Oudkerk
Journal
European Journal of Cancer
Related Product

LCS

Date Published
2025.02
Summary

This study validated the AI software AVIEW LCS version 1.1.39 using the UK lung cancer screening (UKLS) CT dataset to assess its potential for lung cancer detection and reducing reading workload. The AI analyzed baseline CT scans from 1,252 individuals and was compared against human readers and histological lung cancer diagnoses. AI demonstrated fewer misclassifications than human readers and achieved a sensitivity of 99.8%, detecting 31 cases of lung cancer. However, one case was misclassified as negative due to the application of the 100mm³ threshold. Using AI as a first reader was estimated to reduce CT reading workload by up to 79%. The study results indicate that a rule-out approach leveraging AI can enhance the efficiency of lung cancer screening and alleviate the workload of radiologists.

Contact

Please leave your inquiry if you have any questions regarding our products, recruitment, investment, or any other matters.

Contact us