Software using artificial intelligence for nodule and cancer detection in CT lung cancer screening: systematic review of test accuracy studies

Authors
Julia Geppert, Asra Asgharzadeh, Anna Brown, Chris Stinton, Emma J Helm, Surangi Jayakody, Daniel Todkill, Daniel Gallacher, Hesam Ghiasvand, Mubarak Patel, Peter Auguste, Alexander Tsertsvadze, Yen-Fu Chen, Amy Grove, Bethany Shinkins, Aileen Clarke, Sian Taylor-Phillips
Journal
Lung Cancer
Related Product

LCS

Date Published
2024.09
Summary

Julia Geppert et al. conducted a systematic review of AI-based software for lung cancer screening, evaluating 11 studies with 19,770 participants. The analysis compared AI-assisted versus unaided CT readings for detecting pulmonary nodules and malignancies. AI significantly improved sensitivity (+5% to +20%) but reduced specificity (−3% to −8%), detecting an additional 150–750 cancers per million screenings. However, false-positive rates rose, leading to 59,700–79,600 unnecessary follow-ups. The AVIEW Lungscreen software was one of the platforms reviewed, showcasing AI's role in enhancing sensitivity and categorization efficiency. Despite promising results, all studies faced high bias risks and lacked real-world applicability, limiting generalizability. The review calls for prospective trials and advances in AI algorithms to minimize false positives, ensure clinical feasibility, and optimize lung cancer screening protocols. AI's integration into practice can revolutionize early detection but requires evidence-based strategies to address current limitations.

Contact

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

Contact us