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

Collapse authors list.
Lancaster, HL ORCID: 0000-0003-3025-9745, Jiang, B, Davies, MPA ORCID: 0000-0002-7609-4977, Gratama, JWC ORCID: 0000-0001-7483-4997, Silva, M ORCID: 0000-0002-2538-7032, Yi, J ORCID: 0000-0002-7664-9493, Heuvelmans, MA, de Bock, GH ORCID: 0000-0003-3104-4471, Devaraj, A, Field, JK ORCID: 0000-0003-3951-6365 et al (show 1 more authors) and Oudkerk, M ORCID: 0000-0003-2800-4110 (2025) Histological proven AI performance in the UKLS CT lung cancer screening study: Potential for workload reduction European Journal of Cancer, 220. 115324-. ISSN 0959-8049, 1879-0852

PDF Histological proven AI performance in the UKLS CT lung cancer screening.pdf - Open Access published version Download (6MB) | Preview

Abstract

Purpose: Artificial intelligence (AI) could reduce lung cancer screening computer tomography (CT)-reading workload if used as a first-reader, ruling-out negative CT-scans at baseline. Evidence is lacking to support AI performance when compared to gold-standard lung cancer outcomes. This study validated the performance of a commercially available AI software in the UK lung cancer screening (UKLS) trial dataset, with comparison to human reads and histological lung cancer outcomes, and estimated CT-reading workload reduction. Methods: 1252 UKLS-baseline-CT-scans were evaluated independently by AI and human readers. AI performance was evaluated on two-levels. Firstly, AI classification and individual reads were compared to a EU reference standard (based on NELSON2.0-European Position Statement) determined by a European expert panel blinded from individual results. A positive misclassification was defined as a nodule positive read ≥ 100mm³ and no/<100mm³ nodules in the expert read; A negative misclassification was defined as a nodule negative read, whereas an indeterminate or positive finding in the expert read. Secondly, AI nodule classification was compared to gold-standard histological lung cancer outcomes. CT-reading workload reduction was calculated from AI negative CT-scans when AI was used as first-reader. Results: Expert panel reference standard reported 815 (65 %) negative and 437 (35 %) indeterminate/positive CT-scans in the dataset of 1252 UKLS-participants. Compared to the reference standard, AI resulted in less misclassification than human reads, NPV 92·0 %(90·2 %-95·3 %). On comparison to gold-standard, AI detected all 31 baseline-round lung cancers, but classified one as negative due to the 100mm³ threshold, NPV 99·8 %(99·0 %-99·9 %). Estimated maximum CT-reading workload reduction was 79 %. Conclusion: Implementing AI as first-reader to rule-out negative CT-scans, shows considerable potential to reduce CT-reading workload and does not lead to missed lung cancers.

Item Type:	Article
Uncontrolled Keywords:	Lung cancer, Lung nodule, CT, Screening, Artificial intelligence
Divisions:	Faculty of Health & Life Sciences Faculty of Health & Life Sciences > Inst. Systems, Molec & Integrative Biology > Inst. Systems, Molec & Integrative Biology
Depositing User:	Symplectic Admin
Date Deposited:	20 Mar 2025 10:28
Last Modified:	28 Feb 2026 01:06
DOI:	10.1016/j.ejca.2025.115324
Related Websites:	Author Publisher
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3190893
Disclaimer:	The University of Liverpool is not responsible for content contained on other websites from links within repository metadata. Please contact us if you notice anything that appears incorrect or inappropriate.