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

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

doi:10.1016/j.ejca.2025.115324

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

Eur J Cancer. 2025 May 2:220:115324. doi: 10.1016/j.ejca.2025.115324. Epub 2025 Feb 25.

Authors

Affiliations

¹ Department of Epidemiology, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands; Institute for Diagnostic Accuracy, Groningen, the Netherlands.
² Institute for Diagnostic Accuracy, Groningen, the Netherlands.
³ Molecular and Clinical Cancer Medicine, University of Liverpool, Liverpool, United Kingdom.
⁴ Department of Radiology and Nuclear Medicine, Gelre Hospitals, Apeldoorn, the Netherlands.
⁵ Scienze Radiologische, Department of Medicine and Surgery (DiMeC), University of Parma, Parma, Italy; Department of Radiology, University of Massachusetts Memorial Health, University of Massachusetts, Chan Medical School, Worcester, MA, United States.
⁶ Coreline Soft, Seoul, Republic of Korea.
⁷ Department of Epidemiology, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands; Institute for Diagnostic Accuracy, Groningen, the Netherlands; Department of Respiratory Medicine, University of Amsterdam, Amsterdam University Medical Centers, Amsterdam, the Netherlands.
⁸ Department of Epidemiology, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands.
⁹ Royal Brompton Hospital London, Sydney St, Chelsea, London, United Kingdom; National Heart and Lung Institute, Imperial College, London, United Kingdom.
¹⁰ Molecular and Clinical Cancer Medicine, University of Liverpool, Liverpool, United Kingdom. Electronic address: J.K.Field@liverpool.ac.uk.
¹¹ Institute for Diagnostic Accuracy, Groningen, the Netherlands; Faculty of Medical Sciences, University of Groningen, Groningen, the Netherlands. Electronic address: m.oudkerk@rug.nl.

PMID: 40022836
DOI: 10.1016/j.ejca.2025.115324

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.

Keywords: Artificial intelligence; CT; Lung cancer; Lung nodule; Screening.

MeSH terms

Aged
Artificial Intelligence*
Early Detection of Cancer* / methods
Female
Humans
Lung Neoplasms* / diagnostic imaging
Lung Neoplasms* / pathology
Male
Middle Aged
Radiographic Image Interpretation, Computer-Assisted / methods
Tomography, X-Ray Computed* / methods
United Kingdom
Workload*