EUVIMEDEUVIMED entdecken
Systematische ÜbersichtOpen Access · CC BY 4.0

Reference standard methodology in the clinical evaluation of AI chest X-ray algorithms for lung cancer detection: A systematic review

Sean F Duncan, Andrew C Kidd, Jesus Perdomo Lampignano, Paul Cannon, Mark Hall, David B Stobo, John D Maclay, Kevin G Blyth, David J Lowe

European Journal of Radiology · 2025-11 · 192 · 112409

DOI: 10.1016/j.ejrad.2025.112409 ↗

Kurz gefasst

Systematische Übersicht in der Zeitschrift European Journal of Radiology, 2025-11. Verfasst von Sean F Duncan und 8 weiteren Autor:innen. Beteiligte Einrichtung: Digital Health Validation Lab, University of Glasgow, Imaging Centre of Excellence, Langlands Dr, Glasgow G51 4LB, UK; The Beatson West of Scotland C…. Hauptthemen laut MEDLINE: Lung Neoplasms, Algorithms, Radiography, Thoracic, Artificial Intelligence. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 74 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: UK Research and Innovation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND: Lung cancer remains the leading cause of cancer death worldwide, with early diagnosis linked to improved survival. Artificial intelligence (AI) holds promise for augmenting radiologists' workflows in chest X-ray (CXR) interpretation, particularly for detecting thoracic malignancies. However, clinical implementation of this technology relies on robust and standardised reference standard methodology at the patient-level. PURPOSE: This systematic review aims to describe reference standard methodology in the clinical evaluation of CXR algorithms for lung cancer detection. MATERIALS AND METHODS: Searches targeted studies on AI CXR analysis across MEDLINE, Embase, CENTRAL, and trial registries. 2 reviewers independently screened titles and abstracts, with disagreements resolved by a 3rd reviewer. Studies lacking external validation in real-world cohorts were excluded. Bias was assessed using a modified QUADAS-2 tool, and data synthesis followed SWiM guidelines. RESULTS: 1,679 papers were screened with 46 papers included for full paper review. 24 different AI solutions were evaluated across a broad range of research questions. We identified significant heterogeneity in reference standard methodology, including variations in target abnormalities, reference standard modality, expert panel composition, and arbitration techniques. 25 % of reference standard parameters were inadequately reported. 66 % of included studies demonstrated high risk of bias in at least one domain. DISCUSSION: To our knowledge, this is the first systematic description of patient-level reference standard methodology in CXR AI analysis of thoracic malignancy. To facilitate translational progress in this field, researchers undertaking evaluations of diagnostic algorithms at the patient-level should ensure that reference standards are aligned with clinical workflows and adhere to reporting guidelines. Limitations include a lack of prospective studies.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Lung NeoplasmsAlgorithmsRadiography, ThoracicArtificial IntelligenceRadiographic Image Interpretation, Computer-AssistedHumansReference StandardsReproducibility of Results

Autor:innen und Institutionen

  • Digital Health Validation Lab, University of Glasgow, Imaging Centre of Excellence, Langlands Dr, Glasgow G51 4LB, UK; The Beatson West of Scotland Cancer Center, 1053 Great Western Rd, Glasgow G12 0YN, UK.
  • The Beatson West of Scotland Cancer Center, 1053 Great Western Rd, Glasgow G12 0YN, UK; School of Cancer Sciences, University of Glasgow, Garscube Estate, Switchback Road, Bearsden G61 1QH, UK.
  • University of Glasgow Library, Hillhead St, Glasgow G12 8QE, UK; Radiology Department, NHS Greater Glasgow and Clyde, 1345 Govan Rd, Glasgow G51 4TF, UK.
  • University of Glasgow Library, Hillhead St, Glasgow G12 8QE, UK.
  • Radiology Department, NHS Greater Glasgow and Clyde, 1345 Govan Rd, Glasgow G51 4TF, UK.
  • Department of Respiratory Medicine, Glasgow Royal Infirmary, 84 Castle St, Glasgow G4 0SF, UK.

Förderung

  • UK Research and Innovation

Zitieren

APA

Duncan, S. F., Kidd, A. C., Lampignano, J. P., Cannon, P., Hall, M., Stobo, D. B., Maclay, J. D., Blyth, K. G., & Lowe, D. J. (2025). Reference standard methodology in the clinical evaluation of AI chest X-ray algorithms for lung cancer detection: A systematic review. European Journal of Radiology, 192, 112409. https://doi.org/10.1016/j.ejrad.2025.112409

Vancouver

Duncan SF, Kidd AC, Lampignano JP, Cannon P, Hall M, Stobo DB, et al. Reference standard methodology in the clinical evaluation of AI chest X-ray algorithms for lung cancer detection: A systematic review. European Journal of Radiology. 2025;192:112409. doi:10.1016/j.ejrad.2025.112409

BibTeX
@article{Duncan2025,
  author = {Duncan, Sean F and Kidd, Andrew C and Lampignano, Jesus Perdomo and Cannon, Paul and Hall, Mark and Stobo, David B and Maclay, John D and Blyth, Kevin G and Lowe, David J},
  title = {Reference standard methodology in the clinical evaluation of AI chest X-ray algorithms for lung cancer detection: A systematic review},
  journal = {European Journal of Radiology},
  year = {2025},
  volume = {192},
  pages = {112409},
  doi = {10.1016/j.ejrad.2025.112409}
}
Kostenlos starten →