Real-world before-and-after evaluation of AI support for lung cancer diagnosis at three US lung nodule clinics
4 affiliations
- 1Thoracic Surgery and Lung Center of Excellence, Jupiter Medical Center, Jupiter, Florida, United States of America
- 2Interventional Pulmonology, AtlantiCare Regional Medical Center, Atlantic City, New Jersey, United States of America
- 3Department of Medicine, Division of Pulmonary, Critical Care, and Sleep Medicine, University of Connecticut School of Medicine, Farmington, Connecticut, United States of America
- 4Optellum Ltd, Oxford, United Kingdom
Abstract
Background: Health systems and payers require evidence that artificial intelligence (AI)-enabled decision support improves care delivery. Integrating AI into lung nodule management pathways may streamline workflows, improve identification and triage of patients with pulmonary nodules, and enable earlier lung cancer diagnosis. Yet real-world evidence of clinical utility remains limited. This study evaluated the impact of implementing commercially available AI software to support the lung nodule and lung cancer care pathway across three US health systems. Materials and Methods: A real-world before-and-after study compared clinic activity during matched 12.5-month pre-implementation and post-implementation periods at three US lung nodule clinics. Primary outcomes were the number of new patients and the number of patients undergoing at least one invasive procedure. Secondary outcomes were numbers of lung cancers and early-stage (stages I–II) lung cancers, and time from baseline CT to diagnosis. Analyses used Wilcoxon signed-rank and Mann-Whitney U tests, with medians and interquartile ranges (IQRs) reported for continuous measures. Results: Across sites, new patients increased from 138 before to 232 after implementation, equivalent to 11.0 to 18.6 patients per month (p = 0.0002), with the median rising from 9 (IQR 7–14) to 16 (IQR 13–21) patients. Monthly patients undergoing at least one invasive procedure increased from 4.9 to 8.0 (p = 0.0085). Monthly cancers increased from 3.0 to 3.8 (p = 0.3242), while monthly early-stage cancers rose from 1.4 to 2.4 (p = 0.125). Median time to diagnosis was 58 days (IQR 33–158) before and 57 days (IQR 27–85) after implementation (p = 0.26), with IQR narrowing from 125 days to 58 days. Conclusion: Implementation of the AI-enabled software was associated with significantly increased throughput of lung nodule patients and procedural volume. These results, observed prospectively post-implementation, suggest AI-assisted identification and triage can expand access to timely specialist review.
Related
- Court-Documented Adverse Financial Events Before Cancer Diagnosis and Mortality Risk
- Exploring patterns of acute and primary care utilisation before diagnosis, during treatment and for 3 years postcompletion of surgery and adjuvant treatment in women with endometrial cancer: a protocol for a Victorian data linkage study
- Exploring stakeholder support for student-run free clinics to address gaps in non-communicable disease care in Ghana: An exploratory sequential mixed-methods study
- Tumor immune organoids support preclinical evaluation of cancer immunotherapies through comparative modelling and translational investigation
- Temporal EHR Models Detect Rare Disease Years Before Diagnosis
BibTeX
@article{lee2026realworld,
title = {{Real-world before-and-after evaluation of AI support for lung cancer diagnosis at three US lung nodule clinics}},
author = {Kenneth Adam Lee and Amit Borah and Garrett Fiscus and Lisa Petricca and Sarah Dean and Annika Rings and Catarina Santos and Lyndsey C. Pickup and Omar Ibrahim},
journal = {PLOS One},
year = {2026},
month = oct,
doi = {10.1371/journal.pone.0358655},
url = {https://doi.org/10.1371/journal.pone.0358655},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Real-world before-and-after evaluation of AI support for lung cancer diagnosis at three US lung nodule clinics
AU - Kenneth Adam Lee
AU - Amit Borah
AU - Garrett Fiscus
AU - Lisa Petricca
AU - Sarah Dean
AU - Annika Rings
AU - Catarina Santos
AU - Lyndsey C. Pickup
AU - Omar Ibrahim
PY - 2026
DA - 2026/10/14
JO - PLOS One
DO - 10.1371/journal.pone.0358655
UR - https://doi.org/10.1371/journal.pone.0358655
AB - Background: Health systems and payers require evidence that artificial intelligence (AI)-enabled decision support improves care delivery. Integrating AI into lung nodule management pathways may streamline workflows, improve identification and triage of patients with pulmonary nodules, and enable earlier lung cancer diagnosis. Yet real-world evidence of clinical utility remains limited. This study evaluated the impact of implementing commercially available AI software to support the lung nodule and lung cancer care pathway across three US health systems. Materials and Methods: A real-world before-and-after study compared clinic activity during matched 12.5-month pre-implementation and post-implementation periods at three US lung nodule clinics. Primary outcomes were the number of new patients and the number of patients undergoing at least one invasive procedure. Secondary outcomes were numbers of lung cancers and early-stage (stages I–II) lung cancers, and time from baseline CT to diagnosis. Analyses used Wilcoxon signed-rank and Mann-Whitney U tests, with medians and interquartile ranges (IQRs) reported for continuous measures. Results: Across sites, new patients increased from 138 before to 232 after implementation, equivalent to 11.0 to 18.6 patients per month (p = 0.0002), with the median rising from 9 (IQR 7–14) to 16 (IQR 13–21) patients. Monthly patients undergoing at least one invasive procedure increased from 4.9 to 8.0 (p = 0.0085). Monthly cancers increased from 3.0 to 3.8 (p = 0.3242), while monthly early-stage cancers rose from 1.4 to 2.4 (p = 0.125). Median time to diagnosis was 58 days (IQR 33–158) before and 57 days (IQR 27–85) after implementation (p = 0.26), with IQR narrowing from 125 days to 58 days. Conclusion: Implementation of the AI-enabled software was associated with significantly increased throughput of lung nodule patients and procedural volume. These results, observed prospectively post-implementation, suggest AI-assisted identification and triage can expand access to timely specialist review.
SN - 1932-6203
ER -