Caroline M. Godfrey, Ashley A. Leech, Kevin C. McGann, Jinyi Zhu, Hannah N. Marmor, Sophia Pena, Lyndsey C. Pickup, Fabien Maldonado, Evan C. Osmundson, Stacie B. Dusetzina, Eric L. Grogan, Stephen A. Deppen
Abstract
Artificial intelligence-based radiomic approaches have been shown to accurately evaluate indeterminate pulmonary nodules. With the expansion of lung cancer screening and utilization of computed tomography imaging, indeterminate pulmonary nodules requiring diagnostic evaluation are increasingly common.
Introduction
Nearly 15 million Americans are now eligible for annual lung cancer screening with low dose computed tomography (CT) imaging following the 2021 expansion of lung cancer screening guidelines by the United States Preventive Services Task Force [1]. In addition to screening-identified pulmonary nodules, CT is increasingly used for medical diagnostic imaging, leading to a larger number of incidentally-identified pulmonary nodules [2].
Methods
We constructed a decision analysis model to compare IPN stratification performed by the clinician alone to stratification performed by the clinician with AI support from the payer perspective with a lifetime horizon (Fig 1).
Results
For the base case scenario, AI-supported pulmonary nodule risk stratification resulted in an increase of 0.03 life years as compared to clinician alone and an incremental cost of $114. With a 65% pre-test probability of malignancy, AI was cost-effective at a willingness to pay threshold (WTP) of $100,000/LYG with an incremental cost-effectiveness ratio (ICER) of $4,485/LYG (Table 3).
Discussion
As AI is increasingly woven into medical decision-making and these platforms come to market, it is important to evaluate their cost-effectiveness. This study suggests that AI-supported IPN risk stratification, at the observed accuracies from the previously published 12-reader LCP Score study, may be cost-effective in populations with an underlying malignancy rate of 5% or greater [9].
Conclusion
AI-assisted pulmonary nodule risk estimation is cost-effective in most common clinical settings where the malignancy prevalence is at least 5%. With high rates of lung resections performed on benign disease and the necessity of rapid identification of lung cancer to initiate treatment in its earliest, most treatable phase.
Citation: Godfrey CM, Leech AA, McGann KC, Zhu J, Marmor HN, Pena S, et al. (2026) Cost-effectiveness analysis of artificial intelligence-assisted risk stratification of indeterminate pulmonary nodules. PLoS One 21(3): e0343492. https://doi.org/10.1371/journal.pone.0343492
Editor: Jun Hyeok Lim, Inha University Hospital, KOREA, REPUBLIC OF
Received: July 31, 2025; Accepted: February 7, 2026; Published: March 5, 2026
Copyright: © 2026 Godfrey et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files. We have now made all of raw data available in our submission in our tables and figures, and included the full variable list file. Collectively, this provides all data required to duplicate our work. We have additionally included our actual TreeAge software, which if future users wanted/had access to the TreeAge software could be used to run our full model immediately. This is not necessary, however, as the complete raw data and variables list is also provided -- meaning the user could run the model on any software of their choosing.
Funding: Dr. Godfrey and Dr. McGann are supported by the National Institutes of Health (T32CA106183). Dr. Leech is supported by the National Institute on Drug Abuse of the National Institutes of Health (K01DA050740). Dr. Maldonado is supported by the National Institutes of Health (R01CA253923). Dr. Dusetzina is supported by the National Institutes of Health (2P30CA068485). Dr. Grogan is supported by the National Institutes of Health (U01CA152662, R01CA252964) and was previously a recipient of the Department of VA Health Services Research and Development Service Career Development Award (10-024). Dr. Deppen is supported by the National Institutes of Health (U01CA152662). Lyndsey C. Pickup is an employee of Optellum Ltd., a commercial company which produces an artificial intelligence radiomic tool for use in indeterminate pulmonary nodules. Optellum Ltd. provided support in the form of salary for this author [LCP] but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.
Competing interests: We have read the Journal’s policy, and the authors of this manuscript have the following competing interests: Lyndsey C. Pickup is an employee of Optellum Ltd., which produces an artificial intelligence radiomic tool for use in indeterminate pulmonary nodules. Optellum Ltd. provided support in the form of salary for this author [LCP] but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. This does not alter our adherence to PLOS ONE policies on sharing data and materials. The authors do not have any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products to declare.
