Development and cost analysis of a lung nodule management strategy combining artificial intelligence and Lung-RADS for baseline lung cancer screening
Principal Investigator
Name
Paul Babyn
Degrees
M.D.,C.M.
Institution
University of Saskatchewan
Position Title
Professor and Physician Executive
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-722
Initial CDAS Request Approval
Oct 22, 2020
Title
Development and cost analysis of a lung nodule management strategy combining artificial intelligence and Lung-RADS for baseline lung cancer screening
Summary
Artificial intelligence (AI) algorithms have demonstrated high accuracy for lung nodule classification, in some cases surpassing that of radiologists unaided by AI. However, it remains unclear how malignancy risk prediction scores from AI algorithms should be combined with and incorporated into current lung cancer screening management approaches such as Lung-RADS, and what the impact on healthcare utilization and cost may be. In this study, we will develop a lung nodule management strategy combining Lung-RADS with an AI malignancy risk score and determine the impact of this management strategy on follow-up investigations and associated costs in a baseline lung cancer screening population.
Aims
(1) Develop a lung nodule management strategy combining Lung-RADS and an AI risk score to refine categorization at the time of baseline lung cancer screening; and
(2) Determine the potential impact of this strategy on recommendations for further investigations following baseline screening and the associated costs of recommended follow-up investigations from a public payer perspective.
Collaborators
Scott J. Adams, MD, University of Saskatchewan