Recalibration of a Deep Learning Model for Low-Dose Computed Tomographic Images to Inform Lung Cancer Screening Intervals

• Published in: JAMA network open 2023-03, Vol.6 (3), p.e233273-e233273
• Main Authors: Landy, Rebecca, Wang, Vivian L, Baldwin, David R, Pinsky, Paul F, Cheung, Li C, Castle, Philip E, Skarzynski, Martin, Robbins, Hilary A, Katki, Hormuzd A
• Format: Article
• Language: English
• Subjects: Deep Learning; Early Detection of Cancer - methods; Female; Humans; Lung - diagnostic imaging; Lung - pathology; Lung cancer; Lung Neoplasms - pathology; Male; Medical diagnosis; Medical screening; Middle Aged; Oncology; Online Only; Original Investigation; Tomography, X-Ray Computed - methods
• ISSN: 2574-3805; 2574-3805
• DOI: 10.1001/jamanetworkopen.2023.3273

Annual low-dose computed tomographic (LDCT) screening reduces lung cancer mortality, but harms could be reduced and cost-effectiveness improved by reusing the LDCT image in conjunction with deep learning or statistical models to identify low-risk individuals for biennial screening. To identify low-risk individuals in the National Lung Screening Trial (NLST) and estimate, had they been assigned a biennial screening, how many lung cancers would have been delayed 1 year in diagnosis. This diagnostic study included participants with a presumed nonmalignant lung nodule in the NLST between January 1, 2002, and December 31, 2004, with follow-up completed on December 31, 2009. Data were analyzed for this study from September 11, 2019, to March 15, 2022. An externally validated deep learning algorithm that predicts malignancy in current lung nodules using LDCT images (Lung Cancer Prediction Convolutional Neural Network [LCP-CNN]; Optellum Ltd) was recalibrated to predict 1-year lung cancer detection by LDCT for presumed nonmalignant nodules. Individuals with presumed nonmalignant lung nodules were hypothetically assigned annual vs biennial screening based on the recalibrated LCP-CNN model, Lung Cancer Risk Assessment Tool (LCRAT + CT [a statistical model combining individual risk factors and LDCT image features]), and the American College of Radiology recommendations for lung nodules, version 1.1 (Lung-RADS). Primary outcomes included model prediction performance, the absolute risk of a 1-year delay in cancer diagnosis, and the proportion of people without lung cancer assigned a biennial screening interval vs the proportion of cancer diagnoses delayed. The study included 10 831 LDCT images from patients with presumed nonmalignant lung nodules (58.7% men; mean [SD] age, 61.9 [5.0] years), of whom 195 were diagnosed with lung cancer from the subsequent screen. The recalibrated LCP-CNN had substantially higher area under the curve (0.87) than LCRAT + CT (0.79) or Lung-RADS (0.69) to predict 1-year lung cancer risk (P