Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Multicenter Study
. 2024 Feb;165(2):371-380.
doi: 10.1016/j.chest.2023.10.012. Epub 2023 Oct 14.

A Deep Learning-Based Radiomic Classifier for Usual Interstitial Pneumonia

Jonathan H Chung  1 , Lydia Chelala  1 , Janelle Vu Pugashetti  2 , Jennifer M Wang  2 , Ayodeji Adegunsoye  3 , Alexander W Matyga  1 , Lauren Keith  4 , Kai Ludwig  4 , Sahar Zafari  4 , Sahand Ghodrati  5 , Ahmadreza Ghasemiesfe  5 , Henry Guo  6 , Eleanor Soo  7 , Stephen Lyen  7 , Charles Sayer  7 , Charles Hatt  4 , Justin M Oldham  8
Affiliations
Multicenter Study

A Deep Learning-Based Radiomic Classifier for Usual Interstitial Pneumonia

Jonathan H Chung et al. Chest. 2024 Feb.

Abstract

Background: Because chest CT scan has largely supplanted surgical lung biopsy for diagnosing most cases of interstitial lung disease (ILD), tools to standardize CT scan interpretation are urgently needed.

Research question: Does a deep learning (DL)-based classifier for usual interstitial pneumonia (UIP) derived using CT scan features accurately discriminate radiologist-determined visual UIP?

Study design and methods: A retrospective cohort study was performed. Chest CT scans acquired in individuals with and without ILD were drawn from a variety of public and private data sources. Using radiologist-determined visual UIP as ground truth, a convolutional neural network was used to learn discrete CT scan features of UIP, with outputs used to predict the likelihood of UIP using a linear support vector machine. Test performance characteristics were assessed in an independent performance cohort and multicenter ILD clinical cohort. Transplant-free survival was compared between UIP classification approaches using the Kaplan-Meier estimator and Cox proportional hazards regression.

Results: A total of 2,907 chest CT scans were included in the training (n = 1,934), validation (n = 408), and performance (n = 565) data sets. The prevalence of radiologist-determined visual UIP was 12.4% and 37.1% in the performance and ILD clinical cohorts, respectively. The DL-based UIP classifier predicted visual UIP in the performance cohort with sensitivity and specificity of 93% and 86%, respectively, and in the multicenter ILD clinical cohort with 81% and 77%, respectively. DL-based and visual UIP classification similarly discriminated survival, and outcomes were consistent among cases with positive DL-based UIP classification irrespective of visual classification.

Interpretation: A DL-based classifier for UIP demonstrated good test performance across a wide range of UIP prevalence and similarly discriminated survival when compared with radiologist-determined UIP. This automated tool could efficiently screen for UIP in patients undergoing chest CT scan and identify a high-risk phenotype among those with known ILD.

Keywords: deep learning; idiopathic pulmonary fibrosis; interstitial lung disease; progressive pulmonary fibrosis; radiomic; usual interstitial pneumonia.

PubMed Disclaimer

Conflict of interest statement

Financial/Nonfinancial Disclosures The authors have reported to CHEST the following: L. K., K. L., S. Z., and C. H. are employees of Imbio, Inc. H. G., E. S., S. L., and C. S. are paid consultants for Imbio, Inc. J. H. C., A. A., and J. M. O. report consulting fees from Genentech unrelated to this investigation. None declared (L. C., J. V. P., J. M. W., A. W. M., S. G., A. G.).

Figures

None
Graphical abstract
Figure 1
Figure 1
Radiologist-determined ground truth UIP prevalence in training, validation, and performance cohorts. COPDGene = Chronic Obstructive Pulmonary Disease Genetic Epidemiology study; LTRC = Lung Tissue Research Consortium; MIDRC RiCORD = Medical Imaging Data Resource Center’s RSNA International COVID-19 Open Radiology Database; NLST = National Lung Screening Trial; PCT1 = proprietary idiopathic pulmonary fibrosis clinical trial; UIP = usual interstitial pneumonia.
Figure 2
Figure 2
A, B, Plots of estimated predictive values over a range of UIP prevalence (A) and case identification above and below the selected threshold for binary deep learning UIP classification (B). NPV = negative predictive value; PPV = positive predictive value; UIP = usual interstitial pneumonia.
Figure 3
Figure 3
A-C, Kaplan-Meier survival curves for interstitial lung disease clinical cohort according to deep learning UIP classification (A), visual UIP classification (B), and deep learning UIP classification relative to visual UIP classification (C). UIP = usual interstitial pneumonia.
See this image and copyright information in PMC

Comment in

References

    1. Oldham J.M., Vancheri C. Rethinking idiopathic pulmonary fibrosis. Clin Chest Med. 2021;42(2):263–273. - PMC - PubMed
    1. Raghu G., Remy-Jardin M., Richeldi L., et al. Idiopathic pulmonary fibrosis (an update) and progressive pulmonary fibrosis in adults: an official ATS/ERS/JRS/ALAT Clinical Practice Guideline. Am J Respir Crit Care Med. 2022;205(9):e18–e47. - PMC - PubMed
    1. Hutchinson J.P., Fogarty A.W., McKeever T.M., Hubbard R.B. In-hospital mortality after surgical lung biopsy for interstitial lung disease in the United States. 2000 to 2011. Am J Respir Crit Care Med. 2016;193(10):1161–1167. - PubMed
    1. Hutchinson J.P., McKeever T.M., Fogarty A.W., Navaratnam V., Hubbard R.B. Surgical lung biopsy for the diagnosis of interstitial lung disease in England: 1997-2008. Eur Respir J. 2016;48(5):1453–1461. - PubMed
    1. Chung J.H., Oldham J.M., Montner S.M., et al. CT-pathologic correlation of major types of pulmonary fibrosis: insights for revisions to current guidelines. AJR Am J Roentgenol. 2018;210(5):1034–1041. - PubMed

Publication types