Deep Learning Estimation of Small Airway Disease from Inspiratory Chest Computed Tomography: Clinical Validation, Repeatability, and Associations with Adverse Clinical Outcomes in Chronic Obstructive Pulmonary Disease.
Muhammad F A Chaudhary, Hira A Awan, Sarah E Gerard, Sandeep Bodduluri, Alejandro P Comellas, Igor Z Barjaktarevic, R Graham Barr, Christopher B Cooper, Craig J Galban, MeiLan K Han, Jeffrey L Curtis, Nadia N Hansel, Jerry A Krishnan, Martha G Menchaca, Fernando J Martinez, Jill Ohar, Luis G Vargas Buonfiglio, Robert Paine, Surya P Bhatt, Eric A Hoffman, Joseph M Reinhardt
{"title":"Deep Learning Estimation of Small Airway Disease from Inspiratory Chest Computed Tomography: Clinical Validation, Repeatability, and Associations with Adverse Clinical Outcomes in Chronic Obstructive Pulmonary Disease.","authors":"Muhammad F A Chaudhary, Hira A Awan, Sarah E Gerard, Sandeep Bodduluri, Alejandro P Comellas, Igor Z Barjaktarevic, R Graham Barr, Christopher B Cooper, Craig J Galban, MeiLan K Han, Jeffrey L Curtis, Nadia N Hansel, Jerry A Krishnan, Martha G Menchaca, Fernando J Martinez, Jill Ohar, Luis G Vargas Buonfiglio, Robert Paine, Surya P Bhatt, Eric A Hoffman, Joseph M Reinhardt","doi":"10.1164/rccm.202409-1847OC","DOIUrl":null,"url":null,"abstract":"<p><p><b>Rationale:</b> Quantifying functional small airway disease (fSAD) requires additional expiratory computed tomography (CT) scans, limiting clinical applicability. Artificial intelligence (AI) could enable fSAD quantification from chest CT scans at total lung capacity (TLC) alone (fSAD<sup>TLC</sup>). <b>Objectives:</b> To evaluate an AI model for estimating fSAD<sup>TLC</sup>, compare it with dual-volume parametric response mapping fSAD (fSAD<sup>PRM</sup>), and assess its clinical associations and repeatability in chronic obstructive pulmonary disease (COPD). <b>Methods:</b> We analyzed 2,513 participants from SPIROMICS (the Subpopulations and Intermediate Outcome Measures in COPD Study). Using a randomly sampled subset (<i>n</i> = 1,055), we developed a generative model to produce virtual expiratory CT scans for estimating fSAD<sup>TLC</sup> in the remaining 1,458 SPIROMICS participants. We compared fSAD<sup>TLC</sup> with dual-volume fSAD<sup>PRM</sup>. We investigated univariate and multivariable associations of fSAD<sup>TLC</sup> with FEV<sub>1</sub>, FEV<sub>1</sub>/FVC ratio, 6-minute-walk distance, St. George's Respiratory Questionnaire score, and FEV<sub>1</sub> decline. The results were validated in a subset of patients from the COPDGene (Genetic Epidemiology of COPD) study (<i>n</i> = 458). Multivariable models were adjusted for age, race, sex, body mass index, baseline FEV<sub>1</sub>, smoking pack-years, smoking status, and percent emphysema. <b>Measurements and Main Results:</b> Inspiratory fSAD<sup>TLC</sup> showed a strong correlation with fSAD<sup>PRM</sup> in SPIROMICS (Pearson's <i>R</i> = 0.895) and COPDGene (<i>R</i> = 0.897) cohorts. Higher fSAD<sup>TLC</sup> levels were significantly associated with lower lung function, including lower postbronchodilator FEV<sub>1</sub> (in liters) and FEV<sub>1</sub>/FVC ratio, and poorer quality of life reflected by higher total St. George's Respiratory Questionnaire scores independent of percent CT emphysema. In SPIROMICS, individuals with higher fSAD<sup>TLC</sup> experienced an annual decline in FEV<sub>1</sub> of 1.156 ml (relative decrease; 95% confidence interval [CI], 0.613-1.699; <i>P</i> < 0.001) per year for every 1% increase in fSAD<sup>TLC</sup>. The rate of decline in the COPDGene cohort was slightly lower at 0.866 ml/yr (relative decrease; 95% CI, 0.345-1.386; <i>P</i> < 0.001) per 1% increase in fSAD<sup>TLC</sup>. Inspiratory fSAD<sup>TLC</sup> demonstrated greater consistency between repeated measurements, with a higher intraclass correlation coefficient of 0.99 (95% CI, 0.98-0.99) compared with fSAD<sup>PRM</sup> (0.83; 95% CI, 0.76-0.88). <b>Conclusions:</b> Small airway disease can be reliably assessed from a single inspiratory CT scan using generative AI, eliminating the need for an additional expiratory CT scan. fSAD estimation from inspiratory CT correlates strongly with fSAD<sup>PRM</sup>, demonstrates a significant association with FEV<sub>1</sub> decline, and offers greater repeatability.</p>","PeriodicalId":7664,"journal":{"name":"American journal of respiratory and critical care medicine","volume":" ","pages":"1185-1195"},"PeriodicalIF":19.4000,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12264693/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"American journal of respiratory and critical care medicine","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1164/rccm.202409-1847OC","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CRITICAL CARE MEDICINE","Score":null,"Total":0}
引用次数: 0
Abstract
Rationale: Quantifying functional small airway disease (fSAD) requires additional expiratory computed tomography (CT) scans, limiting clinical applicability. Artificial intelligence (AI) could enable fSAD quantification from chest CT scans at total lung capacity (TLC) alone (fSADTLC). Objectives: To evaluate an AI model for estimating fSADTLC, compare it with dual-volume parametric response mapping fSAD (fSADPRM), and assess its clinical associations and repeatability in chronic obstructive pulmonary disease (COPD). Methods: We analyzed 2,513 participants from SPIROMICS (the Subpopulations and Intermediate Outcome Measures in COPD Study). Using a randomly sampled subset (n = 1,055), we developed a generative model to produce virtual expiratory CT scans for estimating fSADTLC in the remaining 1,458 SPIROMICS participants. We compared fSADTLC with dual-volume fSADPRM. We investigated univariate and multivariable associations of fSADTLC with FEV1, FEV1/FVC ratio, 6-minute-walk distance, St. George's Respiratory Questionnaire score, and FEV1 decline. The results were validated in a subset of patients from the COPDGene (Genetic Epidemiology of COPD) study (n = 458). Multivariable models were adjusted for age, race, sex, body mass index, baseline FEV1, smoking pack-years, smoking status, and percent emphysema. Measurements and Main Results: Inspiratory fSADTLC showed a strong correlation with fSADPRM in SPIROMICS (Pearson's R = 0.895) and COPDGene (R = 0.897) cohorts. Higher fSADTLC levels were significantly associated with lower lung function, including lower postbronchodilator FEV1 (in liters) and FEV1/FVC ratio, and poorer quality of life reflected by higher total St. George's Respiratory Questionnaire scores independent of percent CT emphysema. In SPIROMICS, individuals with higher fSADTLC experienced an annual decline in FEV1 of 1.156 ml (relative decrease; 95% confidence interval [CI], 0.613-1.699; P < 0.001) per year for every 1% increase in fSADTLC. The rate of decline in the COPDGene cohort was slightly lower at 0.866 ml/yr (relative decrease; 95% CI, 0.345-1.386; P < 0.001) per 1% increase in fSADTLC. Inspiratory fSADTLC demonstrated greater consistency between repeated measurements, with a higher intraclass correlation coefficient of 0.99 (95% CI, 0.98-0.99) compared with fSADPRM (0.83; 95% CI, 0.76-0.88). Conclusions: Small airway disease can be reliably assessed from a single inspiratory CT scan using generative AI, eliminating the need for an additional expiratory CT scan. fSAD estimation from inspiratory CT correlates strongly with fSADPRM, demonstrates a significant association with FEV1 decline, and offers greater repeatability.
期刊介绍:
The American Journal of Respiratory and Critical Care Medicine focuses on human biology and disease, as well as animal studies that contribute to the understanding of pathophysiology and treatment of diseases that affect the respiratory system and critically ill patients. Papers that are solely or predominantly based in cell and molecular biology are published in the companion journal, the American Journal of Respiratory Cell and Molecular Biology. The Journal also seeks to publish clinical trials and outstanding review articles on areas of interest in several forms. The State-of-the-Art review is a treatise usually covering a broad field that brings bench research to the bedside. Shorter reviews are published as Critical Care Perspectives or Pulmonary Perspectives. These are generally focused on a more limited area and advance a concerted opinion about care for a specific process. Concise Clinical Reviews provide an evidence-based synthesis of the literature pertaining to topics of fundamental importance to the practice of pulmonary, critical care, and sleep medicine. Images providing advances or unusual contributions to the field are published as Images in Pulmonary, Critical Care, Sleep Medicine and the Sciences.
A recent trend and future direction of the Journal has been to include debates of a topical nature on issues of importance in pulmonary and critical care medicine and to the membership of the American Thoracic Society. Other recent changes have included encompassing works from the field of critical care medicine and the extension of the editorial governing of journal policy to colleagues outside of the United States of America. The focus and direction of the Journal is to establish an international forum for state-of-the-art respiratory and critical care medicine.