基于吸气期胸部CT的深度学习对小气道疾病的评估与慢性阻塞性肺疾病患者的第一秒用力呼气容积下降相关。
Deep Learning Estimation of Small Airways Disease from Inspiratory Chest CT is Associated with FEV Decline in COPD.
作者信息
Chaudhary Muhammad F A, Awan Hira A, Gerard Sarah E, Bodduluri Sandeep, Comellas Alejandro P, Barjaktarevic Igor Z, Graham Barr R, Cooper Christopher B, Galban Craig J, Han MeiLan K, Curtis Jeffrey L, Hansel Nadia N, Krishnan Jerry A, Menchaca Martha G, Martinez Fernando J, Ohar Jill, Vargas Buonfiglio Luis G, Paine Robert, Bhatt Surya P, Hoffman Eric A, Reinhardt Joseph M
机构信息
The Roy J. Carver Department of Biomedical Engineering, University of Iowa, Iowa City, IA, 52242.
Center for Lung Analytics and Imaging Research (CLAIR), Division of Pulmonary, Allergy and Critical Care Medicine, University of Alabama at Birmingham, Birmingham, AL 35294.
出版信息
medRxiv. 2024 Sep 11:2024.09.10.24313079. doi: 10.1101/2024.09.10.24313079.
RATIONALE
Quantifying functional small airways disease (fSAD) requires additional expiratory computed tomography (CT) scan, limiting clinical applicability. Artificial intelligence (AI) could enable fSAD quantification from chest CT scan at total lung capacity (TLC) alone (fSAD).
OBJECTIVES
To evaluate an AI model for estimating fSAD and study its clinical associations in chronic obstructive pulmonary disease (COPD).
METHODS
We analyzed 2513 participants from the SubPopulations and InteRmediate Outcome Measures in COPD Study (SPIROMICS). Using a subset ( = 1055), we developed a generative model to produce virtual expiratory CTs for estimating fSAD in the remaining 1458 SPIROMICS participants. We compared fSAD with dual volume, parametric response mapping fSAD. We investigated univariate and multivariable associations of fSAD with FEV, FEV/FVC, six-minute walk distance (6MWD), St. George's Respiratory Questionnaire (SGRQ), and FEV decline. The results were validated in a subset ( = 458) from COPDGene study. Multivariable models were adjusted for age, race, sex, BMI, baseline FEV, smoking pack years, smoking status, and percent emphysema.
MEASUREMENTS AND MAIN RESULTS
Inspiratory fSAD was highly correlated with fSAD in SPIROMICS (Pearson's R = 0.895) and COPDGene (R = 0.897) cohorts. In SPIROMICS, fSAD was associated with FEV (L) (adj.β = -0.034, < 0.001), FEV/FVC (adj.β = -0.008, < 0.001), SGRQ (adj.β = 0.243, < 0.001), and FEV decline (mL / year) (adj.β = -1.156, < 0.001). fSAD was also associated with FEV (L) (adj.β = -0.032, < 0.001), FEV/FVC (adj.β = -0.007, < 0.001), SGRQ (adj.β = 0.190, = 0.02), and FEV decline (mL / year) (adj.β = -0.866, = 0.001) in COPDGene. We found fSAD to be more repeatable than fSAD with intraclass correlation of 0.99 (95% CI: 0.98, 0.99) vs. 0.83 (95% CI: 0.76, 0.88).
CONCLUSIONS
Inspiratory fSAD captures small airways disease as reliably as fSAD and is associated with FEV decline.
原理
量化功能性小气道疾病（fSAD）需要额外的呼气计算机断层扫描（CT），这限制了其临床应用。人工智能（AI）能够仅根据全肺容量（TLC）时的胸部CT扫描来量化fSAD（即fSAD）。
目的
评估一种用于估计fSAD的AI模型，并研究其在慢性阻塞性肺疾病（COPD）中的临床关联。
方法
我们分析了慢性阻塞性肺疾病研究中的亚组和中间结局指标（SPIROMICS）的2513名参与者。使用一个子集（n = 1055），我们开发了一个生成模型，以生成虚拟呼气CT，用于估计其余1458名SPIROMICS参与者的fSAD。我们将fSAD与双容积、参数反应映射fSAD进行比较。我们研究了fSAD与第一秒用力呼气容积（FEV）、FEV/用力肺活量（FVC）、6分钟步行距离（6MWD）、圣乔治呼吸问卷（SGRQ）和FEV下降的单变量和多变量关联。结果在慢性阻塞性肺疾病基因（COPDGene）研究的一个子集（n = 458）中得到验证。多变量模型针对年龄、种族、性别、体重指数、基线FEV、吸烟包年数、吸烟状态和肺气肿百分比进行了调整。
测量和主要结果
吸气fSAD与SPIROMICS队列（Pearson相关系数R = 0.895）和COPDGene队列（R = 0.897）中的fSAD高度相关。在SPIROMICS中，fSAD与FEV（升）（调整后β = -0.034，P < 0.001）、FEV/FVC（调整后β = -0.008，P < 0.001）、SGRQ（调整后β = 0.243，P < 0.001）和FEV下降（毫升/年）（调整后β = -1.156，P < 0.001）相关。在COPDGene中，fSAD也与FEV（升）（调整后β = -0.032，P < 0.001）FEV/FVC（调整后β = -0.007，P < 0.001）、SGRQ（调整后β = 0.190，P = 0.02）和FEV下降（毫升/年）（调整后β = -0.866，P = 0.001）相关。我们发现fSAD比fSAD更具可重复性，组内相关系数为0.99（95%置信区间：0.98，0.99），而fSAD为0.83（95%置信区间：0.76，0.88）。
结论
吸气fSAD与fSAD一样可靠地反映小气道疾病，并与FEV下降相关。
Zotero 插件