Lee Amanda N, Hsiao Albert, Hasenstab Kyle A
From the Computational Science Research Center, San Diego State University, San Diego, Calif (A.N.L.); Department of Radiology, University of California San Diego, La Jolla, Calif (A.H.); and Department of Mathematics and Statistics, San Diego State University, 5500 Campanile Dr, San Diego, CA 92182 (K.A.H.).
Radiol Cardiothorac Imaging. 2024 Dec;6(6):e240005. doi: 10.1148/ryct.240005.
Purpose To measure the benefit of single-phase CT, inspiratory-expiratory CT, and clinical data for convolutional neural network (CNN)-based chronic obstructive pulmonary disease (COPD) staging. Materials and Methods This retrospective study included inspiratory and expiratory lung CT images and spirometry measurements acquired between November 2007 and April 2011 from 8893 participants (mean age, 59.6 years ± 9.0 [SD]; 53.3% [4738 of 8893] male) in the COPDGene phase I cohort (ClinicalTrials.gov: NCT00608764). CNNs were trained to predict spirometry measurements (forced expiratory volume in 1 second [FEV], FEV percent predicted, and ratio of FEV to forced vital capacity [FEV/FVC]) using clinical data and either single-phase or multiphase CT. Spirometry predictions were then used to predict Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage. Agreement between CNN-predicted and reference standard spirometry measurements and GOLD stage was assessed using intraclass correlation coefficient (ICC) and compared using bootstrapping. Accuracy for predicting GOLD stage, within-one GOLD stage, and GOLD 0 versus 1-4 was calculated. Results CNN-predicted and reference standard spirometry measurements showed moderate to good agreement (ICC, 0.66-0.79), which improved by inclusion of clinical data (ICC, 0.70-0.85; ≤ .04), except for FEV/FVC in the inspiratory-phase CNN model with clinical data ( = .35) and FEV in the expiratory-phase CNN model with clinical data ( = .33). Single-phase CNN accuracies for GOLD stage, within-one stage, and diagnosis ranged from 59.8% to 84.1% (682-959 of 1140), with moderate to good agreement (ICC, 0.68-0.70). Accuracies of CNN models using inspiratory and expiratory images ranged from 60.0% to 86.3% (684-984 of 1140), with moderate to good agreement (ICC, 0.72). Inclusion of clinical data improved agreement and accuracy for both the single-phase CNNs (ICC, 0.72; ≤ .001; accuracy, 65.2%-85.8% [743-978 of 1140]) and inspiratory-expiratory CNNs (ICC, 0.77-0.78; ≤ .001; accuracy, 67.6%-88.0% [771-1003 of 1140]), except expiratory CNN with clinical data (no change in GOLD stage ICC; = .08). Conclusion CNN-based COPD diagnosis and staging using single-phase CT provides comparable accuracy with inspiratory-expiratory CT when provided clinical data relevant to staging. Convolutional Neural Network, Chronic Obstructive Pulmonary Disease, CT, Severity Staging, Attention Map © RSNA, 2024.
目的 评估单相CT、吸气-呼气CT及临床数据对基于卷积神经网络（CNN）的慢性阻塞性肺疾病（COPD）分期的作用。材料与方法 这项回顾性研究纳入了2007年11月至2011年4月期间COPDGene I期队列研究（ClinicalTrials.gov：NCT00608764）中8893名参与者（平均年龄59.6岁±9.0[标准差]；53.3%[8893名中的4738名]为男性）的吸气和呼气肺部CT图像以及肺功能测量数据。使用临床数据以及单相或多相CT对CNN进行训练，以预测肺功能测量值（第1秒用力呼气量[FEV]、预测FEV百分比以及FEV与用力肺活量的比值[FEV/FVC]）。然后用肺功能预测值来预测慢性阻塞性肺疾病全球倡议（GOLD）分期。使用组内相关系数（ICC）评估CNN预测的肺功能测量值和参考标准肺功能测量值及GOLD分期之间的一致性，并通过自抽样法进行比较。计算预测GOLD分期、在一个GOLD分期范围内以及GOLD 0期与1 - 4期的准确率。结果 CNN预测的肺功能测量值与参考标准肺功能测量值显示出中度到良好的一致性（ICC，0.66 - 0.79），纳入临床数据后一致性有所提高（ICC, 0.70 - 0.85；P≤0.04），但在有临床数据的吸气期CNN模型中的FEV/FVC（P = 0.35）以及有临床数据的呼气期CNN模型中的FEV（P = 0.33）除外。单相CNN对GOLD分期、在一个分期范围内以及诊断的准确率为59.8%至84.1%（1140名中的682 - 959名），一致性为中度到良好（ICC，0.68 - 0.70）。使用吸气和呼气图像的CNN模型准确率为60.0%至86.3%（1140名中的684 - 984名），一致性为中度到良好（ICC，0.72）。纳入临床数据提高了单相CNN（ICC，0.72；P≤0.001；准确率，65.2% - 85.8%[1140名中的743 - 978名]）和吸气 - 呼气CNN（ICC，0.77 - 0.78；P≤0.001；准确率，67.6% - 88.0%[1140名中的771 - 1003名]）的一致性和准确率，但有临床数据的呼气期CNN除外（GOLD分期ICC无变化；P = 0.08）。结论 当提供与分期相关的临床数据时，基于CNN的使用单相CT的COPD诊断和分期与吸气 - 呼气CT具有相当的准确性。卷积神经网络、慢性阻塞性肺疾病、CT、严重程度分期、注意力图 © RSNA，2024。
