Chassagnon Guillaume, Zacharaki Evangelia I, Bommart Sébastien, Burgel Pierre-Régis, Chiron Raphael, Dangeard Séverine, Paragios Nikos, Martin Clémence, Revel Marie-Pierre
Department of Radiology (G.C., S.D., M.P.R.) and Respiratory Medicine and National Cystic Reference Center (P.R.B.), Groupe Hospitalier Cochin-Hotel Dieu, AP-HP, Université Paris Descartes, 27 Rue du Faubourg Saint-Jacques, 75014 Paris, France; Center for Visual Computing, Ecole CentraleSupelec, Grande Voie des Vignes, Chatenay Malabry, France (G.C., E.I.Z., N.P.); U1016 Inserm, Institut Cochin, Paris, France (G.C., P.R.B., C.M., M.P.R.); Radiology Department (S.B.) and Pulmonary Department (R.C.), Hôpital Arnaud de Villeneuve, CHU de Montpellier, Université de Montpellier, Montpellier, France; ERN-Lung CF Network, France (P.R.B., C.M.); and TheraPanacea, Paris-Biotech-Santé, Paris, France (N.P.).
Radiol Cardiothorac Imaging. 2020 Dec 17;2(6):e200022. doi: 10.1148/ryct.2020200022. eCollection 2020 Dec.
To develop radiomics-based CT scores for assessing lung disease severity and exacerbation risk in adult patients with cystic fibrosis (CF).
This two-center retrospective observational study was approved by an institutional ethics committee, and the need for patient consent was waived. A total of 215 outpatients with CF referred for unenhanced follow-up chest CT were evaluated in two different centers between January 2013 and December 2016. After lung segmentation, chest CT scans from center 1 (training cohort, 162 patients [median age, 29 years; interquartile range {IQR}, 24-36 years; 84 men]) were used to build CT scores from 38 extracted CT features, using five different machine learning techniques trained to predict a clinical prognostic score, the Nkam score. The correlations between the developed CT scores, two different clinical prognostic scores (Liou and CF-ABLE), forced expiratory volume in 1 second (FEV), and risk of respiratory exacerbations were evaluated in the test cohort (center 2, 53 patients [median age, 27 years; IQR, 22-35 years; 34 men]) using the Spearman rank coefficient.
In the test cohort, all radiomics-based CT scores showed moderate to strong correlation with the Nkam score ( = 0.57 to 0.63, < .001) and Liou scores ( = -0.55 to -0.65, < .001), whereas the correlation with CF-ABLE score was weaker ( = 0.28 to 0.38, = .005 to .048). The developed CT scores showed strong correlation with predicted FEV ( = -0.62 to -0.66, < .001) and weak to moderate correlation with the number of pulmonary exacerbations to occur in the 12 months after the CT examination ( = 0.38 to 0.55, < .001 to = .006).
Radiomics can be used to build automated CT scores that correlate to clinical severity and exacerbation risk in adult patients with CF.Supplemental material is available for this article.See also the commentary by Elicker and Sohn in this issue.© RSNA, 2020.
开发基于影像组学的CT评分系统，用于评估成年囊性纤维化（CF）患者的肺部疾病严重程度和急性加重风险。
本双中心回顾性观察研究经机构伦理委员会批准，无需患者签署知情同意书。2013年1月至2016年12月期间，在两个不同中心对215例因未增强胸部CT随访而转诊的CF门诊患者进行了评估。在进行肺部分割后，中心1（训练队列，162例患者[中位年龄29岁；四分位间距{IQR}，24 - 36岁；84例男性]）的胸部CT扫描用于从38个提取的CT特征构建CT评分，使用五种不同的机器学习技术进行训练，以预测临床预后评分Nkam评分。在测试队列（中心2，53例患者[中位年龄27岁；IQR，22 - 35岁；34例男性]）中，使用Spearman秩系数评估所开发的CT评分与两种不同临床预后评分（Liou和CF - ABLE）、第1秒用力呼气容积（FEV）以及呼吸急性加重风险之间的相关性。
在测试队列中，所有基于影像组学的CT评分与Nkam评分（ = 0.57至0.63， <.001）和Liou评分（ = - 0.55至 - 0.65， <.001）显示出中度至强相关性，而与CF - ABLE评分的相关性较弱（ = 0.28至0.38， = 0.005至0.048）。所开发的CT评分与预测的FEV显示出强相关性（ = - 0.62至 - 0.66， <.001），与CT检查后12个月内发生的肺部急性加重次数显示出弱至中度相关性（ = 0.38至0.55， <.001至 = 0.006）。
影像组学可用于构建与成年CF患者临床严重程度和急性加重风险相关的自动化CT评分。本文提供补充材料。另见本期Elicker和Sohn的评论。©RSNA，2020。
