Predicting Pulmonary Function Testing from Quantified Computed Tomography Using Machine Learning Algorithms in Patients with COPD

被引:27
作者
Gawlitza, Joshua [1 ]
Sturm, Timo [2 ]
Spohrer, Kai [2 ]
Henzler, Thomas [1 ]
Akin, Ibrahim [3 ,4 ]
Schonberg, Stefan [1 ]
Borggrefe, Martin [3 ,4 ]
Haubenreisser, Holger [1 ]
Trinkmann, Frederik [3 ,5 ]
机构
[1] Heidelberg Univ, Inst Clin Radiol & Nucl Med, Univ Med Ctr Mannheim, Med Fac Mannheim, Theodor Kutzer Ufer 1-3, D-68167 Mannheim, Germany
[2] Univ Mannheim, Dept Gen Management & Informat Syst, D-68131 Mannheim, Germany
[3] Heidelberg Univ, Med Fac Mannheim, Univ Med Ctr Mannheim, Dept Med Cardiol Angiol Pulm & Intens Care 1, Theodor Kutzer Ufer 1-3, D-68167 Mannheim, Germany
[4] DZHK German Ctr Cardiovasc Res, Partner Site, D-68167 Mannheim, Germany
[5] Heidelberg Univ, Med Fac Mannheim, Univ Med Ctr Mannheim, Dept Biomed Informat,Heinrich Lanz Ctr, Theodor Kutzer Ufer 1-3, D-68167 Mannheim, Germany
来源
DIAGNOSTICS | 2019年 / 9卷 / 01期
关键词
chronic obstructive pulmonary disease; machine learning; thorax; AIR-FLOW OBSTRUCTION; LUNG-FUNCTION; CT; EMPHYSEMA; DENSITY; DISEASE; MORPHOMETRY; MANAGEMENT; MORTALITY; STATEMENT;
D O I
10.3390/diagnostics9010033
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
Introduction: Quantitative computed tomography (qCT) is an emergent technique for diagnostics and research in patients with chronic obstructive pulmonary disease (COPD). qCT parameters demonstrate a correlation with pulmonary function tests and symptoms. However, qCT only provides anatomical, not functional, information. We evaluated five distinct, partial-machine learning-based mathematical models to predict lung function parameters from qCT values in comparison with pulmonary function tests. Methods: 75 patients with diagnosed COPD underwent body plethysmography and a dose-optimized qCT examination on a third-generation, dual-source CT with inspiration and expiration. Delta values (inspiration-expiration) were calculated afterwards. Four parameters were quantified: mean lung density, lung volume low-attenuated volume, and full width at half maximum. Five models were evaluated for best prediction: average prediction, median prediction, k-nearest neighbours (kNN), gradient boosting, and multilayer perceptron. Results: The lowest mean relative error (MRE) was calculated for the kNN model with 16%. Similar low MREs were found for polynomial regression as well as gradient boosting-based prediction. Other models led to higher MREs and thereby worse predictive performance. Beyond the sole MRE, distinct differences in prediction performance, dependent on the initial dataset (expiration, inspiration, delta), were found. Conclusion: Different, partially machine learning-based models allow the prediction of lung function values from static qCT parameters within a reasonable margin of error. Therefore, qCT parameters may contain more information than we currently utilize and can potentially augment standard functional lung testing.
引用
收藏
页数:13
相关论文
共 35 条
[1]   AN INTRODUCTION TO KERNEL AND NEAREST-NEIGHBOR NONPARAMETRIC REGRESSION [J].
ALTMAN, NS .
AMERICAN STATISTICIAN, 1992, 46 (03) :175-185
[2]   Relationship between extent of pulmonary emphysema by high-resolution computed tomography and lung elastic recoil in patients with chronic obstructive pulmonary disease [J].
Baldi, S ;
Miniati, M ;
Bellina, CR ;
Battolla, L ;
Catapano, G ;
Begliomini, E ;
Giustini, D ;
Giuntini, C .
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE, 2001, 164 (04) :585-589
[3]   Spirometrically gated high-resolution CT findings in COPD - Lung attenuation vs lung function and dyspnea severity [J].
Camiciottoli, G ;
Bartolucci, M ;
Maluccio, NM ;
Moroni, C ;
Mascalchi, M ;
Giuntini, C ;
Pistolesi, M .
CHEST, 2006, 129 (03) :558-564
[4]   An official American Thoracic Society/European Respiratory Society statement: research questions in COPD [J].
Celli, Bartolome R. ;
Decramer, Marc ;
Wedzicha, Jadwiga A. ;
Wilson, Kevin C. ;
Agusti, Alvar A. ;
Criner, Gerard J. ;
MacNee, William ;
Make, Barry J. ;
Rennard, Stephen I. ;
Stockley, Robert A. ;
Vogelmeier, Claus ;
Anzueto, Antonio ;
Au, David H. ;
Barnes, Peter J. ;
Burgel, Pierre-Regis ;
Calverley, Peter M. ;
Casanova, Ciro ;
Clini, Enrico M. ;
Cooper, Christopher B. ;
Coxson, Harvey O. ;
Dusser, Daniel J. ;
Fabbri, Leonardo M. ;
Fahy, Bonnie ;
Ferguson, Gary T. ;
Fisher, Andrew ;
Fletcher, Monica J. ;
Hayot, Maurice ;
Hurst, John R. ;
Jones, Paul W. ;
Mahler, Donald A. ;
Maltais, Francois ;
Mannino, David M. ;
Martinez, Fernando J. ;
Miravitlles, Marc ;
Meek, Paula M. ;
Papi, Alberto ;
Rabe, Klaus F. ;
Roche, Nicolas ;
Sciurba, Frank C. ;
Sethi, Sanjay ;
Siafakas, Nikos ;
Sin, Don D. ;
Soriano, Joan B. ;
Stoller, James K. ;
Tashkin, Donald P. ;
Troosters, Thierry ;
Verleden, Geert M. ;
Verschakelen, Johny ;
Vestbo, Jorgen ;
Walsh, John W. .
EUROPEAN RESPIRATORY REVIEW, 2015, 24 (136) :159-172
[5]  
Chan H.-P., 2004, SAMPLE SIZE VALIDATI, P872
[6]   EFFECTS OF SAMPLE-SIZE IN CLASSIFIER DESIGN [J].
FUKUNAGA, K ;
HAYES, RR .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1989, 11 (08) :873-885
[7]   The method and efficacy of support vector machine classifiers based on texture features and multi-resolution histogram from 18F-FDG PET-CT images for the evaluation of mediastinal lymph nodes inpatients with lung cancer [J].
Gao, Xuan ;
Chu, Chunyu ;
Li, Yingci ;
Lu, Peiou ;
Wang, Wenzhi ;
Liu, Wanyu ;
Yu, Lijuan .
EUROPEAN JOURNAL OF RADIOLOGY, 2015, 84 (02) :312-317
[8]   Finding the right spot: Where to measure airway parameters in patients with COPD [J].
Gawlitza, Joshua ;
Haubenreisser, Holger ;
Henzler, Thomas ;
Akin, Ibrahim ;
Schoenberg, Stefan ;
Borggrefe, Martin ;
Trinkmann, Frederik .
EUROPEAN JOURNAL OF RADIOLOGY, 2018, 104 :87-93
[9]   Time to Exhale: Additional Value of Expiratory Chest CT in Chronic Obstructive Pulmonary Disease [J].
Gawlitza, Joshua ;
Trinkmann, Frederik ;
Scheffel, Hans ;
Fischer, Andreas ;
Nance, John W. ;
Henzler, Claudia ;
Vogler, Nils ;
Saur, Joachim ;
Akin, Ibrahim ;
Borggrefe, Martin ;
Schoenberg, Stefan O. ;
Henzler, Thomas .
CANADIAN RESPIRATORY JOURNAL, 2018, 2018
[10]   Comparison of computed density and microscopic morphometry in pulmonary emphysema [J].
Gevenois, PA ;
DeVuyst, P ;
deMaertelaer, V ;
Zanen, J ;
Jacobvitz, D ;
Cosio, MG ;
Yernault, JC .
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE, 1996, 154 (01) :187-192