Function Follows Form: Regression from Complete Thoracic Computed Tomography Scans

Chronic Obstructive Pulmonary Disease (COPD) is a leading cause of morbidity and mortality. While COPD diagnosis is based on lung function tests, early stages and progression of different aspects of the disease can be visible and quantitatively assessed on computed tomography (CT) scans. Many studies have been published that quantify imaging biomarkers related to COPD. In this paper we present a convolutional neural network that directly computes visual emphysema scores and predicts the outcome of lung function tests for 195 CT scans from the COPDGene study. Contrary to previous work, the proposed method does not encode any specific prior knowledge about what to quantify, but it is trained end-to-end with a set of 1424 CT scans for which the output parameters were available. The network provided state-of-the-art results for these tasks: Visual emphysema scores are comparable to those assessed by trained human observers; COPD diagnosis from estimated lung function reaches an area under the ROC curve of 0.94, outperforming prior art. The method is easily generalizable to other situations where information from whole scans needs to be summarized in single quantities.

[1]  Bram van Ginneken,et al.  Automatic Segmentation of Pulmonary Segments From Volumetric Chest CT Scans , 2009, IEEE Transactions on Medical Imaging.

[2]  B. van Ginneken,et al.  The relationship between lung function impairment and quantitative computed tomography in chronic obstructive pulmonary disease , 2011, European Radiology.

[3]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[4]  Nicola A Hanania,et al.  Chronic obstructive pulmonary disease exacerbations in the COPDGene study: associated radiologic phenotypes. , 2011, Radiology.

[5]  Polina Golland,et al.  Unsupervised Discovery of Emphysema Subtypes in a Large Clinical Cohort , 2016, MLMI@MICCAI.

[6]  Scott G. Baginski,et al.  Association of Coronary Artery Calcification and Mortality in the National Lung Screening Trial: A Comparison of Three Scoring Methods. , 2015, Radiology.

[7]  Fei-Fei Li,et al.  Large-Scale Video Classification with Convolutional Neural Networks , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[8]  L. Tanoue,et al.  Identification of Chronic Obstructive Pulmonary Disease in Lung Cancer Screening Computed Tomographic Scans , 2012 .

[9]  Lyle J. Palmer,et al.  Precision Radiology: Predicting longevity using feature engineering and deep learning methods in a radiomics framework , 2017, Scientific Reports.

[10]  J. Seo,et al.  Texture-Based Quantification of Pulmonary Emphysema on High-Resolution Computed Tomography: Comparison With Density-Based Quantification and Correlation With Pulmonary Function Test , 2008, Investigative radiology.

[11]  E. Regan,et al.  Genetic Epidemiology of COPD (COPDGene) Study Design , 2011, COPD.

[12]  Bram van Ginneken,et al.  Toward automated segmentation of the pathological lung in CT , 2005, IEEE Transactions on Medical Imaging.

[13]  N. Richard,et al.  Efficacy and safety of umeclidinium/vilanterol 62.5/25 mcg and tiotropium 18 mcg in chronic obstructive pulmonary disease: results of a 24-week, randomized, controlled trial. , 2014, Respiratory medicine.

[14]  Harry J de Koning,et al.  Diagnosis of chronic obstructive pulmonary disease in lung cancer screening Computed Tomography scans: independent contribution of emphysema, air trapping and bronchial wall thickening , 2013, Respiratory Research.

[15]  Juan F Mosley,et al.  Tiotropium Bromide/Olodaterol (Stiolto Respimat): Once-Daily Combination Therapy for the Maintenance of COPD. , 2016, P & T : a peer-reviewed journal for formulary management.

[16]  Ye Xu,et al.  MDCT-based 3-D texture classification of emphysema and early smoking related lung pathologies , 2006, IEEE Transactions on Medical Imaging.

[17]  Lauge Sørensen,et al.  Quantitative Analysis of Pulmonary Emphysema Using Local Binary Patterns , 2010, IEEE Transactions on Medical Imaging.

[18]  Edwin K Silverman,et al.  CT-Definable Subtypes of Chronic Obstructive Pulmonary Disease: A Statement of the Fleischner Society. , 2015, Radiology.

[19]  Raúl San José Estépar,et al.  Visual Assessment of Chest Computed Tomographic Images Is Independently Useful for Genetic Association Analysis in Studies of Chronic Obstructive Pulmonary Disease , 2017, Annals of the American Thoracic Society.

[20]  Johan H C Reiber,et al.  Volume correction in computed tomography densitometry for follow-up studies on pulmonary emphysema. , 2008, Proceedings of the American Thoracic Society.

[21]  Joyce D. Schroeder,et al.  A Combined Pulmonary-Radiology Workshop for Visual Evaluation of COPD: Study Design, Chest CT Findings and Concordance with Quantitative Evaluation , 2012, COPD.

[22]  David A Lynch,et al.  Texture-based Quantification of Centrilobular Emphysema and Centrilobular Nodularity in Longitudinal CT Scans of Current and Former Smokers. , 2016, Academic radiology.

[23]  Eric A. Hoffman,et al.  Adaptive Quantification and Longitudinal Analysis of Pulmonary Emphysema With a Hidden Markov Measure Field Model , 2014, IEEE Transactions on Medical Imaging.

[24]  Philip F. Judy,et al.  Normalizing computed tomography data reconstructed with different filter kernels: effect on emphysema quantification , 2015, European Radiology.

[25]  Lauge Sørensen,et al.  Classification of COPD with Multiple Instance Learning , 2014, 2014 22nd International Conference on Pattern Recognition.

[26]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[27]  Elsa D. Angelini,et al.  Texton and sparse representation based texture classification of lung parenchyma in CT images , 2016, 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).