Lung tissue classification in severe advanced cystic fibrosis from CT scans

A framework for lung tissue classification in Computed Tomography (CT) scans is presented. The method combines supervised and unsupervised learning techniques, with the aim of classifying four tissue types in lung: (i) inflammation, (ii) air-trapped / hypoperfused, (iii) normal / hyperperfused and (iv) bulla / cyst. The framework has been tested on a large heterogeneous dataset, collected over the last 20 years from 17 sites worldwide. The overall accuracy of the proposed methodology is 72.6% and the mean tissue sensitivity and specificity are 70.1% and 89.6% respectively: these results suggest that the framework can be used for the assessment of Severe Advanced Lung Disease (SALD) in patients affected by Cystic Fibrosis.

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

[2]  T J Williams,et al.  The spectrum of structural abnormalities on CT scans from patients with CF with severe advanced lung disease , 2008, Thorax.

[3]  W. Hop,et al.  227* CT scores correlate with survival in cystic fibrosis patients awaiting lung transplantation , 2011 .

[4]  D. Hansell,et al.  Obstructive lung diseases: texture classification for differentiation at CT. , 2003, Radiology.

[5]  James C. Bezdek,et al.  Pattern Recognition with Fuzzy Objective Function Algorithms , 1981, Advanced Applications in Pattern Recognition.

[6]  E. Silverman,et al.  Inflammatory markers of lung disease in adult patients with cystic fibrosis , 2007, Pediatric pulmonology.

[7]  Henning Müller,et al.  Fusing visual and clinical information for lung tissue classification in high-resolution computed tomography , 2010, Artif. Intell. Medicine.

[8]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[9]  Robert E. Schapire,et al.  The Boosting Approach to Machine Learning An Overview , 2003 .

[10]  Lauge Sørensen,et al.  A Texton-Based Approach for the Classification of Lung Parenchyma in CT Images , 2010, MICCAI.

[11]  Ryan M. Rifkin,et al.  In Defense of One-Vs-All Classification , 2004, J. Mach. Learn. Res..

[12]  Petia Radeva,et al.  A Meta-Learning Approach to Conditional Random Fields Using Error-Correcting Output Codes , 2010, 2010 20th International Conference on Pattern Recognition.

[13]  Bram van Ginneken,et al.  Automated classification of hyperlucency, fibrosis, ground glass, solid, and focal lesions in high-resolution CT of the lung. , 2006, Medical physics.

[14]  Thomas G. Dietterich,et al.  Solving Multiclass Learning Problems via Error-Correcting Output Codes , 1994, J. Artif. Intell. Res..

[15]  Lauge Sørensen,et al.  Image Dissimilarity-Based Quantification of Lung Disease from CT , 2010, MICCAI.

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