Automated classification of normal and pathologic pulmonary tissue by topological texture features extracted from multi-detector CT in 3D

To provide a novel, robust algorithm for classification of lung tissue depicted by multi-detector computed tomography (MDCT) based on the topology of CT-attenuation values and to compare discriminative results with densitometric methods. Two hundred seventy-five cubic volumes of interest (VOI, edge length 40 pixels) were obtained from MDCT chest CT (isotropic voxel size, edge length 0.6 mm) of 21 subjects with and without pathology (emphysema, fibrosis). All VOIs were visually consensus-classified by two radiologists. Texture features based on the Minkowski functionals (MF) as well as on the CT attenuation values are determined. Classification results of both approaches were assessed by receiver-operator characteristic and discriminant analysis. By densitometric (topological) parameters, normal and abnormal VOIs were distinguished with an area under the curve ranging from 0.78 to 0.85 (0.87 to 0.96). Correlation between both groups of parameters was non-significant (p ≥ 0.36). By combined information of densitometric and topological quantities, the radiologists’ ratings were reproduced for 92% of VOIs, ranging from 85.7% (fibrosis) to 98% (normal VOIs). Our method performs well for identification of VOIs containing abnormal lung-tissue. Combined information of densitometry and topology increases the number of correctly classified VOIs further. When extended to CT data depicting whole lungs, topological analysis may allow to enhance density-based analysis and improve monitoring texture changes with progression of pulmonary disease.

[1]  Michael R Kosorok,et al.  Reproducibility of a Scoring System for Computed Tomography Scanning in Cystic Fibrosis , 2006, Journal of thoracic imaging.

[2]  G. van Kaick,et al.  Usual interstitial pneumonia. Quantitative assessment of high-resolution computed tomography findings by computer-assisted texture-based image analysis. , 1997, Investigative radiology.

[3]  M. Stone An Asymptotic Equivalence of Choice of Model by Cross‐Validation and Akaike's Criterion , 1977 .

[4]  R. Castile,et al.  Technique and clinical applications of full-inflation and end-exhalation controlled-ventilation chest CT in infants and young children , 2001, Pediatric Radiology.

[5]  Gary K Grunwald,et al.  Quantitative CT indexes in idiopathic pulmonary fibrosis: relationship with physiologic impairment. , 2003, Radiology.

[6]  N J Morrison,et al.  Limitations of computed tomography in the assessment of emphysema. , 1989, The American review of respiratory disease.

[7]  J. Best,et al.  Measurement of CT lung density in patients with chronic asthma. , 1997, The European respiratory journal.

[8]  L Fasano,et al.  [Quantitative assessment of pulmonary emphysema with computerized tomography. Comparison of the visual score and high resolution computerized tomography, expiratory density mask with spiral computerized tomography and respiratory function tests]. , 1997, La Radiologia medica.

[9]  E. V. van Beek,et al.  Computer-aided classification of interstitial lung diseases via MDCT: 3D adaptive multiple feature method (3D AMFM). , 2006, Academic radiology.

[10]  J. Best,et al.  DIAGNOSIS OF PULMONARY EMPHYSEMA BY COMPUTERISED TOMOGRAPHY , 1984, The Lancet.

[11]  A A Bankier,et al.  Pulmonary emphysema: subjective visual grading versus objective quantification with macroscopic morphometry and thin-section CT densitometry. , 1999, Radiology.

[12]  Nikkila Ea Letter: Serum high-density-lipoprotein and coronary heart-disease. , 1976 .

[13]  T. McLoud,et al.  Normal chest roentgenograms in chronic diffuse infiltrative lung disease. , 1978, The New England journal of medicine.

[14]  Kristel Michielsen,et al.  Integral-geometry morphological image analysis , 2001 .

[15]  C Sanders,et al.  The radiographic diagnosis of emphysema. , 1991, Radiologic clinics of North America.

[16]  E. Hoffman,et al.  Interstitial lung disease: A quantitative study using the adaptive multiple feature method. , 1999, American journal of respiratory and critical care medicine.

[17]  G. Raghu,et al.  Usual Interstitial Pneumonia , 2001, Seminars in respiratory and critical care medicine.

[18]  Bernd A. Berg Markov Chain Monte Carlo Simulations and Their Statistical Analysis , 2004 .

[19]  C. Metz Basic principles of ROC analysis. , 1978, Seminars in nuclear medicine.

[20]  Christian P. Robert,et al.  Monte Carlo Statistical Methods , 2005, Springer Texts in Statistics.

[21]  H Nishitani,et al.  The diagnosis of mild emphysema. Correlation of computed tomography and pathology scores. , 1990, The American review of respiratory disease.

[22]  CT Attenuation Values of Lung Density in Sarcoidosis , 1983, Journal of computer assisted tomography.

[23]  J E Cotes,et al.  Lung volumes and forced ventilatory flows. Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society. , 1993, The European respiratory journal. Supplement.

[24]  J. Best,et al.  CT measurements of lung density in life can quantitate distal airspace enlargement--an essential defining feature of human emphysema. , 1988, The American review of respiratory disease.

[25]  E. Hoffman,et al.  Characterization of the interstitial lung diseases via density-based and texture-based analysis of computed tomography images of lung structure and function. , 2003, Academic radiology.

[26]  H. Coxson,et al.  Use of CT morphometry to detect changes in lung weight and gas volume. , 2005, Chest.

[27]  G. Laszlo,et al.  Computed tomography in pulmonary emphysema. , 1982, Clinical radiology.

[28]  Paul J. Friedman,et al.  American Thoracic Society/European Respiratory Society International Multidisciplinary Consensus Classification of the Idiopathic Interstitial Pneumonias. This joint statement of the American Thoracic Society (ATS), and the European Respiratory Society (ERS) was adopted by the ATS board of directors , 2002, American journal of respiratory and critical care medicine.

[29]  E. Hoffman,et al.  Quantification of pulmonary emphysema from lung computed tomography images. , 1997, American journal of respiratory and critical care medicine.

[30]  Frederick R. Long,et al.  Inspiratory and expiratory CT lung density in infants and young children , 2005, Pediatric Radiology.

[31]  Roland Werthschützky,et al.  Automated CT image evaluation of the lung: a morphology-based concept , 2001, IEEE Transactions on Medical Imaging.