Classification of CT examinations for COPD visual severity analysis

Jun Tan, Bin Zheng, Xingwei Wang, Jiantao Pu, David Gur, Frank C. Sciurba, J. Ken Leader

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this study we present a computational method of CT examination classification into visual assessed emphysema severity. The visual severity categories ranged from 0 to 5 and were rated by an experienced radiologist. The six categories were none, trace, mild, moderate, severe and very severe. Lung segmentation was performed for every input image and all image features are extracted from the segmented lung only. We adopted a two-level feature representation method for the classification. Five gray level distribution statistics, six gray level co-occurrence matrix (GLCM), and eleven gray level run-length (GLRL) features were computed for each CT image depicted segment lung. Then we used wavelets decomposition to obtain the low- and high-frequency components of the input image, and again extract from the lung region six GLCM features and eleven GLRL features. Therefore our feature vector length is 56. The CT examinations were classified using the support vector machine (SVM) and k-nearest neighbors (KNN) and the traditional threshold (density mask) approach. The SVM classifier had the highest classification performance of all the methods with an overall sensitivity of 54.4% and a 69.6% sensitivity to discriminate "no" and "trace visually assessed emphysema. We believe this work may lead to an automated, objective method to categorically classify emphysema severity on CT exam.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2012
Subtitle of host publicationBiomedical Applications in Molecular, Structural, and Functional Imaging
Volume8317
DOIs
StatePublished - May 14 2012
EventMedical Imaging 2012: Biomedical Applications in Molecular, Structural, and Functional Imaging - San Diego, CA, United States
Duration: Feb 5 2012Feb 7 2012

Other

OtherMedical Imaging 2012: Biomedical Applications in Molecular, Structural, and Functional Imaging
CountryUnited States
CitySan Diego, CA
Period2/5/122/7/12

Fingerprint

Chronic Obstructive Pulmonary Disease
Emphysema
emphysema
examination
Lung
lungs
Support vector machines
Wavelet decomposition
Computational methods
Masks
Classifiers
Statistics
occurrences
sensitivity
matrices
classifiers
masks
statistics
low frequencies
decomposition

Keywords

  • Classification
  • COPD
  • Texture analysis
  • Visual severity quantification

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Tan, J., Zheng, B., Wang, X., Pu, J., Gur, D., Sciurba, F. C., & Leader, J. K. (2012). Classification of CT examinations for COPD visual severity analysis. In Medical Imaging 2012: Biomedical Applications in Molecular, Structural, and Functional Imaging (Vol. 8317). [831723] https://doi.org/10.1117/12.911751

Classification of CT examinations for COPD visual severity analysis. / Tan, Jun; Zheng, Bin; Wang, Xingwei; Pu, Jiantao; Gur, David; Sciurba, Frank C.; Leader, J. Ken.

Medical Imaging 2012: Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 8317 2012. 831723.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Tan, J, Zheng, B, Wang, X, Pu, J, Gur, D, Sciurba, FC & Leader, JK 2012, Classification of CT examinations for COPD visual severity analysis. in Medical Imaging 2012: Biomedical Applications in Molecular, Structural, and Functional Imaging. vol. 8317, 831723, Medical Imaging 2012: Biomedical Applications in Molecular, Structural, and Functional Imaging, San Diego, CA, United States, 2/5/12. https://doi.org/10.1117/12.911751
Tan J, Zheng B, Wang X, Pu J, Gur D, Sciurba FC et al. Classification of CT examinations for COPD visual severity analysis. In Medical Imaging 2012: Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 8317. 2012. 831723 https://doi.org/10.1117/12.911751
Tan, Jun ; Zheng, Bin ; Wang, Xingwei ; Pu, Jiantao ; Gur, David ; Sciurba, Frank C. ; Leader, J. Ken. / Classification of CT examinations for COPD visual severity analysis. Medical Imaging 2012: Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 8317 2012.
@inproceedings{d54643be112942499e3898c22c79180d,
title = "Classification of CT examinations for COPD visual severity analysis",
abstract = "In this study we present a computational method of CT examination classification into visual assessed emphysema severity. The visual severity categories ranged from 0 to 5 and were rated by an experienced radiologist. The six categories were none, trace, mild, moderate, severe and very severe. Lung segmentation was performed for every input image and all image features are extracted from the segmented lung only. We adopted a two-level feature representation method for the classification. Five gray level distribution statistics, six gray level co-occurrence matrix (GLCM), and eleven gray level run-length (GLRL) features were computed for each CT image depicted segment lung. Then we used wavelets decomposition to obtain the low- and high-frequency components of the input image, and again extract from the lung region six GLCM features and eleven GLRL features. Therefore our feature vector length is 56. The CT examinations were classified using the support vector machine (SVM) and k-nearest neighbors (KNN) and the traditional threshold (density mask) approach. The SVM classifier had the highest classification performance of all the methods with an overall sensitivity of 54.4{\%} and a 69.6{\%} sensitivity to discriminate {"}no{"} and {"}trace visually assessed emphysema. We believe this work may lead to an automated, objective method to categorically classify emphysema severity on CT exam.",
keywords = "Classification, COPD, Texture analysis, Visual severity quantification",
author = "Jun Tan and Bin Zheng and Xingwei Wang and Jiantao Pu and David Gur and Sciurba, {Frank C.} and Leader, {J. Ken}",
year = "2012",
month = "5",
day = "14",
doi = "10.1117/12.911751",
language = "English (US)",
isbn = "9780819489661",
volume = "8317",
booktitle = "Medical Imaging 2012",

}

TY - GEN

T1 - Classification of CT examinations for COPD visual severity analysis

AU - Tan, Jun

AU - Zheng, Bin

AU - Wang, Xingwei

AU - Pu, Jiantao

AU - Gur, David

AU - Sciurba, Frank C.

AU - Leader, J. Ken

PY - 2012/5/14

Y1 - 2012/5/14

N2 - In this study we present a computational method of CT examination classification into visual assessed emphysema severity. The visual severity categories ranged from 0 to 5 and were rated by an experienced radiologist. The six categories were none, trace, mild, moderate, severe and very severe. Lung segmentation was performed for every input image and all image features are extracted from the segmented lung only. We adopted a two-level feature representation method for the classification. Five gray level distribution statistics, six gray level co-occurrence matrix (GLCM), and eleven gray level run-length (GLRL) features were computed for each CT image depicted segment lung. Then we used wavelets decomposition to obtain the low- and high-frequency components of the input image, and again extract from the lung region six GLCM features and eleven GLRL features. Therefore our feature vector length is 56. The CT examinations were classified using the support vector machine (SVM) and k-nearest neighbors (KNN) and the traditional threshold (density mask) approach. The SVM classifier had the highest classification performance of all the methods with an overall sensitivity of 54.4% and a 69.6% sensitivity to discriminate "no" and "trace visually assessed emphysema. We believe this work may lead to an automated, objective method to categorically classify emphysema severity on CT exam.

AB - In this study we present a computational method of CT examination classification into visual assessed emphysema severity. The visual severity categories ranged from 0 to 5 and were rated by an experienced radiologist. The six categories were none, trace, mild, moderate, severe and very severe. Lung segmentation was performed for every input image and all image features are extracted from the segmented lung only. We adopted a two-level feature representation method for the classification. Five gray level distribution statistics, six gray level co-occurrence matrix (GLCM), and eleven gray level run-length (GLRL) features were computed for each CT image depicted segment lung. Then we used wavelets decomposition to obtain the low- and high-frequency components of the input image, and again extract from the lung region six GLCM features and eleven GLRL features. Therefore our feature vector length is 56. The CT examinations were classified using the support vector machine (SVM) and k-nearest neighbors (KNN) and the traditional threshold (density mask) approach. The SVM classifier had the highest classification performance of all the methods with an overall sensitivity of 54.4% and a 69.6% sensitivity to discriminate "no" and "trace visually assessed emphysema. We believe this work may lead to an automated, objective method to categorically classify emphysema severity on CT exam.

KW - Classification

KW - COPD

KW - Texture analysis

KW - Visual severity quantification

UR - http://www.scopus.com/inward/record.url?scp=84860722649&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860722649&partnerID=8YFLogxK

U2 - 10.1117/12.911751

DO - 10.1117/12.911751

M3 - Conference contribution

SN - 9780819489661

VL - 8317

BT - Medical Imaging 2012

ER -