Explaining Radiological Emphysema Subtypes with Unsupervised Texture Prototypes: MESA COPD Study.

Authors:
Jie Yang
Jie Yang
Stony Brook University Medical Center
Stony Brook | United States
Elsa D Angelini
Elsa D Angelini
Ecole Nationale Supérieure des Télécommunications
Benjamin M Smith
Benjamin M Smith
College of Physicians and Surgeons
Toronto | Canada
John H M Austin
John H M Austin
Washington University School of Medicine
Saint Louis | United States
Eric A Hoffman
Eric A Hoffman
University of Iowa
Iowa City | United States
David A Bluemke
David A Bluemke
National Institutes of Health
Bethesda | United States
R Graham Barr
R Graham Barr
McMaster University
Canada
Andrew F Laine
Andrew F Laine
Columbia University
United States

Med Comput Vis Bayesian Graph Models Biomed Imaging (2016) 2017 1;2017:69-80. Epub 2017 Jul 1.

Dept. of Biomedical Engineering, Columbia University, New York, NY, USA.

Pulmonary emphysema is traditionally subcategorized into three subtypes, which have distinct radiological appearances on computed tomography (CT) and can help with the diagnosis of chronic obstructive pulmonary disease (COPD). Automated texture-based quantification of emphysema subtypes has been successfully implemented via supervised learning of these three emphysema subtypes. In this work, we demonstrate that unsupervised learning on a large heterogeneous database of CT scans can generate texture prototypes that are visually homogeneous and distinct, reproducible across subjects, and capable of predicting accurately the three standard radiological subtypes. These texture prototypes enable automated labeling of lung volumes, and open the way to new interpretations of lung CT scans with finer subtyping of emphysema.

Download full-text PDF

Source
http://link.springer.com/10.1007/978-3-319-61188-4_7
Publisher Site
http://dx.doi.org/10.1007/978-3-319-61188-4_7DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5708576PMC
July 2017
21 Reads

Publication Analysis

Top Keywords

emphysema subtypes
12
texture prototypes
12
emphysema
5
subtypes
5
automated texture-based
4
predicting accurately
4
copd automated
4
capable predicting
4
quantification emphysema
4
subtypes implemented
4
subjects capable
4
accurately three
4
texture-based quantification
4
pulmonary disease
4
standard radiological
4
diagnosis chronic
4
radiological subtypes
4
chronic obstructive
4
obstructive pulmonary
4
reproducible subjects
4

References

(Supplied by CrossRef)
Article in Am. J. Med.
BM Smith et al.
Am. J. Med. 2014
Article in Ann. Intern. Med.
M Dahl et al.
Ann. Intern. Med. 2002
Article in Clin. Chest Med.
SD Shapiro et al.
Clin. Chest Med. 2000
Article in COPD
RG Barr et al.
COPD 2012
Article in IEEE Trans. Med. Imaging
Y Xu et al.
IEEE Trans. Med. Imaging 2006
Article in IEEE Trans. Med. Imaging
L Srensen et al.
IEEE Trans. Med. Imaging 2010

MJ Gangeh et al.
2010
Article in Med. Image Anal.
A Depeursinge et al.
Med. Image Anal. 2014

V Cheplygina et al.
2015
Article in Am. J. Respir. Crit. Care Med.
JP Sieren et al.
Am. J. Respir. Crit. Care Med. 2016

V Roth et al.
2002

Similar Publications