GENERATIVE METHOD TO DISCOVER EMPHYSEMA SUBTYPES WITH UNSUPERVISED LEARNING USING LUNG MACROSCOPIC PATTERNS (LMPS): THE MESA COPD STUDY

Jingkuan Song; Jie Yang; Benjamin Smith; Pallavi Balte; Eric A Hoffman; R Graham Barr; Andrew F Laine; Elsa D Angelini

doi:10.1109/ISBI.2017.7950541

Journal article

GENERATIVE METHOD TO DISCOVER EMPHYSEMA SUBTYPES WITH UNSUPERVISED LEARNING USING LUNG MACROSCOPIC PATTERNS (LMPS): THE MESA COPD STUDY

Jingkuan Song, Jie Yang, Benjamin Smith, Pallavi Balte, Eric A Hoffman, R Graham Barr, Andrew F Laine and Elsa D Angelini

Proceedings (International Symposium on Biomedical Imaging), Vol.2017, pp.375-378

04/2017

DOI: 10.1109/ISBI.2017.7950541

PMCID: PMC5629072

PMID: 28989563

Files and links (1)

url

http://doi.org/10.1109/ISBI.2017.7950541View

Open Access

Abstract

Pulmonary emphysema overlaps considerably with chronic obstructive pulmonary disease (COPD), and is traditionally subcategorized into three subtypes: centrilobular emphysema (CLE), panlobular emphysema (PLE) and paraseptal emphysema (PSE). Automated classification methods based on supervised learning are generally based upon the current definition of emphysema subtypes, while unsupervised learning of texture patterns enables the objective discovery of possible new radiological emphysema subtypes. In this work, we use a variant of the Latent Dirichlet Allocation (LDA) model to discover lung macroscopic patterns (LMPs) in an unsupervised way from lung regions that encode emphysematous areas. We evaluate the possible utility of the LMPs as potential novel emphysema subtypes via measuring their level of reproducibility when varying the learning set and by their ability to predict traditional radiological emphysema subtypes. Experimental results show that our algorithm can discover highly reproducible LMPs, that predict traditional emphysema subtypes.

Lung

texture

LDA

classification

Emphysema

COPD

unsupervised learning

Details

Title: Subtitle: GENERATIVE METHOD TO DISCOVER EMPHYSEMA SUBTYPES WITH UNSUPERVISED LEARNING USING LUNG MACROSCOPIC PATTERNS (LMPS): THE MESA COPD STUDY
Creators: Jingkuan Song - Department of Biomedical Engineering, Columbia University, New York, NY, USA
Jie Yang - Department of Biomedical Engineering, Columbia University, New York, NY, USA
Benjamin Smith - Department of Medicine, Columbia University Medical Center, New York, NY, USA
Pallavi Balte - Department of Medicine, Columbia University Medical Center, New York, NY, USA
Eric A Hoffman - Department of Biomedical Engineering, University of Iowa, Iowa City, IA, USA
R Graham Barr - Department of Epidemiology, Columbia University Medical Center, New York, NY, USA
Andrew F Laine - Department of Biomedical Engineering, Columbia University, New York, NY, USA
Elsa D Angelini - Department of Biomedical Engineering, Columbia University, New York, NY, USA
Resource Type: Journal article
Publication Details: Proceedings (International Symposium on Biomedical Imaging), Vol.2017, pp.375-378
DOI: 10.1109/ISBI.2017.7950541
PMID: 28989563
PMCID: PMC5629072
NLM abbreviation: Proc IEEE Int Symp Biomed Imaging
eISBN: 9781509011728; 1509011722
ISSN: 1945-7928
eISSN: 1945-8452
Publisher: United States
Grant note: N01HC95159 / NHLBI NIH HHS P30 ES005605 / NIEHS NIH HHS N01HC95169 / NHLBI NIH HHS UL1 RR025005 / NCRR NIH HHS UL1 RR024156 / NCRR NIH HHS R01 HL077612 / NHLBI NIH HHS R01 HL093081 / NHLBI NIH HHS R01 HL121270 / NHLBI NIH HHS R01 HL112986 / NHLBI NIH HHS RC1 HL100543 / NHLBI NIH HHS
Language: English
Date published: 04/2017
Academic Unit: Roy J. Carver Department of Biomedical Engineering; Radiology; Internal Medicine
Record Identifier: 9984051764702771

Metrics

15 Record Views

7 Times Cited - Web of Science