Chronic obstructive pulmonary disease (COPD) and emphysema, jointly, are the fourth leading cause of death in the US and globally. Decades of research have not found disease-modifying therapies except for a1 antitrypsin deficiency, a rare disease found by improved phenotyping, characterized by a specific emphysema subtype and caused by gene variants in SERPINA1. This renewal builds on the successful discovery of six new quantitative emphysema subtypes (QES), which were associated independently with greater symptoms, impaired function and increased mortality in addition to several genome-wide significant associations for gene variants. In the renewal, we propose to leverage over 27,000 CTs acquired over 5-7 years of follow-up from 4,500 highly phenotyped and genotyped participants to test if deep and unsupervised learning on new state-of- the-art CTs and CT angiograms will reveal additional deep-learned and molecular QES that suggest mechanistic pathways to treatment. Further, we will collect 60 explanted lungs and use state-of-the art microCT to test if QES have distinct histology and structure. Finally, we will recruit 100 patients undergoing lung cancer CT screening to test if QES will be translatable to clinical low-dose lung cancer screening CT scans acquired on contemporary scanners. Successful completion of these aims will discover new deep QES and molecular QES and validate and translate QES to further subphenotype emphysema and facilitate testing of personalize molecular therapies and, hopefully, replicate the success of therapy for a1-antitrypsin deficiency for more common emphysema subtypes.

Public Health Relevance

In the first funding period, we discovered six QES, which are independently associated with morbidity, mortality and distinct gene variants that suggest targetable molecular pathways. This renewal builds upon our successful UML approach with integration of DL, greatly increased CT and genetic data; validates QES with histology; and translates QES to the clinic in order to facilitate testing of molecular therapies of specific QES and to redefine emphysema subtypes in clinical practice.

Agency
National Institute of Health (NIH)
Institute
National Heart, Lung, and Blood Institute (NHLBI)
Type
Research Project (R01)
Project #
5R01HL121270-06
Application #
9927683
Study Section
Biomedical Computing and Health Informatics Study Section (BCHI)
Program Officer
Postow, Lisa
Project Start
2014-08-01
Project End
2023-04-30
Budget Start
2020-05-01
Budget End
2021-04-30
Support Year
6
Fiscal Year
2020
Total Cost
Indirect Cost
Name
Columbia University (N.Y.)
Department
Internal Medicine/Medicine
Type
Schools of Medicine
DUNS #
621889815
City
New York
State
NY
Country
United States
Zip Code
10032
Smith, Benjamin M; Traboulsi, Hussein; Austin, John H M et al. (2018) Human airway branch variation and chronic obstructive pulmonary disease. Proc Natl Acad Sci U S A 115:E974-E981
Aaron, Carrie P; Schwartz, Joseph E; Hoffman, Eric A et al. (2018) A Longitudinal Cohort Study of Aspirin Use and Progression of Emphysema-like Lung Characteristics on CT Imaging: The MESA Lung Study. Chest 154:41-50
Barr, R Graham (2018) Rethinking Chronic Obstructive Pulmonary Disease. Chronic Pulmonary Insufficiency and Combined Cardiopulmonary Insufficiency. Ann Am Thorac Soc 15:S30-S34
Song, Jingkuan; Yang, Jie; Smith, Benjamin et al. (2017) GENERATIVE METHOD TO DISCOVER EMPHYSEMA SUBTYPES WITH UNSUPERVISED LEARNING USING LUNG MACROSCOPIC PATTERNS (LMPS): THE MESA COPD STUDY. Proc IEEE Int Symp Biomed Imaging 2017:375-378
Yang, Jie; Angelini, Elsa D; Balte, Pallavi P et al. (2017) Unsupervised Discovery of Spatially-Informed Lung Texture Patterns for Pulmonary Emphysema: The MESA COPD Study. Med Image Comput Comput Assist Interv 10433:116-124
Parikh, Megha A; Aaron, Carrie P; Hoffman, Eric A et al. (2017) Angiotensin-Converting Inhibitors and Angiotensin II Receptor Blockers and Longitudinal Change in Percent Emphysema on Computed Tomography. The Multi-Ethnic Study of Atherosclerosis Lung Study. Ann Am Thorac Soc 14:649-658
Feng, Xinyang; Yang, Jie; Laine, Andrew F et al. (2017) Discriminative Localization in CNNs for Weakly-Supervised Segmentation of Pulmonary Nodules. Med Image Comput Comput Assist Interv 10435:568-576
Yang, Jie; Angelini, Elsa D; Smith, Benjamin M et al. (2017) Explaining Radiological Emphysema Subtypes with Unsupervised Texture Prototypes: MESA COPD Study. Med Comput Vis Bayesian Graph Models Biomed Imaging (2016) 2017:69-80
Hoffman, Eric A; Lynch, David A; Barr, R Graham et al. (2016) Pulmonary CT and MRI phenotypes that help explain chronic pulmonary obstruction disease pathophysiology and outcomes. J Magn Reson Imaging 43:544-57
Jie Yang; Xinyang Feng; Angelini, Elsa D et al. (2016) Texton and sparse representation based texture classification of lung parenchyma in CT images. Conf Proc IEEE Eng Med Biol Soc 2016:1276-1279

Showing the most recent 10 out of 12 publications