Journal article
Pulmonary emphysema subtypes defined by unsupervised machine learning on CT scans
Thorax, Vol.78(11), pp.1067-1079
11/2023
DOI: 10.1136/thorax-2022-219158
PMCID: PMC10592007
PMID: 37268414
Abstract
Background Treatment and preventative advances for chronic obstructive pulmonary disease (COPD) have been slow due, in part, to limited subphenotypes. We tested if unsupervised machine learning on CT images would discover CT emphysema subtypes with distinct characteristics, prognoses and genetic associations. Methods New CT emphysema subtypes were identified by unsupervised machine learning on only the texture and location of emphysematous regions on CT scans from 2853 participants in the Subpopulations and Intermediate Outcome Measures in COPD Study (SPIROMICS), a COPD case–control study, followed by data reduction. Subtypes were compared with symptoms and physiology among 2949 participants in the population-based Multi-Ethnic Study of Atherosclerosis (MESA) Lung Study and with prognosis among 6658 MESA participants. Associations with genome-wide single-nucleotide-polymorphisms were examined. Results The algorithm discovered six reproducible (interlearner intraclass correlation coefficient, 0.91–1.00) CT emphysema subtypes. The most common subtype in SPIROMICS, the combined bronchitis-apical subtype, was associated with chronic bronchitis, accelerated lung function decline, hospitalisations, deaths, incident airflow limitation and a gene variant near DRD1, which is implicated in mucin hypersecretion (p=1.1 ×10−8). The second, the diffuse subtype was associated with lower weight, respiratory hospitalisations and deaths, and incident airflow limitation. The third was associated with age only. The fourth and fifth visually resembled combined pulmonary fibrosis emphysema and had distinct symptoms, physiology, prognosis and genetic associations. The sixth visually resembled vanishing lung syndrome.Conclusion Large-scale unsupervised machine learning on CT scans defined six reproducible, familiar CT emphysema subtypes that suggest paths to specific diagnosis and personalised therapies in COPD and pre-COPD.
Details
- Title: Subtitle
- Pulmonary emphysema subtypes defined by unsupervised machine learning on CT scans
- Creators
- Elsa D Angelini - NIHR Imperial Biomedical Research CentreJie Yang - Columbia UniversityPallavi P Balte - Columbia University Irving Medical CenterEric A Hoffman - University of IowaAni W Manichaikul - University of VirginiaYifei Sun - Columbia University Irving Medical CenterWei Shen - Columbia UniversityJohn H M Austin - Columbia UniversityNorrina B Allen - Northwestern UniversityEugene R Bleecker - University of ArizonaRussell Bowler - National Jewish HealthMichael H Cho - Brigham and Women's HospitalChristopher S Cooper - Department of Medicine, University of California, Los Angeles, California, USADavid Couper - University of North Carolina at Chapel HillMark T Dransfield - University of Alabama at BirminghamChristine Kim Garcia - Columbia University Irving Medical CenterMeiLan K Han - University of MichiganNadia N Hansel - Johns Hopkins UniversityEmlyn Hughes - Columbia UniversityDavid R Jacobs - University of MinnesotaSilva Kasela - New York Genome CenterJoel Daniel Kaufman - University of WashingtonJohn Shinn Kim - University of VirginiaTuuli Lappalainen - Columbia University Irving Medical CenterJoao Lima - Johns Hopkins MedicineDaniel Malinsky - Columbia University Irving Medical CenterFernando J Martinez - Cornell UniversityElizabeth C Oelsner - Columbia University Irving Medical CenterVictor E Ortega - Mayo Clinic in ArizonaRobert Paine - University of UtahWendy Post - Johns Hopkins UniversityTess D Pottinger - Columbia University Irving Medical CenterMartin R Prince - Cornell UniversityStephen S Rich - University of VirginiaEdwin K Silverman - Brigham and Women's HospitalBenjamin M Smith - McGill University Health CentreAndrew J Swift - University of SheffieldKarol E Watson - University of California, Los AngelesPrescott G Woodruff - University of California, San FranciscoAndrew F Laine - Columbia UniversityR Graham Barr - Columbia University Irving Medical Center
- Resource Type
- Journal article
- Publication Details
- Thorax, Vol.78(11), pp.1067-1079
- DOI
- 10.1136/thorax-2022-219158
- PMID
- 37268414
- PMCID
- PMC10592007
- NLM abbreviation
- Thorax
- ISSN
- 0040-6376
- eISSN
- 1468-3296
- Publisher
- BMJ Publishing Group Ltd and British Thoracic Society
- Grant note
- K12HL120004; P01HL105339; R01HL113264; U01HL089856 / NIH DK063491; HHSN268201500003I; N01-HC-95159-69; UL1-TR-000040; UL1-TR-001079; UL1-TR-001420; UL1-TR-001881 / National Heart, Lung, and Blood Institute (NHLBI) Not applicable / Foundation for the NIH Not applicable / COPD Foundation (http://dx.doi.org/10.13039/100008184) N02-HL-64278 / NHLBI HHSN268200900013C-20C; R01-HL077612; R01-HL093081; R01-HL103676; R01-HL121270; R01-HL130506; R01-HL131565; R01-HL142028; T32-144442 / NIH/NHLBI
- Language
- English
- Electronic publication date
- 06/02/2023
- Date published
- 11/2023
- Academic Unit
- Roy J. Carver Department of Biomedical Engineering; Radiology; Internal Medicine
- Record Identifier
- 9984426853802771
Metrics
10 Record Views