PT  - JOURNAL ARTICLE
AU  - Felix J.S. Bragman
AU  - Jamie R. McClelland
AU  - Dave J. Hawkes
AU  - John R. Hurst
TI  - Multi-scale analysis of imaging features to study exacerbation susceptible COPD
DP  - 2014 Sep 01
TA  - European Respiratory Journal
PG  - P644
VI  - 44
IP  - Suppl 58
4099  - http://erj.ersjournals.com/content/44/Suppl_58/P644.short
4100  - http://erj.ersjournals.com/content/44/Suppl_58/P644.full
SO  - Eur Respir J2014 Sep 01; 44
AB  - Despite advances in imaging, the link between the distribution of disease and pathophysiology of exacerbations in COPD is unknown.We hypothesised that a new method, using regional analysis of CT-derived imaging features representing lung structure and physiology, can track exacerbation susceptible phenotypes.The study was performed on CT images of GOLD 3 patients from the COPDGene study with exacerbation frequencies of 0 and >6 per year (both n=8). The groups had mean age 67.5 and 60.2 years, FEV1 %predicted 40.5 and 42.1 and FEV1/FVC 42.4 and 47.2 respectively.Registration of paired breath-hold images was used to derive 27 features, representing tissue structure and function over the full lung and individual lobes. These included the mean volume change (determinant Jacobian), anisotropy in strain (Var(e)), an overall measure of strain magnitude (∑(e)) and tissue density (voxel intensity).We observed marked differences between features at the lobar level, principally within lower lobes yet no significance at the whole lung scale.View this table:These pilot data suggest that regional alterations in lung structure and function can be detected, which may inform on COPD exacerbation susceptibility.