Quantifying the spatial clustering characteristics of radiographic emphysema explains variability in pulmonary...

Area of Science:

• Pulmonary Imaging
• Quantitative CT Analysis
• Biomarker Discovery

Background:

• Current emphysema assessment in CT scans relies on density thresholding, losing spatial information.
• This simplification may obscure diverse disease patterns, clinical phenotypes, and outcomes.
• Quantifying emphysema heterogeneity is crucial for a comprehensive understanding of the disease.

Purpose of the Study:

• To compare three novel methods for quantifying emphysema heterogeneity in CT scans.
• To evaluate the association of these spatial metrics with clinical outcomes and plasma biomarkers.
• To determine if spatial analysis improves upon traditional emphysema burden measures.

Main Methods:

• Compared three spatial methods: emphysema cluster size distribution, voxel adjacencies, and spatial point process models.
• Utilized CT scans and plasma protein data from 587 individuals in the COPDGene study.
• Assessed associations with clinical measures (e.g., FEV1) and plasma biomarkers using regression models.

Main Results:

• Spatial measures effectively distinguished heterogeneous emphysema patterns with similar overall burdens.
• Average cluster size from the point process model showed stronger clinical outcome associations than existing metrics.
• Accounting for spatial clustering identified approximately 75% more associated plasma biomarkers.

Conclusions:

• Spatial quantification of emphysema heterogeneity provides deeper insights than traditional density-based measures.
• Advanced spatial analysis, particularly average cluster size, is a more informative imaging biomarker.
• Incorporating spatial clustering enhances the discovery of emphysema-related plasma biomarkers.