Elastic parametric response mapping: quantitative CT scoring for local COPD severity

Area of Science:

• Pulmonary Medicine
• Radiology
• Medical Imaging Analysis

Background:

• Current quantitative CT scans aid chronic obstructive pulmonary disease (COPD) phenotyping.
• These techniques do not fully capture spatial variability or reversible disease in lung parenchyma.

Purpose of the Study:

• To develop and validate elastic parametric response mapping (ePRM) for detailed lung tissue characterization in COPD.
• To assess the longitudinal changes in lung subvolume classification using ePRM.
• To correlate ePRM findings with changes in lung function (FEV1).

Main Methods:

• Elastic principal graphing applied to CT scans from the Genetic Epidemiology of COPD study.
• Development of a tiered scoring system (ePRM tiers 0-3, Op) classifying lung subvolumes.
• Longitudinal analysis of 3631 participants over 5 years, stratified by spirometry (no obstruction, GOLD 1-4).

Main Results:

• Worsening airflow obstruction correlated with increased proportions of tier 2 and 3 subvolumes.
• Tier 1 subvolumes showed potential for reversion to tier 0, especially in less severe obstruction.
• Baseline mean tier 1 position strongly predicted future disease progression across all obstruction levels.
• Higher proportion of lung in tier 1 was associated with slower FEV1 decline.

Conclusions:

• CT-based ePRM effectively categorizes local lung tissue into distinct tiers of disease severity.
• These tiers represent potentially reversible stages of COPD.
• ePRM offers a novel approach to understanding COPD heterogeneity and progression.