Related Experiment Video
Updated: Aug 5, 2026
02:09
Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
Published on: April 12, 2024
Radiomics of the Airway (RadAr): Multi-Scale Airway Phenotyping for Disease Characterization on Routine CT Imaging
Pushkar Mutha1, Juyoung Lee2, George Lucas Silva3
1Wallace H. Coulter Department of Biomedical Engineering, Emory University and Georgia Institute of Technology, Atlanta, GA, USA.
Medrxiv : the Preprint Server for Health Sciences
|August 1, 2026
Summary
Radiomics of the Airway (RadAr) offers advanced airway phenotyping from CT scans, identifying disease-specific signatures linked to outcomes in various respiratory conditions. This tool aids in diagnosis, risk stratification, and monitoring for lung diseases.
Area of Science:
- Pulmonary Medicine
- Radiology
- Medical Imaging Analysis
Background:
- Airway remodeling is a common feature in many respiratory diseases.
- Current CT imaging tools have limitations in comprehensively characterizing the airway tree.
- There is a need for advanced imaging analysis techniques to better understand airway pathology.
Purpose of the Study:
- To introduce Radiomics of the Airway (RadAr), an automated framework for multi-scale airway phenotyping using routine chest CT scans.
- To develop a tool for detailed characterization of airway dimensions, tapering, architectural distortion, and global morphology.
- To provide an interactive platform for analysis and visualization of airway phenotypes.
Main Methods:
- RadAr was evaluated in four distinct clinical settings: fibrotic interstitial lung disease (fILD), COVID-19, progressive pulmonary fibrosis (PPF), and pediatric cystic fibrosis (CF).
- The framework extracts multi-scale airway measurements and was used for mortality prediction, disease severity prediction, structure-function association, and structure-inflammation marker analysis.
- Unsupervised clustering was employed to identify distinct airway phenotypes within the fILD and COVID-19 cohorts.
Main Results:
- In fILD, lower-lobe architectural distortion predicted mortality.
- In COVID-19, luminal dilation was independently associated with severe disease.
- Airway phenotypes in PPF correlated with lung function (e.g., forced vital capacity) and gas exchange impairment.
- In pediatric CF, airway morphology changes were linked to exacerbations and inflammation.
- Five distinct airway phenotypes were identified, showing a spectrum from healthy to severely remodeled airways, associated with increasing disease severity and mortality.
Conclusions:
- RadAr successfully identified interpretable, disease-specific airway signatures across diverse lung diseases (restrictive, obstructive, mixed).
- These signatures are associated with clinical outcomes and physiological function in both adult and pediatric populations.
- The RadAr framework offers a scalable solution for improved diagnosis, risk stratification, and longitudinal monitoring in pulmonary diseases.
