Automated quantification of lung structures from optical coherence tomography images

Area of Science:

• Pulmonary Medicine
• Medical Imaging
• Biomedical Engineering

Background:

• Accurate characterization of lung structure size is crucial for assessing respiratory diseases.
• Existing methods for lung structure analysis can be labor-intensive and subjective.
• Optical coherence tomography (OCT) offers high-resolution imaging of biological tissues.

Purpose of the Study:

• To develop and validate a fully automated algorithm for segmenting and quantifying lung structures in OCT images.
• To characterize lung structure size using the stereological measure of median chord length.
• To explore the potential of this method as an in vivo indicator of respiratory disease-related structural remodeling.

Main Methods:

• A fully automated segmentation and quantification algorithm was developed for OCT images.
• The algorithm delineates lung structures and calculates their size using median chord length.
• The method was demonstrated on ex vivo pig and rat lung tissues using OCT needle probes.

Main Results:

• The automated algorithm accurately segmented and quantified lung structures.
• Computed estimates of lung structure size were validated against manual measurements.
• 3D visualizations of lung structures were generated from the segmentation data.

Conclusions:

• The developed automated algorithm provides a robust method for lung structure size characterization in OCT images.
• This technique shows potential for in vivo assessment of structural changes in respiratory diseases like COPD and pulmonary fibrosis.
• The algorithm facilitates objective and efficient analysis of lung tissue morphology.