Semiautomated Ventilation Defect Quantification in Exercise-induced Bronchoconstriction Using Hyperpolarized Helium-3...

Area of Science:

• Medical Imaging
• Pulmonary Medicine
• Computational Biology

Background:

• Exercise-induced bronchoconstriction (EIB) affects lung ventilation.
• Accurate quantification of ventilation defects is crucial for diagnosis and management.
• Hyperpolarized helium-3 magnetic resonance imaging (MRI) is a sensitive tool for assessing lung ventilation.

Purpose of the Study:

• To compare the performance of a semiautomated adaptive K-means segmentation method with manual segmentation for ventilation defects in EIB using hyperpolarized helium-3 MRI.
• To evaluate the speed, reproducibility, and agreement of the adaptive K-means method against manual segmentation.

Main Methods:

• Six subjects with EIB underwent hyperpolarized helium-3 MRI and spirometry at baseline, post-exercise, and recovery.
• Ventilation defects were segmented manually by two readers and by an adaptive K-means algorithm.
• The adaptive K-means method incorporated coil sensitivity correction and a vesselness filter.

Main Results:

• The adaptive K-means method was approximately five times faster than manual segmentation with a bias under 2%.
• Both methods showed strong, comparable correlations between spirometric measures (FEV1/FVC) and ventilation defect percent (VDP).
• Neither method revealed significant apical/basal or anterior dependence of VDP in this cohort.

Conclusions:

• The adaptive K-means method offers a faster, reproducible, and comparable alternative to manual segmentation for VDP assessment in EIB.
• This semiautomated approach has potential applications in diagnosing and monitoring various lung diseases.