Grant: Project Grant

Background: Cystic fibrosis (CF) is a fatal genetic disease, affecting 1 in ~3600 children born in Canada. CF lung disease is characterized by bronchiectasis and computed tomography (CT) is the gold standard for diagnosis. However, conventional CT exposes patients to radiation which limits repeated use. Methods that do not use radiation and can provided added functional information, such as magnetic resonance imaging (MRI), are preferred. While separating the airways from the lung is challenging with standard MRI, a new approach called ultra-short echo time (UTE) MRI can produce clear images of the airways. However, there are currently no ways to generate automated and accurate UTE MRI airway lumen measurements. Hypothesis: We hypothesize that UTE MRI airway lumen measurements will be repeatable, reproducible and in agreement with CT measurements, and will quantify disease-related changes in children and adults with CF compared to controls. We propose three aims: Aim 1: Technical Performance Validation: 1) Optimize UTE MRI acquisition methods for airway imaging in children and adults with CF and controls, and, 2) Evaluate agreement, repeatability and reproducibility of UTE MRI airway lumen measurements in comparison to industry-standard CT measurements in children and adults with CF and controls. Aim 2: Clinical Validation: 1) Determine if UTE MRI airway lumen measurements are significantly different in children and adults with CF compared with controls, and associated with lung function measurements, and, 2) Evaluate whether UTE MRI airway lumen measurements can detect significant differences over 2 years in children and adults with CF. Aim 3. Develop fully-automated UTE MRI airway tree segmentation by Deep Neural Networks. Expected Outcomes: We will establish the technical development and clinical validation of UTE MRI airway measurements necessary to enable translation of MRI airway imaging as a long-term monitoring tool in CF patients.