Grant: Travel Awards - Institute Community Support

Ultra-short time to echo (UTE) magnetic resonance imaging (MRI) allows us to visualize lung structure and recognize abnormalities. However, MR signal intensity (SI) can vary greatly between healthy and/or disease participants and over short periods of time, even when images are completed within 5 minutes of each other. This limits our ability to reliably measure lung densities. We developed a method to correct SI within the lung. We analyzed free-breathing UTE MRI at baseline and 3-months later in participants with idiopathic pulmonary fibrosis (IPF). The lungs were automatically segmented using and the body was manually segmented. The SI of air and blood in UTE MRI was measured. The lungs were subtracted from the body segmentation to generate a tissue mask, which was used to generate a signal variation map. The original uncorrected UTE MR image was divided by the signal variation map; the median value of air and blood were used to clip (SI<air equal air, SI>blood equal blood) and correct [SIcorrected=(SIuncorrected-SIair)/(SIblood-SIblood)] the SI within the lung, generating corrected image. After correction, the median SI within the lungs was similar at baseline and 3-month follow-up. The percent change from baseline at 3-months for corrected was smaller than for uncorrected. SI in stable IPF was lower (better) than SI in progressive IPF after correction, but was similar before correction. This method may be used to develop quantitative lung density measurements on UTE MRI to assess differences in disease populations, disease progression and treatment response without the ionizing radiation inherent to gold-standard computed tomography.