Proton MRI

Conventional proton magnetic resonance imaging (MRI) of the lung has historically suffered from a poor signal-to-noise ratio (SNR), due to several factors, including the low tissue density, fast signal relaxation, and respiratory/cardiac motion. The image quality in proton lung MRI can be improved using the recently developed ultrashort echo time (UTE) approaches that acquire the proton tissue signal before it decays substantially. UTE MRI allows for direct visualization of lung tissue previously not possible with MRI and with image quality rivaled only by x-ray computed tomography (CT), but without the associated radiation dose. MRI strategies that acquire low frequency information first, such as UTE, are also inherently less sensitive to motion, which is expected to be a considerable benefit for imaging neonates, prone to spontaneous movement.

Comparison of proton MRI using a conventional fast low angle shot (FLASH) sequence (left) and a spiral ultrashort echo time (UTE) sequence (right). The top row shows images from a pediatric healthy volunteer, and the bottom row shows images from a pediatric cystic fibrosis (CF) patient. All images shown above were acquired in separate 10 second breath-holds. The UTE images show an improved tissue signal from the lung parenchyma, and images acquired in CF show are greater sensitivity to structural abnormalities, including mucous plugging and bronchiectasis (arrows).

Comparison of high-resolution UTE imaging performed in a healthy adult (left) and a healthy pediatric participant (right). The UTE images were acquired during free breathing with prospective gating navigators to reduce motion effects. Subjects were asked to breathe normally throughout the 5 – 10 minute acquisition, and the final reconstructed image has a 1.25 mm3 isotropic resolution.

Fourier decomposition

Fourier decomposition (FD) is a proton-based MRI technique that uses lung images acquired during free-breathing. A single slice image is acquired at a frame rate of approximately 3 images per second. Since the patient is free-breathing, the resulting images capture motion from both the lungs and heart. Following image registration, a pixel-by-pixel Fourier decomposition of the proton signal in the time domain yield maps that are related to ventilation and perfusion. FD MRI has been successfully validated with comparisons to single photon emission computed tomography (SPECT), dynamic contrast enhanced perfusion MRI, and hyperpolarized gas MRI. In patients with cystic fibrosis, FD MRI ventilation maps can potentially visualize ventilation defects; however, FD MRI has not yet been directly compared to hyperpolarized gas MRI in cystic fibrosis patients.