Hyperpolarized Gas Diffusion MRI using Steady State Free Precession Pulse Sequences

Hyperpolarized noble gas magnetic resonance imaging (MRI) provides a unique view of the airspaces in human lungs. However, images created with this technique have a fundamental resolution limit due in part to the gas diffusion within the air spaces during the image acquisition. The process of diffusion can be used to provide a method for extracting structural information below the resolution limit, via short-time diffusion MR. In free space, the area a gas particle explores in a given amount of time (the diffusion coefficient) is a constant that does not depend on the duration over which the measurement is made. In a highly restrictive area like the lung airspaces, the diffusion coefficient varies greatly with the duration of the measurement. For very short times measurement times, the diffusion coefficient approaches the free space value, while at longer measurement times the surrounding walls prevent the particle from traveling. This time dependence is related to the surface to volume ratio of the confining space. The goal of this work was to develop a method of making diffusion-weighted measurements at diffusion times less than∼ 1 ms to detect this time dependence in restrictive environments such as the human lung. In order to make measurements at these short times, we turned to an MRI technique known as Steady State Free Precession (SSFP). SSFP pulse sequences are coherent, which means the transverse magnetization is not zero at the application of the next RF pulse. An advantage of iii using an SSFP pulse sequence is that it produces a very high signal level on which to measure the small diffusion attenuation imparted by short-time measurements. We developed several modifications to an SSFP pulse sequence which include diffusion sensitization, and investigated the behavior of each of these methods through the use of a magnetization simulation. We made global apparent diffusion coefficient (ADC) measurements, as well as created images with the resulting pulse sequences. In the global version, we were able to make ADC measurements over a range of diffusion times from 300-800 µs in glass-bead phantoms and fit the time-dependent ADC to extract the packing volume fraction φ for each of the phantoms. Multiple diffusion-time global ADC measurements made in human subjects highlighted the differences between healthy and emphysmatic lungs. In the imaging experiments, we generated ADC maps at a diffusion time of 500 µs in several human subjects.

Note: Abstract extracted from PDF text