Hyperpolarized Xenon-129 3D Chemical Shift Imaging of Human Lungs

Magnetic resonance imaging (MRI) is in general, capable of providing an immense amount of structural and functional information. Proton MRI in the lungs is challenging due to the inherently low water density and short T2* in lung tissues compared to other tissues in the human body. However, there has been progress in proton MRI of the lungs using ultra-short echo time sequences to overcome these challenges. An alternative to proton imaging is hyperpolarized (HP) gas imaging, which uses Xenon-129 (Xe-129) and Helium-3 (He-3) as inhaled gaseous contrasting agents, and has allowed for unique approaches in evaluating lung structure and function. In this work, we demonstrated the feasibility of 3D Single Breath Chemical Shift Imaging (3D-SBCSI) for assessing regional lung ventilation and gas uptake and exchange in 3D within a single breath hold (less than 10 seconds). Having this regional information of the lungs may allow for a better understanding of disease progression. Additionally, we present a post-processing software package termed “Tools for Automated Spectral Processing” (TASP) for the automated processing and quantification of 3D-SBCSI data.

In this study, a total of 17 subjects including: 6 healthy, 8 interstitial lung disease (ILD), and 3 lung cancer (LC) subjects underwent 3D-SBCSI. To visualize the regional exchange of Xe-129 from the alveoli into the tissue and red blood cell (RBC) compartments, ratio maps of Tissue/Gas and RBC/Gas were generated. Tissue/RBC maps were also generated to visualize the Xe-129 exchange between the tissue and RBC compartments. Healthy subjects showed a uniform distribution of gas signal intensities and Tissue/RBC ratios throughout the lungs, whereas ILD subjects showed a heterogeneous distribution. Moreover, ILD subjects had a significantly higher mean Tissue/RBC than healthy subjects. One of the advantages of 3D-SBCSI was that it acquired the whole spectrum for each voxel, which allowed for the identification of previously unknown MR spectral peaks. With this technique, a new peak was identified in a lung cancer subject. This new peak was found to be associated with the presence of a tumor.