Driehuys, BastiaanDai, Haoran2023-06-082023https://hdl.handle.net/10161/27849<p>Hyperpolarized (HP) 129Xe Magnetic Resonance Imaging (MRI) is a promising approach for the non-invasive diagnosis of pulmonary pathophysiology and has recently received FDA approval. This technique has emerged as a valuable diagnostic tool for investigating various lung diseases and holds significant potential for non-invasive investigation of lung disorders. HP 129Xe MRI is typically utilized to evaluate regional pulmonary ventilation and gas exchange functions. Hyperpolarized (HP) 129Xe MR spectroscopy has emerged as an equally valuable addition to the repertoire of imaging diagnostic tools available. The application of HP 129Xe MR spectroscopy enables us to investigate blood oxygenation levels, as evidenced through 129Xe red blood cell (RBCs) shifts. It also provides an insight into hemodynamics through cardiogenic oscillations. Despite its potential clinical utility, the use of HP 129Xe MR spectroscopy has been limited by the lack of established standards for measuring and interpreting its signal quality. In particular, there is currently no consensus on how to calculate the signal-to-noise ratio (SNR) of HP 129Xe spectroscopic data or how to determine the appropriate dose of HP 129Xe necessary to achieve high-quality NMR signals suitable for clinical analysis. To address these issues, this study developed a reliable approach in calculating the 129Xe MR spectroscopic SNR from clinical data using recommended acquisition protocols. We focused on developing a method for estimating the SNR that takes into account the inherent variability of HP 129Xe signals and the effects of noise and artifact, such as those arising from cariogenic signal oscillations, in the acquired data. Our method involves acquiring multiple repetitions of the HP 129Xe signal and using a combination of statistical techniques and signal processing algorithms to estimate the SNR from the resulting time series data. In addition to developing a method for calculating the HP 129Xe SNR, we also investigated the relationship between the administered dose equivalent of the HP 129Xe and the resulting SNR. By acquiring NMR signals over a range of doses, we were able to establish a quantitative relationship between dose and SNR that can be used to guide the selection of an appropriate HP 129Xe dose for a given clinical application. In addition to the findings of a recent investigation that evaluated the quality of scans conducted in our laboratory with a 5-point Likert scale, our study has defined empirical thresholds for requisite SNR and dosage necessary to achieve high-quality static measurements of RBC/Membrane ratio, RBC chemical shifts, and dynamic measurements of RBC amplitude oscillations using HP 129Xe MRS. Overall, our approach represents an important step towards standardizing the use of HP 129Xe MR spectroscopy in clinical practice. Through the provision of a trustworthy and quantitative measurement of SNR, as well as the establishment of a clear correlation between dosage and SNR, our study ensures that the acquisition and analysis of HP 129Xe MRS scan is executed in a consistent and reproducible manner across different institutions and studies. Besides its potential impact on clinical practice across different sites, this study also has implications for ongoing research efforts. By providing a more robust method for measuring signal quality, this study will be helpful to facilitate future studies aimed at advancing our understanding of the underlying physiology of pulmonary disease and evaluating the efficacy of new treatments. </p>Medical imagingMaster's thesis