Room: Exhibit Hall | Forum 8
Purpose: MRI of hyperpolarized pyruvate is an emerging technique for non-invasively quantifying tumor metabolism. The dimensionality of the measurement is often reduced by encoding spectral information through the use of spectral-spatial excitation pulses. Due to the non-renewable nature of hyperpolarized signal, slice selective excitation can result in highly dynamic signal profiles over the course of a dynamic study. Using simulation dynamic spectral-spatial slice profiles were shown to impart quantification errors in measurement of hyperpolarized metabolism.
Methods: Using a custom-built simulation platform written in MatLab and an open-source suite for designing spectral-spatial pulses developed by colleagues at UCSF (http://www.radiology.ucsf.edu/research/labs/larson/software), a pulse acquire sequence was simulated using a numerical phantom that models metabolically active perfused tissue. Simulations were run with a range of excitation angles and repetition times. The resulting dynamic spectra were characterized using a pharmacokinetic model for HP signal evolution to determine the apparent conversion rate of hyperpolarized pyruvate to lactate (Kpl). Errors imposed by the sequence were calculated as the difference between the Kpl determined from the fit and the values used in simulation.
Results: Errors in Kpl were dependent on the excitation angle and repetition time. Generally, errors were larger for large excitation angles and shorter repetition times. This is expected, as large excitation angles paired with short TR intervals will have the most dramatic slice profile non-idealities. Errors ranged from a few percent at smaller excitation angles to over 40% overestimation of Kpl.
Conclusion: This work demonstrates that slice profile errors that arise from spectral-spatial pulses can impose sizable bias in metabolic quantification. While this work focused on pulse acquire measurement, the errors imposed by spectral-spatial slice non-uniformities should translate into imaging studies. These results suggest that the highly dynamic slice profile associated with hyperpolarized pyruvate can obfuscate metabolic quantification and must be considered during protocol design.
Not Applicable / None Entered.