Sparse multispectral processing for accelerated EPR oximetry in three dimensions

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Electron paramagnetic resonance (EPR) spectroscopy enables noninvasive measurement of free radicals in biological samples. EPR can be used to measure the partial pressure of oxygen by observing oxygen-induced broadening in the lineshape of an introduced paramagnetic probe. In tumors, the oxygen concentration is useful in determining the response to different treatment options. Likewise, the presence of oxygen plays a critical role in the pathophysiology of myocardial injury during both ischemia and subsequent reperfusion. Therefore, the ability of EPR to measure oxygen over time can provide vital information to characterize the progression of a disease state and to determine the efficacy of different treatment options. Unfortunately, long data acquisition times have limited the use of EPR for these applications. The objective of this proposal is to develop and verify processing methods that reduce acquisition time for mapping partial pressure of oxygen in three dimensions using EPR imaging. The two aims in the research plan present a strategy for reducing by a factor of 80 the acquisition time when imaging with particulate spin probes. The opportunity for the proposed improvement in acquisition time is created by the recent development of particulate paramagnetic probes. In the first aim, the proposed approach fully utilizes all energy in the resonance signal by jointly processing multiple harmonics of both the absorption and dispersion components. With multi-spectral processing, we access approximately four times more signal energy than is present in the first harmonic absorption spectrum alone. In the second aim, we incorporate the sparse, but unstructured, distribution of spins presented by particulate probes. These two aims provide processing gains that result in fewer projections, narrower sweep widths and faster sweep rates for achieving a desired image resolution. Preliminary results demonstrate feasibility of the hypothesized ambitious reductions in acquisition time. First, our team has developed particulate spin probes capable of sensing and reporting cellular and tissue pO2 with remarkable oxygen sensitivity (better than 0.1 Torr), repeatability, bio-stability, and narrow line width. Second, for two dimensional spectral-spatial imaging we have demonstrated a 38:1 improvement in acquisition time by directly estimating Lorentzian line parameters, rather than first imaging a spectral-spatial object. Third, we have demonstrated oxygen mapping in three dimensions using sparse distribution of particulate spin probes to reduce the number of projections from 1024 to 32. Thus, preliminary results give optimism for turning hours to minutes and minutes to seconds. PUBLIC HEALTH RELEVANCE: Electron paramagnetic resonance (EPR) oximetry provides a direct, minimally invasive measure of oxygen concentration. Oximetry is useful in predicting the response of tumors to different treatment options and in understanding the pathophysiology of cardiac recovery following injury due to cardiovascular disease. However, weak signal strength and long data acquisition times prohibit widespread use of EPR oximetry. The objective of this proposal is to develop signal processing methods that accelerate data acquisition. The approach fully accesses all observable signal energy in the radio frequency resonance signal and exploits the inherent sparseness of newly developed particulate spin probes. Preliminary results demonstrate the feasibility of reducing acquisition time from tens of minutes to tens of seconds.