a) Scaling up tissue oxygen imaging by EPR imaging in larger objects: To enable study of larger size objects by time-domain Electron Paramagnetic Resonance Imaging, we implemented the RF chain, with the associated pulse programming;receive chain and the digitizer/average in a magnet/gradient assembly with a clear 80 cm bore. This allows placement of larger objects in the active site of the magnet. However, resonant structures capable of assessment of larger objects were not available. We developed a proof-of-concept resonant structure with four elements in different configurations to allow studies on larger objects that cannot be fully covered by a single resonator element. Each of the four elements in the resonant structure was arranged in a plane or at suitable angles with the other elements to mimic a volume resonator. The image from a larger object was obtained by computationally combining images collected from each individual element. By multiplexing or interleaving measurements from each element, we have demonstrated the feasibility to study larger objects relevant to human scale and at the same time reducing the radiofrequency power requirements, while increasing the sensitivity. A paper on this is published in Magnetic Resonance in Medicine. Enomoto A, Hirata H, Matsumoto S, Saito K, Subramanian S, Krishna MC, Devasahayam N. Four-channel surface coil array for 300-MHz pulsed EPR imaging: Proof-of-concept experiments. Magn Reson Med. 2013 b) Implementation of Novel Image data acquisition to improve temporal resolution: Image data acquisition in EPR imaging is fundamentally limited as a result of the necessity of using static magnetic field gradients. However, in conventional MRI, gradients are switched on and off allowing the use of partial k-space imaging which allows significant acceleration of image data acquisition without a loss in resolution. We have implemented partial k-space imaging in EPR even with the use of static field gradients by making use of the Hermitian symmetry of the k-space or by a judicious choice of sub-regions of k-space. This allowed significant reduction in imaging times. This image data acquisition approach has been implemented in all the scanners and the method has been reported. Subramanian S, Chandramouli GV, McMillan A, Gullapalli RP, Devasahayam N, Mitchell JB, Matsumoto S, Krishna MC Evaluation of partial k-space strategies to speed up time-domain EPR imaging. Magn Reson Med. 70, 745-753 (2013) c) Gridding and k-space extrapolation for improving temporal resolution in EPRI: We examined gridding as a method to reconstruct images with equal field of view to enable generating reliable pO2 maps from a single data set compared to the current practice of generating pO2 maps from three data sets. By utilizing gridding and k-space extrapolation, we have been able to generate accurate pO2 maps from a single data set, thereby realizing a 60 % saving in image collection time without a loss in resolution. A paper on this approach has been accepted.Jang H, Subramanian S, Devasahayam N, Saito K, Matsumoto S, Krishna MC, McMillan AB. Single acquisition quantitative single-point electron paramagnetic resonance imaging. Magn Reson Med. 2013. d) Examining the capability of pO2 maps from EPRI to guide therapy. Tumor pO2 status is an important determinant in response to radiotherapy or chemotherapy. Here we tested if EPRI can predict the treatment outcome of radiotherapy or chemotherapy with a hypoxic cytotoxin, TH-302 using three pancreatic carcinoma xenografts, HS766t, MiaPaca2, and Su8686. The tumor pO2 in these follow the order Su8686 (17 mm Hg) MiaPaca2 (12 mm Hg) Hs766t (9 mm Hg). The radio-response followed the order Su8686 MiaPaca2 Hs766t where as the response to TH-302 followed the order Hs766tMiaPaca2 Su8686. These results support the value of EPRI derived pretreatment pO2 values in guiding therapeutic decisions. e) Pre-treatment pO2 predicts treatment outcome in radiotherapy: It is known that hypoxia is an important determinant in response to treatment with radiotherapy. We selected MiaPaca2 tumor xenograft and measured pre-treatment pO2 values and monitored the time for tumor to reach 2.5 times the pretreatment values. We found that the more hypoxic the tumor is, the faster it took to reach the endpoint. This once again support the value of EPRI in predictive capability. f) Post-radiation reoxygneation profile of tumor pO2. In clinic radiotherapy is conventionally given as daily fractions to allow the tumor to reoxygenate to respond better to the subsequent fractions. This is currently done empirically. We studied tumor reoxygenation profile in SCCVII tumor after a 3 Gy single dose and monitored tumor pO2 values for the next 24 hours. We found that the tumor pO2 decreases immediately after radiation following a recovery to the pre-treatment value 24 hours post radiation.