Room: AAPM ePoster Library
Purpose: Diffusion weighted imaging (DWI) provides critical tumor biomarker information for radiotherapy. This work compares apparent diffusion coefficient (ADC) values from DWI echo planar imaging (EPI) acquired at 0.35, 1.5, and 3 T.
Methods: DWI-EPI was acquired using a NIST diffusion phantom with 13 vials of aqueous polyvinylpyrrolidone (PVP), at concentrations ranging from 0% to 50%. Diffusion weighted images were acquired on 3.0T and 1.5T wide bore systems, and two 0.35T MRgRT systems (MR-6°60Co and MR-LINAC b = 0, 500, 900, 2000s/mm²2 for the 1.5T and 3.0T systems, and b = 0, 200, 500, 800s/mm²2 for the MRgRT systems due to gradient limits. Three measurements were performed to assess repeatability. ADC values were calculated using non-linear least squares fitting method of intensity versus b-value. Post processing included denoising utilizing noise floor subtraction and eddy current correction with affine registration.
Results: ADC repeatability was consistent for all field strengths. Low ADC values were measured on both 0.35T systems, e.g., by as much as 14.4% for the lowest PVP concentration vials. Vials with a high concentration of PVP yielded poor ADCs due to complete loss of signal for the b=800s/mm²2 acquisition. Signal loss was due to the high diffusion gradient causing low signal-to-noise ratios. The ADC values and raw signal intensities changed with gantry angle on the MRI-Linac.
Conclusion: ADC values measured at low field are subject to underestimation and gantry angle variations. ADC and signal intensity correction will require future studies.
Funding Support, Disclosures, and Conflict of Interest: Dr. Green has received honoraria and travel grants from ViewRay Inc. Dr. Mutic has consulted for ViewRay and Varian. Washington University in St. Louis has a master research agreement and receives research funding from ViewRay unrelated to this study.
Not Applicable / None Entered.