Optimizing MRI Readout of Fricke Gel Dosimeters On a 1.5T MR-Linac
M MacCallum1*, I Malajovich2, S Serai3, T Salzillo4, A Dresner5, G Ibbott6, L Kim7, (1) University of Pennsylvania, Philadelphia, PA, (2) MD Anderson Cancer Center at Cooper, Camden, NJ, (3) CHOP, Gladwyne, PA, (4) MD Anderson Cancer Center, Houston, TX, (5) Philips Healthcare, ,,(6) The American Board of Radiology, Tucson, AZ, (7) MD Anderson Cancer Center at Cooper, Camden, NJ
Presentations
PO-GePV-M-261 (Sunday, 7/10/2022) [Eastern Time (GMT-4)]
ePoster Forums
Purpose: The increasing availability of 1.5T MR-Linacs brings renewed relevance of Fricke gel dosimetry for machine QA and patient-specific-QA. The primary method of imaging Fricke gels in the past has been T1-mapping; however, the current version of the Unity MR-Linac (Elekta AB) makes T1-mapping a challenge. In this work, we present solutions for overcoming the Unity’s issues of rescaling and acquiring T1-maps. These solutions include using T1-weighted imaging and using off-line processing to obtain T1-maps.
Methods: We created and optimized imaging protocols including T1-TSE and T1-weighted (T1-W) with varying flip angles (VFA) and TR time. Protocols were designed to avoid automatic rescaling between scans and allow the creation of T1-maps. The VFA images were converted to T1-maps and quantitative analysis of scans was performed using open-source pMRI software and in-house MATLAB code. Sample sizes were 1-cm², 3.5-mL cuvettes or 50-mL centrifuge tubes with 2.9-cm diameter. We modified standard FXG formulation by removing xylenol orange to increase sensitivity of the gels. Samples were placed in water, and half of each sample was irradiated to doses between 0-6Gy using a 7MV FFF beam.
Results: We evaluated our readout techniques by obtaining contrast-to-noise ratios (CNR) as a function of dose. We compared T1-W, T1-TSE and T1-map images. T1-map and T1-W images show the CNR increasing with dose, as expected. The T1-TSE images were obtained without rescaling-avoidance; as a result, the CNR increase with dose increase was masked. Preliminary data shows T1-TSE images are most sensitive in the <2Gy range, while T1-maps show the best linearity and contrast in the 3-6Gy range.
Conclusion: We present multiple methods for imaging Fricke gels in clinically relevant dose ranges, and tools for overcoming imaging limitations on a commercial 1.5T MR-Linac. Further work on T1 map generation and T1-TSE image optimization is currently in progress.
Keywords
Image-guided Therapy, Quality Assurance, Gel Dosimeter
Taxonomy
Contact Email