Premium
SU‐GG‐J‐112: Integration and Verification of 3T Magnetic Resonance Spectroscopic Imaging in Radiation Therapy Treatment Planning
Author(s) -
Muruganandham M,
Bayouth J,
Kearney W,
Smith M,
Buatti J
Publication year - 2008
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.2961661
Subject(s) - magnetic resonance spectroscopic imaging , imaging phantom , nuclear medicine , scanner , radiation treatment planning , voxel , image registration , magnetic resonance imaging , computer science , radiation therapy , medicine , artificial intelligence , radiology , image (mathematics)
Purpose: The purpose of this work is to test the feasibility of MRSI integration into brain tumor RT planning using a custom‐built MRSI head phantom as well as a case study. Method and Materials: A special MRSI head phantom was constructed with spatially well‐defined tube inserts. Each tube contains different pre‐determined concentrations of metabolite solutions that mimic different degrees of brain tumor burden and were immersed in a background solution that emulate normal brain metabolite levels. The 3D MRSI data were acquired on a Siemens 3T TRIO TIM scanner using PRESS (point resolved spectroscopy) sequence with TE/TR: 135/1510 ms, 16×16×8 matrix, FOV: 16×16×8 resulting in MRSI voxel size of 10×10×10 mm 3 . Registration of MRSI (gray scale DICOM) data with planning CT and creation of metabolite‐derived target contours were performed on Pinnacle treatment planning system. Target contours created using MRSI‐derived tumor metabolic abnormalities were compared with those created based on morphological (T2 and T1 contrast weighted) MRI scans. Results: The CT‐MR (T1) registration accuracy was found to be within 1 mm, with no spatial distortion during MRSI. Registration errors were found in the TPS due to field‐of‐view size and origin difference between studies. These were corrected using a simple algorithm to determine the translation. Re‐scaled MRSI pixel values allowed automated segmentation of metabolic ratios and were consistent with those identified on the control console of the MRI scanner. When applied to a clinical case, a 42 cc (or 58%) reduction in the treated volume of the PTV was possible, assuming the CNI ratio of 2.5 included all microscopic extension of disease requiring the boost dose. Conclusion: MRSI derived functional information can be incorporated into RT planning for brain tumors well within the registration tolerance limits of current planning process.