Abstract:
MRI can improve delineation of tumor and normal tissues for radiation treatment planning in brain tumor. However, MRI cannot provide radiologic properties information as in the case of CT image. The approach can solve this problem by bulk density method. This method works by overriding the original pixel value over the interested area with average bulk density. The purpose of this study is to evaluate the dosimetric accuracy in MRI based compared with CT base treatment planning. Twenty eight brain tumor patients were scanned both CT and MRI simulator with the same position and then imported both of them to Radiotherapy Treatment Planning (RTP) unit. Registration images and delineation of tumor and OARs were created by radiation oncologist while bone and air cavity were defined with auto-segmentation in treatment planning. Determination of 3 types of bulk density: First, average individual bulk density: the average CT number of all slices for PTV, bone and air for each patient were undertaken. Second, average mean bulk density: the average individual bulk density for all slices and all patients for PTV, bone and air were calculated. And third, ICRU bulk density: recommendation from ICRU number 46 for PTV, bone and air were employed. CT full density plan were created in IMRT and VMAT as a reference plan of this study. MRI with 3 types of bulk density plans and MRI with water equivalent plan were created in both IMRT and VMAT. The plans were compared using D_95% in the dose volume histogram. The result of IMRT and VMAT plan, show that all the dose differences from CT full density plan of MRI average individual and average mean bulk density plan are within 0.27% and 1.54% for average and maximum dose difference, respectively, and the dose difference of MRI with ICRU bulk density plan were within 0.38% and 0.93% for average and maximum dose difference, respectively, The water equivalent plans were within 1.68% and 3.83% for average and maximum dose difference, respectively. The treatment planning of head region using MR image was quite accurate with bulk density method. The dose differences in both of average individual and average mean plan were comparable so the average bulk density for this group of study could be employed in the head planning instead of determining bulk density for individual. The ICRU bulk density can also be used in head planning. The MR water equivalent plan may not be suitable to use in head region.