Program Information

Bone Detection in MR Images and Absorbed Dose in a Material Behind Bones in Radiotherapy

J Korhonen*, M Kapanen, J Keyriläinen, T Seppälä, L Tuomikoski, M Tenhunen, Helsinki University Central Hospital, Department of Oncology, Helsinki, Finland

SU-E-J-54 Sunday 3:00:00 PM - 6:00:00 PM Room: Exhibit Hall

Purpose: To determine whether bones could be localized accurately by using MR images only in radiotherapy treatment planning. Furthermore, to measure absorbed dose in a material behind different parts of the bone, and to evaluate dose calculation error in a pseudo-CT image by assuming a single electron density for the bones.

Methods: A dedicated phantom was constructed using fresh deer bones and gelatine. The accuracy of the bone edge location and the bone diameter in MR images were evaluated by comparing those in the images with the actual measures. The absorbed dose behind the bones was measured by a matrix detector at 6 and 15 MV. The dose calculation error in the bulk density pseudo-CT image was quantified by comparing the calculation results with those obtained in a standard CT image by superposition and Monte Carlo algorithms (TPSs: Xio 4.60 and Monaco 3.00, Elekta CMS Software).

Results: The examination of bone position revealed that the bones can be localized within a 1-mm-pixel-size in the MR images. The measured dose behind less than 2.5-cm-thick femur indicated that the absorbed dose behind the middle part of the bone is approximately one percentage unit (6 MV: 1.3%, 15 MV: 0.9%) smaller than that of the physically narrower bone edge. The calculations illustrated that the bulk density pseudo-CT image used causes errors up to nearly 2 % to the dose behind the middle part, but also, the edge of the femur.

Conclusions: This research ascertains that the bone localization is not a restrictive issue for radiotherapy treatment planning by using MR image only. The work indicates also that the decrease in absorbed dose is not necessarily dependent on the diameter of the bone. Future research should investigate the generation of more complex pseudo-CT images and the dose calculations by using these.

Contact Email