Three-dimensional visualization and measurement of conformal dose distributions using magnetic resonance imaging of...

The measurement of complex dose distributions (those created by irradiation through multiple beams, multiple sources, or multiple source dwell positions) requires a dosimeter that can integrate the dose during a complete treatment. Integrating dosimeter devices generally are capable of measuring only dose at a point (ion chamber, diode, TLD) or in a plane (film). With increasing use of conformal dose distributions requiring shaped, noncoplanar beams, there will be an increased requirement for a dosimeter that can record and display a 3D dose distribution. The use of a 3D dosimeter will be required to confirm the accuracy of treatment plans produced by the current generation of 3D treatment-planning computers.

The use of a Fricke-infused gel and magnetic resonance imaging (MRI) to demonstrate the localization of stereotactic beams has been demonstrated (11). The recently developed BANG polymer gel dosimetry system (MGS Research, Inc., Guilford, CT), based on radiation-induced chain polymerization of acrylic monomers dispersed in a tissue-equivalent gel, surpasses the Fricke-gel method by providing accurate, quantitative dose distribution data that do not deteriorate with time (6, 9). The improved BANG2 formulation contains 3% N,N'-methylene-bisacrylamide, 3% acrylic acid, 1% sodium hydroxide, 5% gelatin, and 88% water, where all percentages are by weight. The gel was poured into volumetric flasks, of dimensions comparable to a human head. The gels were irradiated with complex beam arrangements, similar to those used for conformal radiation therapy. Images of the gels were acquired using a Siemens 1.5T imager and a Hahn spin-echo pulse sequence (90 degrees-tau-180 degrees-tau-acquire, for different values of tau). The images were transferred via network to a Macintosh computer for which a data analysis and display program was written. The program calculates R2 maps on the basis of multiple TE images, using a monoexponential nonlinear least-squares fit based on the Levenberg-Marquardt algorithm. The program also creates a dose-to-R2 calibration function by fitting a polynomial to a set of dose and R2 data points, obtained from gels irradiated in test tubes to known doses. This function can then be applied to any other R2 map, so that a dose map can be computed and displayed.

Through exposure to known doses of radiation, the gel has been shown to respond linearly with dose in the range of 0 to 10 Gy, and its response is independent of the beam energy or modality. Dose distributions have been imaged in orthogonal planes, and can be displayed in a convenient form for comparison with isodose plans. The response of the gel is stable; the gel can be irradiated at any time after its manufacture, and imaging can be conducted any time following a brief interval after irradiation.

The polymer gel dosimeter has been shown to be a valuable device for displaying three-dimensional dose distributions. The imaged dose distribution can be compared easily with calculated dose distributions, to validate a treatment planning system. In the future, gels may be prepared in anthropomorphic phantoms, to confirm unique patient dose distributions.