An investigation of the accuracy of an IMRT dose distribution using two- and three-dimensional dosimetry techniques

Abstract

Complex dose delivery techniques like intensity-modulated radiation therapy (IMRT) require dose measurement in three dimensions for comprehensive validation. Previously, we demonstrated the feasibility of the "PRESAGE/optical-computed tomography (CT)" system for the three-dimensional verification of simple open beam dose distributions where the planning system was known to be accurate. The present work extends this effort and presents the first application of the PRESAGE/optical-CT system for the verification of a complex IMRT distribution. A highly modulated 11 field IMRT plan was delivered to a cylindrical PRESAGE dosimeter (16 cm in diameter and 11 cm in height), and the dose distribution was readout using a commercial scanning-laser optical-CT scanner. Comparisons were made with independent GAFCHROMIC EBT film measurements, and the calculated dose distribution from a commissioned treatment planning system (ECLIPSE). Isodose plots, dose profiles, gamma maps, and dose-volume histograms were used to evaluate the agreement. The isodose plots and dose profiles from the PRESAGE/optical-CT system were in excellent agreement with both the EBT measurements and the ECLIPSE calculation at all points except within 3 mm of the outer edge of the dosimeter where an edge artifact occurred. Excluding this 3 mm rim, gamma map comparisons show that all three distributions mutually agreed to within a 3% (dose difference) and 3 mm (distance-to-agreement) criteria. A 96% gamma pass ratio was obtained between the PRESAGE and ECLIPSE distributions over the entire volume excluding this rim. In conclusion, for the complex IMRT plan studied, and in the absence of inhomogeneities, the ECLIPSE dose calculation was found to agree with both independent measurements, to within 3%, 3 mm gamma criteria.

Figure 1

Frontal (a), sagital (b), and axial (c)–(e) views of the PRESAGE™ dosimeter showing the segmented structures used for treatment planning; the external contour (BODY), OAR, and PTV, respectively. The locations of the three axial views (c–e) are indicated as the dotted lines 1–3 in (a). In the axial view, the multicomponent PTV appears to gradually change from an upset schematic face (c), to neutral (d), and to a happy schematic face (e). There is a 4 mm margin between the PTV and the surrounding OAR. Delivery of a homogenous uniform dose to the multicomponent PTV, while minimizing the dose to the surrounding OAR, represents an extremely challenging treatment planning problem for the ECLIPSE® IMRT algorithm.

Figure 2

(a) The calibration curve for EBT film to 6 MV radiation. (b) After the PRESAGE™ dosimeter was scanned by optical-CT, it was cut into four axial sections. The locations of the cut planes correspond to the optimal representations of the upset, neutral, and happy face representations shown in Fig. 1. EBT films (∼16 cm diameter) were inserted at the cut planes, as shown, to provide independent measurement of the planar dose distributions in these planes.

Figure 3

A complete set of optical-CT scan data of the PRESAGE™ dosimeter acquired with the OCTOPUS™ optical-CT scanner. Data from the central slice are shown and are representative of all other slices. The first row [(a) and (b)] shows the pre- and post-irradiation sinograms of projection data for the entire slice (laser light intensity, x axis=projection number, y axis=mm along the projection). Individual pre- and post-irradiation projection profiles are shown in (c). Pre- and post-irradiation reconstructed optical attenuation maps are shown in (d) and (e), respectively. Corresponding profile data through the reconstructed images along the indicated dashed lines are shown in (f).

Figure 4

Comparison of dose distributions in the three selected slices (see Fig. 1) between PRESAGE™, ECLIPSE®, and EBT film. The first row images [(a)–(c)] are dose distributions from PRESAGE™. The second [(d)–(f)] and third [(g)–(i)] rows are corresponding images from the ECLIPSE® and EBT distributions. The cross mark in (b) indicates the point where the dose distributions were normalized to convert to relative dose. The percent isodose lines (100%, 90%, 80%, 70%, 50%, and 30%) are superimposed onto the dose maps to aid comparison.

Figure 5

Line profiles through the ECLIPSE®, PRESAGE™, and EBT dose distributions along the dashed lines shown in Fig. 4i.

Figure 6

Isodose overlay plots of the 100%, 90%, 80%, 70%, 50%, and 30% isodose lines for the three selected slices (Fig. 1) between PRESAGE™, ECLIPSE®, and EBT film dose distributions. The first row [(a)–(c)] compares isodose lines between EBT (blue) and ECLIPSE® (red). The second row [(d)–(f)] compares isodose lines between EBT (blue) and PRESAGE™ doses (red). The last row [(g)–(i)] compares isodose lines between ECLIPSE® (blue) and PRESAGE™ (red) doses.

Figure 7

Sagittal view of the dose distributions and superimposed isodose lines from ECLIPSE® (a) and PRESAGE™ (b). (c) is the overlay of isodose lines—100%, 90%, 80%, 70%, 50%, and 30%.

Figure 8

Dose-volume histogram comparison between the PRESAGE™ and ECLIPSE® dose distributions. (a), (b), and (c) show overlay of dose volume histograms for the body, OAR, and PTV structures, respectively.

Figure 9

Gamma maps on the central axial slice (criteria of 3% dose difference and 3 mm distance to agreement) between ECLIPSE® and EBT (a), PRESAGE™ and EBT (b), and ECLIPSE® and PRESAGE™ (c). (d) shows the gamma map between PRESAGE™ and ECLIPSE® on a sagittal view. PTV structures (white) are also overlaid. Plots (e)–(h) are the corresponding line profiles along the dashed lines.