The gantry moves during the delivery of each projection, unmodeled by the treatment planning system, describing a small arc around the bore center marked with a cross. In (a), the target near the center of the bore experiences a similar level of fluence throughout the projection, and the planning system's approximation is a good one. In (b), the beam edge gets appreciably closer to the PTV periphery at the start and end of the projection, resulting in an overestimation of dose in this area.
The TPS models each beamlet as a single gantry angle with some relative intensity, temporally located at the center of the projection. The modified version of CheckTomo decomposes this single gantry angle into several subprojections, better approximating the true delivery technique shown at the bottom of this figure.
Observed change in D95% and D2% for plan M between SP = 5 and SP = 1 calculations for a number of different slice spacings, expressed as (inplane spacing, longitudinal spacing).
Overprediction of D95% in calculations utilizing different numbers of subprojections, compared with the case with 21 subprojections modeled per projection.
Overprediction of D2% in calculations utilizing different numbers of subprojections, compared with the case with 21 subprojections modeled per projection.
Profiles of SP = 1 (bold) and SP = 21 (dashed) calculations for plan F (a) and (b) and plan D (c) and (d). (a) Profile from center to periphery of 8 cm target, 15 cm off-axis, in a direction parallel to the off-axis direction. (b) Profile from center to periphery of 8 cm target, 15 cm off-axis in a direction perpendicular to the off-axis direction. (c) Profile across 2 cm target, 15 cm off-axis in a direction parallel to the off-axis direction. (d) Profile across 2 cm target, 15 cm off-axis in a direction perpendicular to the off-axis direction.
Variation of overprediction of dose statistics by SP = 1 calculation compared with SP = 21 calculation against actual modulation factor for plans G–L.
Dose under- and overprediction of the SP = 1 calculation relative to the SP = 21 calculation for (a) plan D, (b) clinical case 1, (c) clinical case 2, and (3) clinical case 3. Each calculation is of the central axial slice of the PTV, with a 3 mm calculation point spacing except (a) which uses a 2 mm spacing. The PTV is displayed as a black outline in each case.
Plan D means normalized leaf opening times of a series of MLC positions for projections of gantry angle closest to 90°. As the variation in beam profile is small across these central positions, this value can be considered a surrogate for the fluence across the profile for these beam positions, exemplifying the concavity of the fluence distribution.
The same portion of the treatment sinogram for (a) plan I and (b) plan M. Darker shades indicate longer leaf opening times.
Central slice dose distributions of (a) plan I and (b) plan M. The spherical PTV outline can be seen at the centre of the dose distribution, surrounded by an avoidance structure to ensure conformality of dose. The cylindrical OAR can be seen to the posterior of the target.
Vertical dose profiles through the center of PTV in plan D. (Black triangles) Original TTPS plan without supersampling. (Hollow squares) TTPS recalculation with supersampling. (Bold Black line) CheckTomo SP = 1 calculation. (Bold gray line) CheckTomo SP = 3 calculation. (Dashed black line) SP = 21 calculation.
Key plan parameters for the 13 TomoTherapy phantom plans used in this study.
Optimization parameter sets used in the creation of plans within the TomoTherapy treatment planning system.
Plan details for clinical cases.
Overestimation of dose statistics for SP = 1 calculations relative to the corresponding SP = 21 calculation, tabulated against target size and position.
Article metrics loading...
Full text loading...