Independent evaluation of an in-house brachytherapy treatment planning system using simulation, measurement and calculation methods

Abstract

Accuracy of treatment planning systems may significantly influence the efficacy of brachytherapy. The purpose of this work is a detailed, varied and independent evaluation of an in-house brachytherapy treatment planning software called STPS. Operational accuracy of STPS was investigated. Geometric tests were performed to validate entry and reconstruction of positional information from scanned orthogonal films. MCNP4C Monte Carlo code and TLDs were used for simulation and experimental measurement, respectively. STPS data were also compared with those from a commercial planning system (Nucletron PLATO). Discrepancy values between MCNP and STPS data and also those of PLATO and STPS at Manchester system dose prescription points (AL and AR) of tandem and ovoid configurations were 2.5% ± 0.5% and 5.4% ± 0.4%, respectively. Similar results were achieved for other investigated configurations. Observed discrepancies between MCNP and STPS at the dose prescription point and at 1 cm from the tip of the vaginal applicator were 4.5% and 25.6% respectively, while the discrepancy between the STPS and PLATO data at those points was 2.3%. The software showed submillimeter accuracy in its geometrical reconstructions. In terms of calculation accuracy, similar to PLATO, as attenuation of the sources and applicator body is not considered, dose was overestimated at the tip of the applicator, but based on the available criteria, dose accuracy at most points were acceptable. Our results confirm STPS's geometrical and operational reliability, and show that its dose computation accuracy is comparable to an established commercial TPS using the same algorithm.

Figure 1. The source point pattern used for determination of the geometrical accuracy of the STPS.

Figure 2. The source train used in treatment of three patients.

Figure 3. The XY plane of the simulated geometry used in treatment of a patient using a tandem‐ovoid applicator set (T1).

Figure 4. The central slab of the two phantoms to measure in the: a) XY plane, and b) YZ plane.

Figure 5. The phantom configurations at the time of irradiations to measure in the: a) XY plane, and b) YZ plane.

Figure 6. Summary of the intercomparisons in this investigation.

Figure 7. Interpolated isodose contour maps around a cylinder loaded with 10 active sources obtained by MC, STPS, PLATO, and TLD: a) XY plane, and b) YZ plane. (Dose rates values are in units of cGy/hr.)

Figure 8. Comparison of the dose rates at different points on the 100% isodose curve within the XY plane for patient T1.

Figure 9. Interpolated isodose contour maps around pellet configurations of patient T2 in YZ plane obtained by STPS, PLATO, and MC calculations. Superior (cranial) and posterior directions are toward the right and top of the diagram, respectively. (Dose rates values are in units of cGy/hr).

Figure 10. Comparison of the dose rates at different points on the 100% isodose curve in the XY plane for patient T2.

Figure 11. Interpolated isodose contour maps around pellet configurations of patient O1 in the XY plane obtained from STPS, PLATO, and MC calculations. (Dose rates values are in units of cGy/hr.)

Figure 12. Comparison of the dose rates at different points on the 100% isodose curve in the XY plane for patient O1. (Note: each point's coordinate is 1:(−1,1.65,0), 2:(1,1.55,0), 3:(−2.7,0,0), 4:(2.7,0,0), 5:(−1,−1.65,0), 6:(−1,−1.7,0), 7:(0,1.6,0) and 8:(−0,−1.7,0).)