Generation of material-specific energy deposition kernels for kilovoltage x-ray dose calculations

A novel analytical method for computing dose from kilovoltage beams used in Image-Guided radiation therapy

PURPOSE

A modified convolution/superposition algorithm is proposed to compute dose from the kilovoltage beams used in IGRT. The algorithm uses material-specific energy deposition kernels instead of water-energy deposition kernels.

METHODS

Monte Carlo simulation was used to model the Elekta XVI unit and determine dose deposition characteristics of its kilovoltage beams. The dosimetric results were compared with ion chamber measurements. The dose from the kilovoltage beams was then computed using convolution/superposition along with material-specific energy deposition kernels and compared with Monte Carlo and measurements. The material-specific energy deposition kernels were previously generated using Monte Carlo.

RESULTS

The obtained gamma indices (using 2%/2mm criteria for 95% of points) were lower than 1 in almost all instances which indicates good agreement between simulated and measured depth doses and profiles. The comparisons of the algorithm with measurements in a homogeneous solid water slab phantom, and that with Monte Carlo in a head and neck CT dataset produced acceptable results. The calculated point doses were within 4.2% of measurements in the homogeneous phantom. Gamma analysis of the calculated vs. Monte Carlo simulations in the head and neck phantom resulted in 94% of points passing with a 2%/2mm criteria.

CONCLUSIONS

The proposed method offers sufficient accuracy in kilovoltage beams dose calculations and has the potential to supplement the conventional megavoltage convolution/superposition algorithms for dose calculations in low energy range.