Comparison of inhomogeneity correction algorithms in small photon fields.

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Dosimetry

Background:

• Traditional inhomogeneity correction algorithms (convolution superposition, Batho, equivalent pathlength) were developed for large fields (>5x5 cm²) under electronic equilibrium conditions.
• Modern intensity-modulated radiation therapy (IMRT) uses small beamlets that often fail to achieve electronic equilibrium, especially in inhomogeneous media.
• Monte Carlo (MC) simulations are increasingly vital for understanding dosimetry in small photon fields interacting with low-density materials.

Purpose of the Study:

• To compare the accuracy of convolution superposition, Batho, and equivalent pathlength algorithms against Monte Carlo simulations for inhomogeneity corrections in small radiation fields.
• To evaluate the impact of field size and medium density on dose calculation accuracy.
• To determine the suitability of these algorithms for accurate dosimetry in contemporary radiation therapy.

Main Methods:

• Depth dose data from MC simulations were compared with results from convolution superposition, Batho, and equivalent pathlength algorithms.
• Dose perturbation factor (DPF) and dose correction factor (DCF) were calculated to quantify discrepancies.
• Simulations were performed for a 6 MV photon beam across various field sizes and densities.

Main Results:

• For small fields in low-density media, MC and convolution superposition showed decreasing central axis dose, while Batho and equivalent pathlength showed increasing dose.
• Convolution superposition demonstrated better agreement with MC data (DCF close to 1.0) compared to Batho and equivalent pathlength algorithms, especially at interfaces.
• Algorithm agreement improved as field size increased, with all methods converging at fields >3 cm diameter where electronic equilibrium is restored.

Conclusions:

• Convolution superposition is more accurate than Batho or equivalent pathlength algorithms for inhomogeneity corrections in small photon fields.
• Simple algorithms like Batho and equivalent pathlength should be avoided for precise dosimetry of small fields near inhomogeneities.
• Accurate dosimetry in small fields within or near inhomogeneities necessitates advanced methods like convolution superposition or Monte Carlo simulations.