Background: In treating patients with radiation, the degree of accuracy for the delivery of tumor dose is recommended to be within ± 5% by ICRU in report 24. The experimental studies have shown that the presence of low-density inhomogeneity in areas such as the lung can lead to a greater than 30% change in the water dose data. Therefore, inhomogeneity corrections should be used in treatment planning especially for lung cancer. The usual methods for inhomogeneity correction are the Tissue-Air Ratio (TAR) method, the power low tissue-air ratio (Batho) method, and the Equivalent Tissue-Air Ratio (ETAR) method. But they are not able to calculate the dose with required accuracy in all cases. New and more accurate methods are based on Monte Carlo methods. They are able to account for all aspects of photon and electron transport within a heterogeneous medium. The focus of this paper is the application of MCNP (Monte Carlo
N-Particle) code in radiotherapy treatment planning.

Materials and methods: Some special test phantoms were made of cork and Perspex instead of lung and normal tissue respectively (with electron densities relative to water equal to 0.2 and 1.137 respectively). Measurements were obtained using cobalt-60 radiation for four different fields. Then the results of RTAR, Batho and MCNP methods were compared to the measurements.

Results: RTAR method has an error equal to 10% approximately. Also Batho method has an error especially in the low-density material. At least, MCNP method calculates correction factors very accurately. Its average error is less than 1% but it takes a long time to calculate the dose.

Conclusion: Monte Carlo method is more accurate than other methods and it is currently used in the process of being implemented by various treatment planning vendors and will be available for clinical use in very near future. Iran . J. Radiat. Res., 2003 1(3): 143 - 149