Background: The artificial neural networks (ANNs) are useful in solving nonlinear processes, without the need for mathematical models of the parameters. Since the relationship between the CT numbers and material compositions is not linear, ANN can be used for obtaining tissue density and composition.Objective: The aim of this study is to utilize ANN for determination of the composition and mass density of different tissues to be used in Monte Carlo simulation in treatment planning of brachytherapy.Methods: The ANN were used for mass density calibration. The density and composition of several human body tissues, along with their corresponding CT numbers are used as the training samples. Finally, when the ANN is trained, the neural network would give us the material information, i.e. mass density, electron density, and material composition, by entering the CT numbers of different tissues into the network as its input. The tissue compositions and densities predicted by the ANN for each CT number were compared with the real values of such parameters. The tissue parameters predicted by the ANN were used as the phantom materials for obtaining the dose at different distances from Pd-103 and Cs-137 brachytherapy sources. Finally, the doses at different distances of the real phantoms were compared with doses inside the phantoms predicted by Neural Network.Results: According to the results of these studies, the Neural Network algorithm used in this investigation can be used for accurate prediction of the material compositions of different tissues. For example, it can give the mass densities of bone, muscle, and water with the percentage differences of 0.52%, -0.95%, and 0% respectively. Comparison of the dose distribution inside the water phantom predicted by ANN and the real water phantom shows a percentage difference of less than 0.66% and 2% for Cs-137 and Pd-103, respectively.Conclusion: The results of this study indicate that the Artificial Neural Networks are applicable in determination of tissue density and material compositions from the CT images data, and the material compositions and density of the phantoms (bone, muscle, and water) obtained by this method can be used for material definition in Monte Carlo simulations.