Background: Inclusion of air-filled cavities in the head and neck treatment fields due to electronic disequilibrium may lead to uncertainties in predicting dose distribution by treatment planning systems (TPS). In this study the Full Scatter Convolution Algorithm (FSC) used in TiGRT treatment planning systems was evaluated using a novel 3D film dosimetry method. Materials and Methods: 9 pieces of EBT2 films were embedded in a rectangular inhomogeneous head and neck phantom. Three approximately small field sizes including sandwiched EBT films which embedded between phantom slabs were exposed with 6 MV X-ray photons. A homemade computer code was developed in MATLAB for the creation of a 3D dose map of irradiated films and calculation of enclosed volumes which were surrounded by isodose lines in films. Then the calculated dose volumes were compared with the same quantities derived from Dose Volume Histogram (DVH) which is available in TPS outputs. Results: Our results showed significant differences between the results of the film dosimetry and DVH values. The maximum difference of calculated and measured values was observed in volumes surrounded by 95% isodose curves and 3x3cm² field size (P=0.035) and the minimum difference was observed in volumes surrounded by 95% isodose curves and 5x5cm² field size (P=0.047). Conclusions: The results of this study demonstrated overestimated results in volumetric dose calculation of the FSC dose calculation algorithm in the presence of inhomogeneities and the 3x3cm² field size. Also, these results confirmed that the utilization of an accurate and comprehensive method such as 3D film dosimetry can be useful for TPS verification.