This work presents a metaheuristic hybrid optimization technique developed to synthesize frequency selective surface (FSS) structures, composed of triangular patch elements and printed on FR-4 dielectric substrates for microwave filtering applications. The optimization technique is based on the combination of genetic algorithm (GA) and Multilayer Perceptron (MLP) artificial neural network (ANN). The developed MLP model for FSS synthesis was performed for efficient evaluation of cost function in GA iterations. The advantages in the optimal design of FSS through the proposed GA-ANN technique are discussed in terms of convergence and computational cost. An optimized FSS prototypes were fabricated and measured. The accuracy of the proposed optimization technique was verified through the good agreement obtained by means of comparison between theoretical and experimental results.