Media optimization to maximize mushroom mycelium production is an essential process. In this study, a mycelium growth model based on medium concentration was developed using two widely employed techniques: Response Surface Methodology (RSM) and Artificial Neural Networks (ANN). The accuracy of the resulting models was statistically verified, and the concentrations of glucose, malt extract, and peptone were utilized as independent variables to develop a predictive model for the optical density (OD) of the mycelium. Both RSM and ANN models produced statistically robust fits and were subsequently used to determine the optimal concentrations of the nutrients. The optimal glucose concentration was identified as 26.02 g/L according to the RSM model and 34.93 g/L according to the ANN model. Under these optimized conditions, the mycelium OD reached 23.31 for RSM and 26.14 for ANN combined with a genetic algorithm (GA-ANN). These results demonstrate that while both methods are effective, the GA-ANN approach provides superior predictive performance and a higher optimization yield for mushroom mycelium production compared to the traditional RSM.