Modern agriculture relies on crop tissue culture technology for fast propagation, genetic improvement, and protection of plant species. Conventional media like Murashige and Skoog (MS) usually lack optimal conditions for embryogenesis, necessitating the development of improved media tailored to specific crop requirements. In this article, we introduce an Efficient Grid Identified-Deep Feedforward Neurons (EGI-DFFN) to identify the ideal nutrient and vitamin levels of the crop plants for improving crop tissue culture aimed to improve crop plant growth in a lab setting by predicting callogenesis rate (CGR), embryogenesis rate (EGR), and somatic embryo number (SEN), shoot regeneration rate (SRR), rooting rate (RR).Different concentrations of ionic macronutrients, bio-molecular, and vitamins of the crop plant are the input to the predictive model, which is collected through the laboratory Callus Induction Experiment (CIE). Z-score normalization is used to preprocessing the CIE data to ensure consistent scales across different input features and improve model training performance. DFFN used discriminates to predict complex relationships and interactions between CGR, EGR, SEN, SRR, and RR with EGI tuning. The EGI-DFFN model has significantly improved crop tissue culture growth by accurately predicting the CGR, EGR, SEN, SRR, and RR respectively. The EGI-DFFN model enhances understanding of how ionic macronutrients and vitamins impact plant growth. It identifies optimal concentrations of the biomolecular to enhance somatic embryo formation and plantlet development, providing insights for optimizing crop tissue culture conditions for optimal growth outcomes.