Feature selection is among the key components of machine learning optimizing models, as it enables reduction of dimensionality, enhances computing efficiency and builds trust in predictions. In this study, we developed an intelligent structure incorporating genetic algorithms (GAs) and deep learning aimed at multi-feature selections and classification. The suggested method demonstrates the ability of GAs in evolution, retaining the most instructional item while discovering the answer to the optimal sub-attribute from the high dimensional datasets. In this work, we proposed a novel deep learning-based assist system for glaucoma detection that is powered by genetic algorithms. The proposed system has two stages. In the first stage after performing the above-mentioned preprocessing of the disease data for the glaucoma, using the normalization and (mean absolute deviation technique) glued to the above-mentioned deep learning system the application of the artificial algae optimization as below shown sub point. Experimental results show that the structure achieves better accuracy, lower computational costs, and greater generalization capability compared to conventional feature selection and classification methods. Such methods are particularly well-suited for applications in areas where it is necessary to work with large and complex datasets, such as in financial modeling, image processing, and bioinformatics domains.