Diagnosing brain tumors is a time-consuming process heavily reliant on the expertise of radiologists, whose workload has surged alongside increasing patient numbers. Traditional methods have become costly and inefficient, prompting researchers to explore faster and more accurate algorithms for tumor detection and classification. Deep Learning (DL) techniques, particularly Convolutional Neural Networks (CNNs), have gained popularity for automating this task, enabling quicker and more precise diagnoses. The proposed Brain Tumor Classification Model based on CNN (BCMCNN) employs an adaptive dynamic sine-cosine fitness grey wolf optimizer (ADSCFGWO) algorithm to optimize CNN hyperparameters.