A brain tumor is an intracranial mass consisting of irregular growth of brain tissue cells. Medical imaging plays a vital role in discovering and examining the precise performance of organs The performance of object detection has increased dramatically by taking advantage of recent advances in deep learning. This paper presents a Convolu-tional Neural Network (CNN) architecture model-based classification approach for brain tumor detection from Magnetic Resonance Imaging (MRI) images. The network training was carried out in both the original dataset and the augmented dataset. Whereas the whole brain MRI images were scaled to fit the input image size of each pre-trained CNN network. Moreover, a comparative study between the proposed model and other pre-trained models was made in terms of accuracy, precision, specificity, sensitivity, and F1-score. Finally, experimental results reveal that without data augmentation, the pro-posed approach achieves an overall accuracy rate of 96.35 percent for a split ratio of 80:20. While the addition of data augmentation boosted the accuracy to 97.78 percent for the same split ratio. Thus, the obtained results demonstrate the effectiveness of the proposed approach to assist professionals in Automated medical diagnostic services