This research focuses on classifying brain tumors using Convolutional Neural Networks (CNNs) for Magnetic Resonance Imaging (MRI) images. The study aims to develop tools that can assist healthcare professionals in quickly and accurately diagnosing brain tumors, reducing reliance on clinical experience alone. The dataset used consists of 7023 MRI images, categorized into 1621 Glioma, 1645 Meningioma, 1757 Pituitary, and 2000 non-tumor images. Two architectural schemes of CNN models were evaluated, with F1-Score values of 96% and 98% for Scheme 1 and Scheme 2, respectively, and Accuracy values of 98% and 99%. These results surpass those of previous studies, which reported accuracy values ranging from 94.39% to 97.54%. The research highlights the effectiveness of CNNs in brain tumor classification, making it a valuable tool for medical professionals.This research focuses on classifying brain tumors using Convolutional Neural Networks (CNNs) for Magnetic Resonance Imaging (MRI) images. The study aims to develop tools that can assist healthcare professionals in quickly and accurately diagnosing brain tumors, reducing reliance on clinical experience alone. The dataset used consists of 7023 MRI images, categorized into 1621 Glioma, 1645 Meningioma, 1757 Pituitary, and 2000 non-tumor images. Two architectural schemes of CNN models were evaluated, with F1-Score values of 96% and 98% for Scheme 1 and Scheme 2, respectively, and Accuracy values of 98% and 99%. These results surpass those of previous studies, which reported accuracy values ranging from 94.39% to 97.54%. The research highlights the effectiveness of CNNs in brain tumor classification, making it a valuable tool for medical professionals.