The pituitary gland regulates crucial body functions by controlling other endocrine glands. Despite technological advances, the classification of pituitary tumors remains a formidable challenge due to their varied forms and types, posing a significant threat to patients. This study proposes a Pituitary Tumor Classification Model using Convolutional Neural Networks (CNNs). The model analyzes MRI images of the pituitary gland to classify tumors, leveraging data augmentation techniques to improve accuracy and reduce overfitting. Two experiments were conducted: the first without data augmentation, achieving an accuracy of 92.31%, and the second with data augmentation, resulting in a significantly improved accuracy of 94.87%. These findings demonstrate the effectiveness of incorporating data augmentation in enhancing model performance. The proposed model provides a more objective, efficient, and accurate diagnostic tool, potentially reducing the burden of subjective diagnosis and aiding healthcare professionals in making timely decisions.