Deep learning model for glioma, meningioma and pituitary classification

<p>One of the common causes death is a brain tumor. Because above mentioned, early detection tumor critical for faster treatment, and therefore there are many techniques used to visualize One these magnetic resonance imaging (MRI). On other hand, machine learning, deep convolutional neural network (CNN) state art technologies in recent years solving medical image-related problems such as classification. In this research, three types tumors were classified using namely glioma, meningioma, pituitary gland on based CNN. The dataset work includes 233 patients total 3,064 contrast-enhanced T1 images. paper, comparison presented between model models demonstrate superiority our over others. Moreover, difference outcome pre- post-data preprocessing augmentation was discussed. highest accuracy metrics extracted from confusion matrices are; precision 99.1% pituitary, sensitivity 98.7% specificity 99.1%, pituitary. overall obtained 96.1%.</p>