Advancing Multi-Class Brain Tumor Classification With DenseNet201 CNN Based Model Using MRI Images

This study focuses on the crucial area of early identification and multi-class categorization of brain tumors using deep learning techniques, notably utilizing the DenseNet201 model. Brain tumors, regardless of whether they are benign or malignant, can cause elevated pressure inside the skull, which is inflexible. This can lead to serious outcomes, including life-threatening situations. The study employs a dataset consisting of 7023 MRI scans of the human brain, which are classified into four categories: glioma, meningioma, no tumor, and pituitary. Preprocessing includes picture augmentation using the ImageDataGenerator for both the training and validation sets. The DenseNet201 model has exceptional accuracy, reaching a 99% level, with precision, recall, and f1-score metrics providing evidence of the effectiveness of the classification. The importance is in the prompt identification of brain tumors, as fast detection helps in choosing suitable treatment approaches to reduce possible brain harm and enhance patient results. This research makes a valuable contribution to the growing field of medical imaging and deep learning by offering effective strategies to improve the diagnosis of brain tumors.