Developed a Convolutional Neural Network (CNN) model for the detection and multi-class classification of brain tumors using a combined dataset comprising 7,023 MRI images categorized into glioma, meningioma, pituitary, and no tumor. The project utilized pre-processing techniques, including resizing and margin removal, to enhance model accuracy. Focused on improving early diagnosis and segmentation of tumor locations to aid in selecting appropriate treatment methods.
Dataset link: https://www.kaggle.com/datasets/masoudnickparvar/brain-tumor-mri-dataset