The goal of this project is to develop various convolutional neural network (CNN) to classify brain tumors from MRI images. By automating aspects of tumor diagnosis, this system aims to reduce diagnostic time, provide decision support to radiologists, and create a foundation for future improvements in medical image analysis.

Data Collection & Preprocessing
• Use publicly available MRI brain tumor datasets.
•Apply preprocessing steps (normalization, resizing, augmentation) to ensure data consistency.

Model Development
• Implement various CNN architectures with multiple convolutional and pooling layers, followed by fully connected layers for classification (tumor vs no tumor, Glioblastomas vs others, IDH mutation classifier)
•Train the models to distinguish between different tumor types.

Acknowledgment
This project idea was proposed by Stella Oh.