A deep learning project for classifying brain MRI images into four tumor categories by comparing multiple transfer learning models.
Tech Used: Python, TensorFlow, Keras, NumPy, Pandas, Matplotlib, Scikit-learn
- Transfer Learning - Trained and compared ResNet50, MobileNetV2, DenseNet121, InceptionV3, and VGG16
- MRI Classification - Classified brain MRI scans into four categories using a multiclass CNN model
- Data Augmentation - Applied image preprocessing and augmentation to improve model generalization
- Performance Evaluation - Compared models using accuracy, precision, recall, F1-score, and confusion matrices
- Training Visualization - Generated training and validation accuracy graphs for each architecture
- Model Comparison - Evaluated multiple CNN architectures to determine the best balance between accuracy and training speed
