This project implements an image classification model for brain tumor detection using deep learning. The model is trained on augmented MRI scan images and classifies them into four categories:
- Glioma
- Meningioma
- Pituitary
- Healthy
The project includes:
- Data preprocessing and augmentation
- Hyperparameter optimization with Keras Tuner
- Model training with early stopping
- Final evaluation and sample visualizations
- Activation Function:
relu - Number of Convolutional Layers:
3 - Filters per Layer:
- Layer 1:
128 filters - Layer 2:
64 filters - Layer 3:
32 filters
- Layer 1:
- Dense Layer Units:
96 - Dropout Rate:
0.2 - Learning Rate:
0.0001
📎 Project Files (Notebook + Model + Resources)
📂 Access via Google Drive