Part of the Deep Learning and Image Recognition class at the University of Chicago's Master of Science in Analytics program
For this project we have developed our own Image Recognition model and compared it with VGG16 (transfer learning) for predicting the severity of Alzheimer's using MRI Images. The model is hosted on a streamlit app to view with different MRI images.
- CNN
- Transfer Learning
- Image Augmentation
- StreamLit
- Python
- Google Colab
Dataset with more information can be found here
The dataset consists of 6400 MRI images (128x128), broken into 4 categories:
- Mild Demented
- Very Mild Demented
- Moderate Demented
- Non Demented
To classify MRI images we compared two different models:
- Convolutional Neural network with 2 layers
- VGG 16 Model
For each model we compared to datasets, one that was augmented and the original
Name | GitHub Handle |
---|---|
Priyank Shroff | @shroffp05 |
Nuka Gvilia | @Nuka-Gvilia |
Anisha BharathSingh | @AnishaB95 |