I've always been very keen into life-sciences and healthcare, so when I decided to become a Data Scientist and discovered how powerful and useful Machine and Deep Learning could be, I started to think on creating something to help others.

This project is the result of pursuing that goal: I've created a high accuracy model using Machine Learning and neural networks to diagnose pneumonia reading chest X-ray images.

Neural networks (CNN) | tensorflow | keras | 93% accuracy | 0.98 recall

• images: folder contains representative images from this project.
• models: folder contains the json file with the final model.history details and the url to get the trained model.
• presentation: folder contains the file used for presenting this project.
• X-ray code.ipynb: contains all the code.

Data downloaded from kaggle.

• 5,863 images.
• 2 categories: PNEUMONIA, NORMAL.

The final model was trained in kaggle notebooks using GPU.