Thoracic disease detection using Deep learning

This project aimed to train a neural network - deep neural network to identify the 14 diseases from the NIH Chest X-Ray dataset. This dataset is very huge containing almost 100000+ frontal images of the X-ray .

The dataset had to be preprocessed as it was very biased with the number of X-Rays that indicated no findings. The Neural network used in this project was MobileNet.

This CNN would distinguish the different features of the images and classify them accordingly. The accuracy given by the MobileNet after training it more number of times over 5epochs is 80%