Adi's work from the TEWH Malaria Project
I worked on a machine learning project to diagnose malaria in developing countries. Our final product was an Andriod app that could take a picture of a blood smear, identify the cells in the image, and report the likelihood that the cell was infected.
I specifically designed Fast Super Resolution Convolutional Neural Networks (FSRCNNs). This type of network increases the resolution of images. It was developed in case the person using our app did not have a 128x128 pixel resolution camera. The FSRCNN would increase the resolution so that the image could be fed into our model.