Aki Health

We propose a mobile application called Aki. It would help healthcare workers and patients in Sub-Saharan Africa and South Asia to diagnose illnesses which have symptoms that manifest themselves physically on a patient’s skin. Such ailments include melanoma, shingles, measles, cellulitis and impetigo. Our application would use some of the most recent advances in computer vision, cloud computing, and machine/deep learning. The more pictures users submit, the more accurate the application becomes. It will enable users to take a picture of a lesion or rash, upload it to our cloud server. The server will then perform an analysis and output a diagnosis with a confidence rating and links to medication and resources. All of this can be done with an average smartphone, which has a camera and an internet connection.