X-READ

Objective:

To aid Americans seeking health care amidst the Rural Health Crisis by automating the diagnosis of common Cardiothoracic illnesses, with an accuracy comparable to that of a physician.

Outcome:

CNN was able to classify the illness exhibited in the inputted x-ray image in seconds with an accuracy of 65-75% on average . The most accurately diagnosed illness was that of X-rays that exhibited a visible mass, and exhibited an accuracy of 91%.

Tools Utilized:

CNN programmed and trained in Python3 in an Anaconda environment,using the TensorFlow and Keras libraries. Training and testing data sourced from an NIH X-RAY imagery database. Front end programmed in Java using JGrasp.