Abstract
Lung cancer is a serious health problem. In the United States alone, approximately 225,000 people each year are diagnosed with lung cancer. Early detection is a crucial part of giving patients the best chance of recovery. Deep learning gives us an opportunity to increase the accuracy of the automated initial diagnosis. Here we present an ensemble of Convolution Neural Networks(CNN) using multiple preprocessing methods to increase the accuracy of the automated labeling of the scans. We have done this by implementing ensembles of CNNs along with a voting system to get the consensus of the two networks. The initial results of our best method show both a consistently high accuracy (97.5%) and a low percentage of false positives (