Identifies and describes lung nodule. TumorNet is based on an end-to-end trainable multi-view deep convolutional neural network (CNN) for the nodule characterization. This network can extract features from the input image and exploits a Gaussian Process (GP) regression to generate a malignancy score. This software performs median intensity projection to combine three different views from an image and uses a concatenation in a tensor form of a single image with three channels.
Center for Research in Computer Vision (CRCV) at University of Central Florida, Orlando, FL, USA; H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL, USA; CuraCloud Corporation, Seattle, WA, USA