Computer-Aided Diagnosing of Tumor Diseases Using CT Images

In the paper, we describe an algorithm for diagnosing tumor diseases using CT images. It is based on a modified U-Net neural network. We developed a post-processing block that makes a decision about the presence of a cancerous tumor in a given area and a classification block that is responsible for detecting the fact of the presence of a tumor. The segmentation accuracy according to the Dice metric was 89% and was 83 % in more complex cases.