Development of Convolutional Neural Network for Defining a Renal Pathology Using Computed Tomography Images

It is known that impaired kidney function leads to a deterioration of human health and quality of life. Modern medicine offers a method of computed tomography (CT) for the diagnosis of various types of renal pathology. At the same time, the risk of incorrect diagnosis by a doctor is still high, even with wide practical experience. The quality of CT scan analysis can be significantly improved by Convolutional Neural Networks (CNNs). This paper represents a construction of a CNN for the task of multiple classification of renal pathology, as well as a modification of the algorithm using a variety of activation functions.