Neural network model and classifier training algorithm for processing human serum gel electrophoresis data

The analysis of biomedical images of proteinograms obtained as a result of gel electrophoresis is a research area of current interest. As a result of the study of various methods and means of analyzing electrophoregrams, the authors proposed a resource-efficient and fast model of convolutional neural network, which allows the classification of human blood serum proteinograms with high accuracy at low requirements to computing resources of the computer.