Rekonstrukce obrazu z naměřených dat v elektrické impedanční tomografii

Abstract

This thesis consists of theoretical and practical parts. In theoretical part are analyzes problems of bioimpedance, methods of electrical impedance tomography, deterministic and stochastic methods of image reconstruction and problems of neural networks. The practical part is dedicated to design and implementation of image reconstruction in EIT using neural networks. For convenience working with the program was designed user interface. Into this interface belongs generating training and testing sets of models and image reconstruction through the use of neural network. In case of doubt, there is help, where is subscribed how concrete button works.