A novel deep neural network method for electrical impedance tomography

Abstract

Image reconstruction for Electrical Impedance Tomography (EIT) is a highly nonlinear and ill-posed inverse problem. It requires the design and employment of feasible reconstruction methods capable to guarantee trustworthy image generation. Deep Neural Networks (DNN) have a powerful ability to express complex nonlinear functions. This research paper introduces a novel framework based on DNN aiming to achieve EIT image reconstruction. The proposed DNN model, comprises of the following two layers, namely: The Stacked Autoencoder (SAE) and the Logistic Regression (LR). It is trained using the large lab samples which are obtained by the COMSOL simulation software (a cross platform finite elements analysis solver). The relationship between the voltage measurement and the internal conductivity distribution is determined. The untrained voltage measurement samples are used as input to the trained DNN, and the output is an estimate for image reconstruction of the internal conductivity distribution. The results show that the proposed model can achieve reliable shape and size reconstruction. When white Gaussian noise with a signal-to-noise ratio of 30, 40 and 50 were added to test set, the proposed DNN structure still has good imaging results, which proved the anti-noise capability of the network. Furthermore, the network that was trained using simulation data sets, would be applied for the EIT image reconstruction based on the experimental data that were produced after preprocessing.

Electrical Tomography Reconstruction Using Reconfigurable Waveforms in a FPGA

“…In the same way, the algorithm selected is classical and needs to be considered in any benchmark. Advanced modern algorithms using machine learning can be considered as a comparison, for example, Bayesian learning [ 42 , 43 ] or neural networks [ 8 , 44 ]. A comparative analysis of reconstruction strategy, excitation configuration and waveform modulation is part of further work in preparation.…”

Liver Electrical Impedance Tomography for Early Identification of Fatty Infiltrate in Obesity

“…For instance, particle swarm optimization (PSO) was applied via paradigm shift from the conventional Gauss-Newton methods for fast convergence and high spatial resolution to solve EIT 43,44 . Recent studies have applied machine learning, including Convolutional Neural Networks, to solve the non-linear ill-posed inverse problem for accurate EIT reconstruction 45,46 . Hamilton et al obtained absolute EIT images by combining the D-bar method with subsequent processing using convolutional neural networks (CNN) technique for sharpening EIT reconstruction 45 .…”

“…Hamilton et al obtained absolute EIT images by combining the D-bar method with subsequent processing using convolutional neural networks (CNN) technique for sharpening EIT reconstruction 45 . Li et al utilized deep neural networks (DNN) to directly obtain a nonlinear relationship between the one-dimensional boundary voltage and the internal conductivity 46 . Experimentally, the accuracy of EIT reconstruction may be further enhanced by increasing the electrode arrays at multiple levels around the abdomen.…”

A two-stage image reconstruction strategy for electrical impedance tomography

“…During the measurement, two adjacent electrodes are successively excited by a safe alternating current. Then, a sensitive field is established and boundary voltage can be measured from other adjacent electrode pairs based on which tomography can be realized (Li et al, 2019). The EIT technique has been proved to be potential and competitive in visualizing industrial process and monitoring biomedical disease (Ma et al, 2020; Pellegrini et al, 2020).…”