Electrical Resistivity Tomography Method Research Articles

Reconstruction of conductivity distribution variation with Tikhonov and wavelet frame combined method for electrical impedance tomography

Electrical impedance tomography (EIT) is receiving considerable research interest in a variety of applications. With this technique, it is possible to reconstruct conductivity distribution inside the detected region. Note that reconstruction of conductivity distribution is an ill-posed inverse problem. To deal with this problem, Tikhonov regularization method has been preferred. However, this method suffers from poor reconstruction quality. In this work, a novel approach which combines Tikhonov regularization method with wavelet frame is proposed for image reconstruction of EIT. The proposed method is superior to the popular Tikhonov regularization method in enforcing sparse of the solution and enhancing sharp feature. Simulation work has been conducted to validate the performance of the proposed method. Compared with Landweber method, total variation (TV) regularization method, and Tikhonov regularization method, reconstructions of six different models show that images reconstructed by the proposed method have better quality and are more robust to noise. In addition, image reconstruction is also performed based on phantom experimental data. The results further demonstrate that the proposed method outperforms other three methods since the inclusion has been more accurately reconstructed, and there are almost no artifacts in the reconstructed images.

Approbation of seismic and electrical geophysical techniques complex for solving the permafrost monitoring problems on the pile-foundation building

In the urban conditions in Salekhard city, a complex of methods, including electrical resistivity tomography (ERT), ground penetrating radar (GPR), a seismic survey on refracted waves (REF), on surface waves (MASW) and the standing waves selection, was tested on solving geocryological problems and monitor the permafrost at the building base. As a model object, a residential building on a pile foundation, built according to the principle I and equipped with a continuous geothermal monitoring system, was chosen. It has been established that the ERT, REF, and MASW methods are ineffective in the conditions of a reinforced concrete pile foundation in a grillage and a concrete floor screed with a layer of loose bulk soil under it. The best result was shown by GPR at frequencies of 150 and 400 MHz and seismic exploration with the standing wave selection. A combination of these methods is recommended for determining the upper permafrost boundary under buildings, determining the actual piles-length, the quality of their fixing, assessing the water content of soils in the area around the piles, and establishing the residual building life.

Evaluation of lung function in a German single center cohort of young patients with sickle cell disease using EIT and standard techniques.

Sickle cell disease (SCD) is a very common autosomal recessive hemoglobinopathy leading to multiple pulmonary complications that are closely associated with mortality. The pathophysiology of chronic pulmonary involvement is not yet fully understood and no specific therapies are available. The aim of this cross-sectional study was to characterize the lung function of children and young adolescents with SCD in a German single-center cohort and to extend conventional lung function testing by the use of a new imaging method. We performed spirometry and body plethysmography in 35 children and young adults with hemoglobin SS, SC, S/β-thalassemia as well as 50 controls. These data were compared with clinical characteristics and typical laboratory parameters of hemolysis and disease activity in SCD. To identify lung inhomogeneities, for example due to atelectasis, hyperinflation, air trapping or vascular occlusions, we used the promising new method of electrical impedance tomography (EIT) and calculated global inhomogeneity indices. Lung function of patients with SCD was significantly reduced compared to that of healthy controls. When the result was found to be pathological, the most commonly observed type of breathing disorder was classified as restrictive. Laboratory parameters showed typical features of SCD including decreased levels of hemoglobin and hematocrit and elevated levels of leucocytes, platelets, lactate dehydrogenase and total bilirubin. However, there was no correlation between blood values and reduced lung function. Electrical impedance tomography (EIT) revealed no abnormalities in SCD patients compared to healthy controls. In particular, we were unable to demonstrate any regional inhomogeneities in lung ventilation. In our study, SCD patients showed impaired lung function, with a relevant percentage of patients suffering from restrictive breathing disorder. Signs of obstruction could not be detected. Electrical impedance tomography (EIT) measurements revealed no unevenness that would suggest air entrapment, blockage of blood vessels, excessive inflation, obstruction, or other forms of lung disease. Additionally, the reduction in lung function observed in SCD patients was not related to the disease severity or laboratory test results.

Integrated GIS, Remote Sensing, and Electrical Resistivity Tomography Methods for the Delineation of Groundwater Potential Zones in Sangaw Sub-Basin, Sulaymaniyah, KRG-Iraq

In the Sangaw region, groundwater is the primary supply of water for drinking, residential purposes, livestock, and summer farming activities. Therefore, the main objective of this research is to delineate groundwater potential zones (GWPZs) in the Sangaw sub-basin, Sulaymaniyah, KRG-Iraq, by integrating geographic information system (GIS), remote sensing (RS), analytical hierarchy process (AHP), and electrical resistivity tomography (ERT) techniques. Seven different thematic layers, including geology, rainfall, slope, lineament density, land use/land cover, drainage density, and topographic position index, were chosen as the prediction factors. The analytical method of an analytical hierarchy process pair-wise matrix was used to evaluate the normalized weight of these thematic layers. All the layers and their corresponding classes were assigned ranks and weights based on their impact on groundwater potential. Using ArcGIS, these thematic maps were combined to precisely determine the groundwater potential map within the research area. Five different potential zones were generated for the resulting map, namely, very low (55.4 km2), low (90.4 km2), moderate (68.1 km2), high (100 km2), and very high (62.4 km2). The findings revealed that almost 43.2% of the study region is characterized by high to very high groundwater potential zones. In contrast, the very low to low groundwater potential covers around 38.7%, and the moderate groundwater potential occupies 18.1% of the study region. The final map was then validated using results from the two-dimensional inverse sections of eight electrical resistivity tomography profiles. The validation data confirmed that groundwater potential classes strongly overlap with the subsurface water-bearing or non-bearing lithology, and groundwater productivity zones in the given area. The novelty of this research lies in the application of electrical resistivity tomography validation to the groundwater potential mapping approach, which illustrates the robustness of the overall methodology for data-scarce areas. Furthermore, this is one of the very few groundwater potential studies in Iraq and the first in the Sangaw sub-basin, which can assist decision-makers with groundwater prospecting and management, and enable further exploration in the region.

Research on Time-Domain Airborne EM Full-Field Apparent Resistivity Imaging Method for Arbitrary Transmitting Waveform

The time-domain airborne electromagnetic (AEM) system can do fast electromagnetic (EM) surveys over mountainous areas by carrying its detection equipment on an airplane. Due to the dense sampling of the AEM method, it generates a huge amount of data. As a result, resistivity imaging methods have become the first choice for data interpretation because they are fast. However, the traditional imaging methods do not fully consider the influence of the transmitting waveform. When they are used to deal with the AEM data with complex current waveform, the imaging results are seriously affected. Therefore, we develop a universal full-field apparent resistivity imaging method for AEM data with arbitrary transmitting waveform. Firstly, we calculate the convolution of the time derivative of the current waveform and step response to obtain the time-domain AEM response of the arbitrary transmitting waveform. Then the full-field apparent resistivity imaging method based on the inverse function theorem is used to complete the rapid imaging of AEM data with complex waveform. Finally, we apply our imaging code to both synthetic and field data to verify its correctness.

Characterization of potential Cu-Au mineralization using electrical resistivity and induced polarization methods: a case study in Eastern Labo, Bicol, Philippines

Abstract Electrical resistivity and induced polarization (ER-IP) surveys were carried out in Eastern Labo, Bicol, Philippines, to assess the potential Cu-Au mineralization in pyroclastic-covered areas. An electrical resistivity tomography method with a dipole–dipole array was used, and the L1 norm (robust) inversion approach was employed to generate the 16 2D ER-IP models. The analysis of the resistivity and chargeability classified the underlying lithology into eight zones. Zones from C to F are the potential mineralized zones, while Zones A and B (Labo Volcanic Complex) and Zones G and H (Tumbaga Formation and Tamisan Diorite) are the non-mineralized zones. Zone C, represented by low chargeability (<15 mV V−1) and intermediate resistivity (15–100 Ω-m) values, is characterized by a high concentration of disseminated sulfide minerals (e.g. chalcopyrite). Zone D has a chargeability range comparable to Zone C but higher resistivity (>100 Ω-m), which might be due to silicification. Zone E corresponds to the oxide zone with concentrated sulfide minerals along clayey, weathered, fractured areas; it has moderate to high chargeability (>15 mV V−1) and low resistivity (<15 Ω-m) signatures. Zone F has a similar resistivity range to Zone E but has lower chargeability (<15 mV V−1) values. The interpreted underlying lithological units were confirmed using borehole data. Because of the extensive occurrence of high chargeability zones with moderate resistivity anomalies, potentially mineralized areas for further investigation (e.g. drilling) were identified in the north-eastern portion of the study area.

Application of VES and 2D Resistivity Methods for Groundwater Exploration in Kutigi-Enagi Region, Northern Bida Basin, Nigeria

The study integrates electrical resistivity tomography (ERT) and vertical electrical sounding (VES) methods for investigation of the groundwater potential of the study area. Five ERT was conducted to determine the variation in resistivity of the shallow subsurface rocks both vertically and laterally in order to infer aquifer zones therefrom while twenty-four VES was carried out to compliment the information obtained from the ERT, determine vertical variation in resistivity, infer lateral variation in resistivity, identify depth to aquifer and infer groundwater prospect of the area. Three geoelectric layers were obtained with layer resistivity and thickness increasing northwards with shallower depth to aquifer obtained at Kutigi. K (ρ1<ρ2>ρ3) and Q (ρ1>ρ2>ρ3) resistivity type curves constitute 75% of obtained curves signifying increasing groundwater saturation and salinity, and are indication of increasing aquifer prospect with depth. The groundwater prospects in the area is generally high with properly constructed boreholes in these areas will produce prolific groundwater at 40 meters with higher yield expected at the southern regions of the study area around Enagi.

Classification and Characterization of Damage in Composite Laminates Using Electrical Resistance Tomography and Supervised Machine Learning

Electrical resistance tomography (ERT) is a nondestructive evaluation technique that uses the internal conductivity variations of materials to assess structural integrity. Due to the low instrumentation required, the widespread use of ERT in the aerospace industry for monitoring the accumulation of damage in aircraft components can lead to significant reductions in inspections and maintenance costs. However, implementing the ERT method for mapping the damage state of structural components made of carbon fiber reinforced polymeric (CFRP) composites is challenging due to the inability of this method to distinguish between damage modes such as delamination and matrix cracking. This article explores the combined use of ERT and machine learning algorithms such as neural networks, random forests, k-nearest neighbors, and support vector machines to classify and characterize delamination and matrix cracking damage in CFRP laminates. Results show that the proposed classification algorithms can successfully estimate the damage severity of delaminated composites in the presence of matrix cracking. Similarly, the classification algorithms can characterize these independent damage modes with an accuracy of 95%. The algorithms showed robustness to predict the electrical resistance variations of damaged composites and characterize delamination and matrix cracking damage even when intrinsic noise was considered. Although neural networks characterized damage with the highest accuracy, these algorithms were also the most sensitive to noise. For applications where instrumentation noise cannot be completely removed from the ERT signals, the use of nearest neighbors is thus recommended.

Time‐Lapse Geophysical Investigation of Geyser Dynamics at Spouter Geyser, Yellowstone National Park: Geyser Dynamics II

AbstractGeysers are unique hydrothermal features, requiring specific geometry, fluid, and vapor input, and heat to produce eruptions. In a geyser eruption, the decompression of superheated hydrothermal fluids results in a conversion of thermal to kinetic energy. Recent studies conclude that a laterally offset cavity structure (bubble trap) plays an integral role in the geyser eruption process. In this second study of Spouter Geyser, Yellowstone National Park (YNP), we build on the geyser structure developed in the first publication and explore how structural components, such as the bubble trap, control eruption dynamics. We utilize time‐lapse electrical resistivity tomography (ERT) and transient electromagnetics (TEM) geophysical methods to track changes in the saturation of conductive hydrothermal fluid and resistive vapor through the eruption cycle. Additionally, we use pressure‐temperature transducer data to measure eruption and recharge durations over the past 23 years at Spouter Geyser, identifying long‐period trends in geyser behavior. The geophysics results support the bubble trap model, capturing an increase in resistivity in the bubble trap structure during the recharge phase interpreted as a vapor saturation increase. The TEM also captures a resistivity increase in the 20–30 m depth interval during the eruption phase, interpreted as a vapor‐dominated “flash eruption zone” where the superheated hydrothermal fluids flash into steam and drive fluid out of this zone during the eruption. From the temperature time‐series detailing 23 years of eruption and recharge durations, we find that eruption durations have remained constant at ∼2 hr while recharge durations have decreased linearly at 6.6 min/year.

Joint Analysis of InSAR, ERT, and in-Situ Geological Survey in Landslide Prone Area: A Case in Laeya Village, North Buton Regency, Indonesia

Landslide investigation in Laeya Village of North Buton Regency, Indonesia, is necessary since this area is frequently experiencing landslides. In this study, we applied a joint analysis of interferometric synthetic aperture radar (InSAR), electrical resistivity tomography (ERT), and in-situ geology observation. The objectives of this multidisciplinary study are: a) to estimate ground deformation through the InSAR method, b) estimation of subsurface characteristics based on the electrical resistivity of rocks, as well as estimation of the shape of the slip plane and the type of landslide that resulted; and c) to identify the surface geology conditions of the site from an in-situ geology survey. The InSAR data of the Sentinel-1A acquired from 2015 to 2017 were processed using the Sentinel toolbox. The resistivity data collected utilizing the electrodes array of the Wenner-Schlumberger configuration were processed employing the RES2DINV. The in-situ geology survey involving the observation of elevation, slope, and soil and rock type was also carried out in the landslide area. The InSAR method may estimate the deformation of slope material of the investigated sites with a magnitude ranging from -53 mm to +55 mm. The negative mark of the magnitude means that the ground deformation is ground subsidence and the positive means ground uplift. The ERT method has successfully estimated the existence of the saturated zones, the clay layer that acts as a slip plane, and the slip surface type. There are at least two types of slip planes estimated, namely the planar and uneven curvature slip planes that generate the translational and the complex landslides, respectively. In-situ observations of geological conditions support the results obtained from the InSAR and ERT methods.

Distribution and Degradation Processes of Isolated Permafrost near Buried Oil Pipelines by Means of Electrical Resistivity Tomography and Ground Temperature Monitoring: A Case Study of Da Xing’anling Mountains, Northeast China

Human engineering activities and climate warming induce permafrost degradation in the Da Xing’anling Mountains, which may affect the distribution of permafrost and the safety of infrastructure. This study uses the electrical resistivity tomography method, in combination with field surveys and ground temperature monitoring, to investigate the distribution and degradation characteristics of permafrost and influencing factors at a typical monitoring site (MDS304) near the China-Russia Crude Oil Pipeline (CRCOP). The results show that the isolated permafrost in this area is vulnerable to further degradation because of warm oil pipelines and thermal erosion of rivers and ponds. The isolated permafrost is degrading in three directions at the MDS304 site. Specifically, the boundary between permafrost and talik is on both sides of the CRCOP, and permafrost is distributed as islands along a cross-section with a length of about 58–60 m. At present, the vertical hydrothermal influence range of the CRCOP increased to about 10–12 m. The active layer thickness has increased at a rate of 2.0 m/a from about 2.4–6.8 m to 2.5–10.8 m from 2019 to 2021 along this cross-section. Permafrost degradation on the side of the CRCOP’s second line is more visible due to the river’s lateral thermal erosion, where the talik boundary has moved eastward about 12 m during 2018–2022 at a rate of 3.0 m/a. It is 2.25 times the westward moving speed of the talik boundary on one side of the CRCOP’s first line. In contrast, the talik boundary between the CRCOP’s first line and the G111 highway also moves westward by about 4 m in 2019–2022. Moreover, the maximum displacement of the CRCOP’s second line caused by the thawing of frozen soil has reached up to 1.78 m. The degradation of permafrost may threaten the long-term stability of the pipeline. Moreover, the research results can provide a useful reference for decision-makers to reduce the risk of pipeline freeze-thaw hazards.

Integrated Interpretation of Electrical Resistivity Tomography for Evaporite Rock Exploration: A Case Study of the Messinian Gypsum in the Sorbas Basin (Almería, Spain)

In this study, we conduct an investigation of the Sorbas Basin (Almería, Spain) on the Messinian gypsum unit using geophysical prospecting methods. Geophysical electrical resistivity tomography (ERT) methods were applied to study the subsurface of this gypsum unit, the exploitation of which could be of interest economically, with different commercial specifications for alabastrine and selenitic gypsums. For the interpretation of the different ERT images, the data for the surface geology, borehole cores, and seismic refraction conducted at a point within the ERT profiles were used. The results obtained from this investigation can be used as a reference for other similar studies in other regions. It was observed that selenitic gypsum is more resistive than alabastrine gypsum; therefore, the diagenetic processes of dehydration (anhydritization) and hydration (gypsification) increase the “percolation” phenomenon through possibly ensuring a greater connectivity of the shale matrix. Fracturing and moisture can be used to fully determine the resistivity of the purest and most resistive gypsum, to the point of considerably lowering the resistivity in an entire area affected by fracturing. The use of different tests with different lengths for the same profile can help one better understand the structure of the gypsum body in the subsurface, especially when there are shale intercalations or more- or less-pure levels of gypsum that do not reach a value of a few meters in thickness, because these thinner levels of a few meters are not defined in the ERT images when the test is performed at depths of up to 75 m.

3D Electrical Resistivity Imaging to Determine the Internal Causal factor for leaning of Minaret of Mosque in Baghdad

The cheapest, easiest, fastest and relatively accurate technique in geophysics exploration is the two-dimensional (2D) and the three-dimensional (3D) resistivity imaging method. 2D electrical resistivity image has been used to image the subsurface structure to discover the causal factor of leaning of the minaret of mosque in Baghdad- Iraq. Eight 2D images have been made to create 3D image for imaging subsurface structure of the study area. The length and width of the image are 80 m with a depth about 15 m. The results showed that there are four types of materials and weak and saturated area has been discovered in area of study. The conclusions have been qualified (confirmed) by using borehole data from the research region.

Computer model for acquisition, processing, analysis and visualization of measurement data by electrical impedance tomography

In this article, a computer model for the collection, processing, analysis and visualization of measurement data by the method of electrical impedance tomography has been developed. The essence of the work performed is to build a computer model of the phantom, electrodes, conductive medium and inhomogeneities in the COMSOL Multiphysics environment to study the conductivity of objects in order to reduce the error in determining the size of the inhomogeneity in the reconstructed image. The study is performed by placing an inhomogeneity phantom in the conductive region. Based on the data obtained, a matrix of potential differences is constructed using the “near neighbor” algorithm. On its basis, in the GNU Octave environment, using the EIDORS package, a reconstruction of the conducted studies is obtained. With the help of the developed software that implements the method and algorithm for refining the boundaries, filtered images of reconstructions of the results of the studies are obtained. The use of this algorithm makes it possible to reduce the measurement error by a factor of 1.44.

The Methods for Measuring the Area of Heartwood and Sapwood

The presence of heartwood is one of the most important parameters in economic value that can affect the final use of wood. For construction materials, the presence of a large heartwood is very necessary, considering this will have a positive effect on the basic properties of the wood. Meanwhile, its heartwood proportion has a negative impact on the quality of pulp and paper. Therefore, this study aims to review the methods of calculating the area of the heartwood by using suitable heartwood measurement methods for better stand management and selection of raw materials in the industry. The results of the review showed that non-destructive methods were developed to measure the area of heartwood-sapwood such as computer tomography (CT-image), electrical resistivity tomography (ERT), x-ray densitometry, time-of-flight secondary ion mass spectrometry (TOF-SIMS), and near-infrared spectroscopy (NIRS), so far. This is different from the destructive methods that are commonly used for visual inspection of wood color and anatomical features. During the selection of methods, the factors to be considered include cost, time, tree damage, number of samples, and efficiency in applying the methods. The visual inspection is the best option for determining the area of heartwood and sapwood. But, when the heartwood and sapwood boundaries color is unclear, chemical indicators can be performed. In addition, the ERT method can be the second alternative for tropics species. Furthermore, the method used in a study of the presence of heartwood depends on the five limiting factors mentioned previously.

Nonlinear Electrical Impedance Tomography Method Using a Complete Electrode Model for the Characterization of Heterogeneous Domains

This paper presents an electrical impedance tomography (EIT) method using a partial-differential-equation-constrained optimization approach. The forward problem in the inversion framework is described by a complete electrode model (CEM), which seeks the electric potential within the domain and at surface electrodes considering the contact impedance between them. The finite element solution of the electric potential has been validated using a commercial code. The inverse medium problem for reconstructing the unknown electrical conductivity profile is formulated as an optimization problem constrained by the CEM. The method seeks the optimal solution of the domain’s electrical conductivity to minimize a Lagrangian functional consisting of a least-squares objective functional and a regularization term. Enforcing the stationarity of the Lagrangian leads to state, adjoint, and control problems, which constitute the Karush-Kuhn-Tucker (KKT) first-order optimality conditions. Subsequently, the electrical conductivity profile of the domain is iteratively updated by solving the KKT conditions in the reduced space of the control variable. Numerical results show that the relative error of the measured and calculated electric potentials after the inversion is less than 1%, demonstrating the successful reconstruction of heterogeneous electrical conductivity profiles using the proposed EIT method. This method thus represents an application framework for nondestructive evaluation of structures and geotechnical site characterization.

Response to soil compaction of the electrical resistivity tomography, induced polarisation, and electromagnetic induction methods: a case study in Belgium

Context Soil compaction acts at different scales and is challenging to measure on field scales. Aims To evaluate soil compaction under a controlled traffic experiment, using three different geophysical methods. Methods Electrical Resistivity Tomography (ERT), Electromagnetic Induction (EMI), and Induced Polarisation (IP) were selected to map soil compaction. Two different ERT arrays and EMI geometries were selected with different spacings. The influences of configuration, electrode spacing, and the Depth of Investigation Index (DOI) were evaluated. Soil physical properties were measured in the Laboratory and in the field. Error models were developed to assess the accuracy of the ERT profiles and later correlated with EMI and soil physical results. Key results Penetration resistance measurements identified a compacted layer at 25 to 35 cm depth with a maximum value of 5 MPa under fixed tracks and bulk density of 1.52 Mg m−3, while lowest values were 1.4 MPa and 1.36 Mg m−3. The dipole–dipole (DD) 10 cm array was more accurate towards both soil properties and locating the zones of high resistivity. The IP method identified chargeability anomalies at the same depth as the resistivity anomalies, possibly indicating a similar origin. The EMI test was less successful in accurately determining the locations of the conductive areas. Conclusions A clear relationship between the absolute value of the resistivity/conductivity signals with the level of compaction was not found, yet patterns of lateral variations in resistivity were identified. Implications Further studies are needed to establish the concrete relationship between soil compaction and geophysical signals.

Application of Electrical Resistivity Imaging (ERI) to analyse geological structure in Paloh, Gua Musang, Kelantan

This study is conducted in Paloh, Gua Musang district, Kelantan and focussed on Electrical Resistivity Imaging (ERI) method which was carried out to assess the subsurface geological structures of Paloh area, Gua Musang. Two survey lines of 200 m are conducted in the study area by using Gradient array. The data obtained from ERI is processed by using RES2DINV software to produce a pseudosection model. Variable resistivity values ranging from 1Ωm to >3000Ωm with a depth of investigation of approximately 40m showed in the pseudosection model. Based on the model, the structural analysis of the study area resulted in several fault lines occurring in the subsurface area. The fault line indicated the occurrence of subsurface movement. This study suggests further geophysical investigation (seismic survey) be carried out as it would be able to give the extent of information on subsurface geological structures for geoengineering study for site investigation.

Uncertainty Quantification in Intelligent-Based Electrical Resistivity Tomography Image Reconstruction With Monte Carlo Dropout Strategy

The Electrical Resistivity Tomography (ERT) method plays an essential role in researching electrical properties at the core scale, and the resistivity image inversion reconstruction technique is the key to constraining measurement accuracy. In recent years, with the improvement of computer operation capabilities and the acquisition of a large amount of geophysical data, the inversion algorithms represented by machine learning (ML) have made it possible to reconstruct ERT images automatically. However, the solution of ERT image inversion has large uncertainty due to high nonlinearity. Meanwhile, traditional ML methods are not initially developed for evaluating the uncertainty of such reconstruction results. In this paper, we propose a novel MC-Net ML scheme to quantitatively estimate the uncertainty of the ERT image reconstruction by introducing the Monte-Carlo Dropout strategy with a multiple stochastic method to approximate Bayesian inference. As one important evaluation index, the correlation coefficient (CC) between target image and reconstructed image is used to compare the effects of six types of simulated data carried by different ML schemes. The inversion results show that MC-Net increases lower limit of CC on the premise of ensuring average CC. Moreover, the accuracy of regression and classification results defined in this paper also obtained the highest value by MC-Net. With the proposed scheme, we can further provide a quantification of the uncertainty to determine whether reconstructed images are reliable after reconstructing the ERT images, which is of great significance for actual intelligent inversion tasks.

Machine Learning Applied to EIT: Image Reconstruction, Displaced Electrodes and Boundary Shape

Electrical Impedance Tomography (EIT) is a noninvasive, nonradioactive technique used to obtain an internal image of a body or object. This work addresses several EIT-related problems using Artificial Neural Networks (ANNs). The EIT method consists of performing measurements of electrical potentials in a set of electrodes while different patterns of electrical current are applied. This work proposes the use of a modified version of LeNet-5 to determine the conductivity distribution within the domain, resulting in an image of the interior of the boundary in which the electrodes are positioned. There are two sources of errors in the usage of EIT in medical images. The first one is the displacement of the electrodes, which are supposed to be equidistant; and the second one is the modification of the boundary shape. The human body is not rigid, and due to breathing the boundary shape changes, leading to the displacement of electrodes. A modified version of LeNet-5 is also used to determine the changes of the boundary shape, and classifiers, such as Support Vector Machine (SVM), Random Forest and K-nearest Neighbors (KNN), are used to determine the displaced electrodes. In this work, all methods are used apart. Then, the reconstruction algorithm does not yet consider these sources of errors.