Waldemar Tomasz Smolik

Impulse-based impedance sensing with capacitively coupled electrodes is introduced as a fast, non-contact, and simplified complementary method to conventional capacitive impedance measurements. Unlike frequency-domain methods, the proposed approach derives effective resistive and capacitive properties of a sample from the transient response to a single rectangular pulse. The equivalent circuit model comprises three elements: sample resistance, sample capacitance, and electrode coupling capacitance. From this model, analytical expressions of the transient response were derived, enabling accurate simulation of measured signals and providing the basis for both phantom verification and machine learning training. Importantly, the coupling capacitance, typically considered a limitation in contactless methods, is estimated alongside the sample parameters, providing insight into electrode–object coupling conditions. A machine-learning model trained on simulated circuit responses, including noise and temporal variability, is employed as a low-latency estimator for extracting parameters from measured transient signals. Experimental validation was carried out using a configurable lumped-element equivalent circuit and NaCl solutions of controlled conductivity, cross-verified with conductometric measurements and numerical probe simulations. Across a tested conductivity range, the method achieved estimation errors of 2–8%. The proposed approach is intended as a low-latency measurement strategy for simplified capacitively coupled impedance sensing, with potential relevance to future capacitively coupled electrical impedance tomography systems, where rapid acquisition of boundary measurements is prioritized over full frequency-resolved impedance spectroscopy.

The aim of the article was to analyze the potential simultaneous use of magnetic nanoparticles as contrast agents in MRI imaging and for magnetic hyperthermia. The study proposed characterizing the nanoparticles using various measurement methods in order to investigate the relationships between different properties. The first stage involved measuring images of nanoparticle samples using scanning transmission electron microscopy (TEM) and dynamic light scattering (DLS). The diameter distribution of nanoparticles was determined based on image segmentation. The next step involved measuring relaxation properties of nanoparticles in low and high magnetic fields. The research was carried out for nanoparticle solutions of various concentrations and properties. The last step was measuring calorimetric properties of nanoparticles as a thermal source under alternating magnetic field excitation conditions. The range of nanoparticle diameters (20–25 nm) for which maximum losses occur in an alternating magnetic field corresponds to the diameter range in which the maximum r₂ relaxivity is observed.

This study investigates the influence of variable electrode positions on image reconstruction in capacitively coupled electrical impedance tomography (CC-EIT) of the human thorax. Images were reconstructed by an adversarial neural network trained on a synthetic dataset generated using a tomographic model that included a wearable elastic band with 32 electrodes attached. Dataset generation was conducted using a previously developed numerical phantom of the thorax, combined with a newly developed algorithm for random selection of electrode positions based on physical limitations resulting from the elasticity of the band and possible position inaccuracies while putting the band on the patient’s chest. The thorax phantom included the heart, lungs, aorta, and spine. Four training and four testing datasets were generated using four different levels of electrode displacement. Reconstruction was conducted using four versions of neural networks trained on the datasets, with random ellipses included and noise added to achieve an SNR of 30 dB. The quality was assessed using pixel-to-pixel metrics such as the root-mean-square error, structural similarity index, 2D correlation coefficient, and peak signal-to-noise ratio. The results showed a strong negative influence of electrode displacement on reconstruction quality when no samples with displaced electrodes were present in the training dataset. Training the network on the dataset containing samples with electrode displacement allowed us to significantly improve the quality of the reconstructed images. Introducing samples with misplaced electrodes increased neural network robustness to electrode displacement while testing.

The aim of the study was to analyze the impedance characteristics of the transmitting coil used in a system for AC magnetometry and the measurement of the Specific Absorption Rate (SAR) of magnetic nanoparticles. A theoretical relationship for the current distribution in a multilayer air-core coil was derived. The formulas for the coil’s stray capacitance were modified to account for additional spacing between layers, introduced to reduce the interlayer capacitance. The developed theory was applied to a two-layer air-core coil with an additional gap between the layers. The frequency dependence of the coil impedance was measured. The measurements confirmed an extension of the useful operating frequency range of the constructed coil.

The assessment of superparamagnetic nanoparticle heating is crucial for effective hyperthermia. AC magnetometry can be used to determine the specific absorption rate (SAR) of nanoparticles, assuming proper calorimetric calibration. We show that an AC magnetometer developed in our laboratory can be used simultaneously as a calorimeter for calibrating measurements. An electrical circuit with lumped parameters that are equivalent to the non-adiabatic calorimeter and that incorporates the effects of heat flow from the excitation coil, the surrounding environment, and the sample is presented. Quantitative thermal system identification was performed using global optimization, which fitted the temperature measured by the three fiber-optic probes to the simulated temperature transient curves. The identified model was used to estimate the thermal power generated in the measurement sample using a resistor with a controlled current value. The results demonstrate significant error reduction, particularly at lower heating powers, where external heat transfer becomes more influential. At low heating power values (around 25 mW), the error was reduced from 16.09% to 2.36%, with less pronounced improvements at higher power levels. The model achieved an overall accuracy of less than 2.5% across the 20–200 mW calibration range, a substantial improvement over the corrected-slope method. The value of the true thermal power of nanoparticles can be obtained using the calibrated calorimeter.

This study investigated the potential of the generative adversarial neural network (cGAN) image reconstruction in industrial electrical capacitance tomography. The image reconstruction quality was examined using image patterns typical for a two-phase flow. The training dataset was prepared by generating images of random test objects and simulating the corresponding capacitance measurements. Numerical simulations were performed using the ECTsim toolkit for MATLAB. A cylindrical sixteen-electrode ECT sensor was used in the experiments. Real measurements were obtained using the EVT4 data acquisition system. The reconstructed images were evaluated using selected image quality metrics. The results obtained using cGAN are better than those obtained using the Landweber iteration and simplified Levenberg–Marquardt algorithm. The suggested method offers a promising solution for a fast reconstruction algorithm suitable for real-time monitoring and the control of a two-phase flow using ECT.

This study investigated the potential of machine-learning-based stroke image reconstruction in capacitively coupled electrical impedance tomography. The quality of brain images reconstructed using the adversarial neural network (cGAN) was examined. The big data required for supervised network training were generated using a two-dimensional numerical simulation. The phantom of an axial cross-section of the head without and with impact lesions was an average of a three-centimeter-thick layer corresponding to the height of the sensing electrodes. Stroke was modeled using regions with characteristic electrical parameters for tissues with reduced perfusion. The head phantom included skin, skull bone, white matter, gray matter, and cerebrospinal fluid. The coupling capacitance was taken into account in the 16-electrode capacitive sensor model. A dedicated ECTsim toolkit for Matlab was used to solve the forward problem and simulate measurements. A conditional generative adversarial network (cGAN) was trained using a numerically generated dataset containing samples corresponding to healthy patients and patients affected by either hemorrhagic or ischemic stroke. The validation showed that the quality of images obtained using supervised learning and cGAN was promising. It is possible to visually distinguish when the image corresponds to the patient affected by stroke, and changes caused by hemorrhagic stroke are the most visible. The continuation of work towards image reconstruction for measurements of physical phantoms is justified.

Electrical impedance tomography (EIT) is a non-invasive technique for visualizing the internal structure of a human body. Capacitively coupled electrical impedance tomography (CCEIT) is a new contactless EIT technique that can potentially be used as a wearable device. Recent studies have shown that a machine learning-based approach is very promising for EIT image reconstruction. Most of the studies concern models containing up to 22 electrodes and focus on using different artificial neural network models, from simple shallow networks to complex convolutional networks. However, the use of convolutional networks in image reconstruction with a higher number of electrodes requires further investigation. In this work, two different architectures of artificial networks were used for CCEIT image reconstruction: a fully connected deep neural network and a conditional generative adversarial network (cGAN). The training dataset was generated by the numerical simulation of a thorax phantom with healthy and illness-affected lungs. Three kinds of illnesses, pneumothorax, pleural effusion, and hydropneumothorax, were modeled using the electrical properties of the tissues. The thorax phantom included the heart, aorta, spine, and lungs. The sensor with 32 area electrodes was used in the numerical model. The ECTsim custom-designed toolbox for Matlab was used to solve the forward problem and measurement simulation. Two artificial neural networks were trained with supervision for image reconstruction. Reconstruction quality was compared between those networks and one-step algebraic reconstruction methods such as linear back projection and pseudoinverse with Tikhonov regularization. This evaluation was based on pixel-to-pixel metrics such as root-mean-square error, structural similarity index, 2D correlation coefficient, and peak signal-to-noise ratio. Additionally, the diagnostic value measured by the ROC AUC metric was used to assess the image quality. The results showed that obtaining information about regional lung function (regions affected by pneumothorax or pleural effusion) is possible using image reconstruction based on supervised learning and deep neural networks in EIT. The results obtained using cGAN are strongly better than those obtained using a fully connected network, especially in the case of noisy measurement data. However, diagnostic value estimation showed that even algebraic methods allow us to obtain satisfactory results.

Efficient electrical impedance imaging fault detection in grounding grid is necessary. In this paper, we propose an optimized differential-Multigrid-Homotopy algorithm for grounding grid impedance imaging. To enhance the iteration speed, we introduced an optimized differential algorithm to obtain the Jacobian matrix and derive calculation methods for the two-dimensional constraint operator and interpolation operator. Moreover, we utilize multigrid fast iteration and homotopy methods simultaneously to improve the rapidity and ill-posedness of the inverse problem. We validate the proposed algorithm through numerical simulations and experimental studies. First, we verify the time and convergence of the optimized differential algorithm for solving the Jacobian matrix. Secondly, we investigate the CPU time and convergence of image reconstruction based on the Tikhonov algorithm, HT algorithm, and Multigrid-Homotopy (MG-HT) algorithm under different ground grid fault scenarios and noise levels. The simulation and experimental results show that the optimized differential-Multigrid-Homotopy algorithm has better robustness and convergence.

The characterization of nanoparticles is crucial in several medical applications, such as hyperthermic therapy, which heats superparamagnetic nanoparticles with an external electromagnetic field. The knowledge of heating ability (magnetic losses) in AC magnetic field frequency function allows for selecting the optimal excitation. A hybrid system for the characterization of superparamagnetic nanoparticles was designed and tested. The proposed setup consists of an excitation coil and two sensing probes: calorimetric and magnetic. The measurements of the imaginary part of the complex magnetic susceptibility of superparamagnetic nanoparticles are possible in the kilohertz range. The system was verified using a set of nanoparticles with different diameters. The measurement procedure was described and verified. The results confirmed that an elaborated sensor system and measuring procedures could properly characterize the magnetic characteristics of nanoparticles. The main advantage of this system is the ability to compare both characteristics and confirm the selection of optimal excitation parameters.