A Hybrid Particle Swarm Optimization (HPSO) algorithm is proposed to optimize the Radial Basis Function Neural Network (RBFNN) for the image reconstruction of Electrical Impedance Tomography (EIT). This algorithm combines the non-linear approximation ability of RBFNN with the mutation ability of HPSO based on the simulated annealing, which has advantages of the strong global search ability and high robustness. A non-linear mapping relationship model between the boundary voltage and the conductivity distribution is established through HPSO-RBFNN to realize the image reconstruction, skipping the solution of the Jacobian matrix. The Structural similarity (SS) and correlation coefficient (CC) are used as the evaluation criteria for the quality of the reconstructed images in the simulations and experiments. Firstly, the HPSO-RBFNN method is compared with the Landweber and RBFNN methods through a single-target simulation, which verifies the accuracy of the image reconstruction. Secondly, the Gaussian noise is added to the ideal simulation data, the results show an anti-noise capability of the proposed method. Then, the multi-target simulation is conducted using the single-target training model, which verifies the generalization ability of the proposed method. Finally, the experiments with multiple fish eggs are performed to further verify the effectiveness of the HPSO-RBFNN method. The results show that the method proposed in this paper has higher SS and CC, which indicates that the HPSO-RBFNN method has a higher accuracy.

The poor spatial resolution of electrical impedance tomography (EIT) limits its use in medical devices because of its highly nonlinear and ill-posed natures. A novel hierarchical block sparse Bayesian learning (BSBL) method is designed for lung respiratory monitoring with EIT. It is its excellent modeling capability and noise robustness that allows BSBL to adaptively explore and exploit the internal conductivity distribution, e.g., block sparsity and intra-block correlation. Firstly, K-Nearest Neighbor (KNN) strategy is employed to automatically group each block based on the clustering property and incorporated to constrain the intra-block correlation matrix based on the spatial distance. Secondly, a three-layer hierarchical BSBL model using weighted Laplace (WL) prior is considered to enhance the recovery performance. Finally, an efficient bound optimization (BO) method is performed for Bayesian inference, avoiding the tedious parameter adjustment. This leads to the improvements in recovery performance, algorithm robustness and computational efficiency. Moreover, root mean square error (RMSE), image correlation coefficient (ICC), and relative size coverage ratio (RCR) are applied for quantitative comparisons of image quality. Numerical simulations and in vivo lung respiratory experiments showed that the proposed KNN-BSBL-WL method outperforms existing referenced methods in terms of reconstruction accuracy and computational time. On average, KNN-BSBL-WL obtains the RMSE = 0.07±0.02, ICC = 0.96±0.01, and RCR = 0.98±0.05, which are closest to the true values of the recorded snapshots obtained in the experiments. Meanwhile, the average time of KNN-BSBL-WL is 0.16 second, achieving an acceleration ratio close to ~3 compared with the referenced KNN-BSBL method. Therefore, the proposed method is an efficient and robust means of imaging reconstruction, which further improves the feasibility of EIT in respiratory clinical applications.

A flexible Electrical Impedance Tomography (EIT) sensor with combined electrodes is proposed to reconstruct 3D image. The 3D EIT sensor consists of 24 combined electrodes, each consisting of a center electrode and an around electrode. The accuracy of measurement data has been improved by using the combined electrodes. In this study, an innovative U2-Net neural network structure is implemented to solve the inverse problem of 3D EIT. Significant amounts of priori information obtained from simulation are stored in the U2-Net, which facilitates the improvement of image reconstruction quality. The proposed 3D EIT sensor is verified by simulation and experiment. In the simulations, U2-Net shows better performance than U-Net. Combined electrodes show better robustness than conventional structure which shows higher accuracy when the signal to noise ratio (SNR) is 20 dB or 30 dB. In the experiments, structural similarity index measure (SSIM) is proposed to evaluate the image reconstruction quality. Experimental results show that the proposed 3D EIT sensor using U2-Net improves accuracy of image reconstruction. In addition, the proposed 3D EIT sensor improves the robustness of measurement results, which is satisfactory for the requirements of 3D image reconstruction.

Pneumatic muscle actuator (PMA) similar to biological muscle is a new type of pneumatic actuator. The flexible manipulator based on PMAs was constructed to simulate the actual movement of the human upper arm. Considering the model errors and external disturbances, the fuzzing sliding mode control based on the saturation function was proposed. Compared with other fuzzy control methods, fuzzy control and saturation function are used to adjust the robust terms to improve the tracking accuracy and reduce the high-frequency chattering.

Bioelectrical Impedance Spectroscopic Tomography (BIST) has been proposed for the detection of tongue tumors. The proposed method is based on Electrode Array Sensor that enables the localization of tongue tumor and quantitative evaluation of the maximum tumor dimension ${d}_{\max}$ . In this study, the detection capability of Electrical Impedance Tomography (EIT) at different positions of the Electrode Array Sensor is investigated by numerical analysis. In addition, the detection sensitivity of Bioelectrical impedance spectroscopy (BIS) of tumor tissue at different positions of the sensor is also investigated. The sensitivity of the sensor is higher than the other excitation fields when the tumor tissue is located in the center of the excitation field. Adipose tissue from beef and porcine is selected to mimic tongue tissue. Five tongue models prepared with ${d}_{\max} =3$ mm, ${d}_{\max}=4.5$ mm, ${d}_{\max}=6$ mm, ${d}_{\max}=7.5$ mm and ${d}_{\max}=9$ mm. The results of BIST detection experiments showed that BIST method is able to localize tumor tissue of different dimensions on the tongue, with the best EIT imaging at Pos2 of the sensor. The ${Z}''_{{\mathrm {relax}}}$ is found to be different dimensions of tumor tissue by BIS detection, which increased with the increase of ${d}_{\max}$ . Finally, curve fitting is performed between ${Z} ''_{{\mathrm {relax}}}$ and ${d}_{\max}$ of the tumor tissue. The ${d}_{\max}$ of tumor tissue is calculated according to ${Z} ''_{{\mathrm {relax}}}$ , which helps doctors determine the Tumor Node Metastasis (TNM) classification of tumor tissue.

A hybrid Genetic Expression Program and Genetic Algorithm (GEP-GA) is proposed for auto-modeling the Electrical Equivalent Circuit (EEC) of particle structure measurement with Electrochemical Impedance Spectroscopy (EIS). It combines the GEP with the ability of searching mathematical models and the GA with the ability of electrical parameters global searching. The GEP is used to modeling EEC of EIS, the GA is used to calculate the electrical parameters and fitness of the EEC. The hybrid method has the advantage of no need manually set. Firstly, the method is used to reconstruct EEC of simulation model and simultaneously calculate its electrical parameters. The results show that the auto-modeling EEC is same as the simulation’s, and the accuracy of calculating electrical parameters over 99.5%. Secondly, a portable and detachable sensor is proposed to measure the EIS of particle solution. Finally, the GEP-GA is verified by auto-modeling the EEC of 10 μm Poly Methyl Methacrylate (10 PMMA), the linear correlation coefficient of EIS between experiment and auto-modeling EEC R2=0.99874. Meanwhile, the EIS of 10 μm Polystyrene Magnetic particles (10 PSM), and 10 μm, 20 μm, 30 μm Polystyrene particles (10 PS, 20 PS, 30 PS) are also fitted with the EEC model of 10 PMMA. The results show that their linear correlation coefficient R2>0.998. It indicates that the hybrid of GEP-GA can auto-modeling the EEC and simultaneously calculate the electrical parameters. Furthermore, the electrical characters of the particle suspension and the particle structure can be measured with these electrical parameters.