Metal Particle Detection by Integration of a Generative Adversarial Network and Electrical Impedance Tomography (GAN-EIT) for a Wet-Type Gravity Vibration Separator

The minor copper (Cu) particles among major aluminum (Al) particles have been detected by means of an integration of a generative adversarial network and electrical impedance tomography (GAN-EIT) for a wet-type gravity vibration separator (WGS). This study solves the problem of blurred EIT reconstructed images by proposing a GAN-EIT integration system for Cu detection in WGS. GAN-EIT produces two types of images of various Cu positions among major Al particles, which are (1) the photo-based GAN-EIT images, where blurred EIT reconstructed images are enhanced by GAN based on a full set of photo images, and (2) the simulation-based GAN-EIT images. The proposed metal particle detection by GAN-EIT is applied in experiments under static conditions to investigate the performance of the metal detection method under single-layer conditions with the variation of the position of Cu particles. As a quantitative result, the images of detected Cu by GAN-EIT ψ̿GAN in different positions have higher accuracy as compared to σ*EIT. In the region of interest (ROI) covered by the developed linear sensor, GAN-EIT successfully reduces the Cu detection error of conventional EIT by 40% while maintaining a minimum signal-to-noise ratio (SNR) of 60 [dB]. In conclusion, GAN-EIT is capable of improving the detailed features of the reconstructed images to visualize the detected Cu effectively.

A noise-controlling method by hybrid current-stimulation and voltage-measurement for electrical impedance tomography (HCSVM-EIT)

Image reconstruction in electrical impedance tomography (EIT) is a typical ill-posed inverse problem, from which the stability of conductivity reconstruction affects the reliability of physiological parameters evaluation. In order to improve the stability, the effect of boundary voltage noise on conductivity reconstruction should be controlled. A noise-controlling method based on hybrid current-stimulation and voltage-measurement for EIT (HCSVM-EIT) is proposed for stable conductivity reconstruction. In HCSVM-EIT, the boundary voltage is measured by one current-stimulation and voltage-measurement pattern (high-SNRpattern) with a higher signal-to-noise ratio (SNR); the sensitivity matrix is calculated by another current-stimulation and voltage-measurement pattern (low-condpattern) with a lower condition number; the boundary voltage is then transformed from thehigh-SNRpattern into thelow-condpattern by multiplying by an optimized transformation matrix for image reconstruction. The stability of conductivity reconstruction is improved by combining the advantages of thehigh-SNRpattern for boundary voltage measurement and thelow-condpattern for sensitivity matrix calculation. The simulation results show that the HCSVM-EIT increases the correlation coefficient (CC) of conductivity reconstruction. The experiment results show that theCCof conductivity reconstruction of the human lower limb is increased from 0.3424 to 0.5580 by 62.97% compared to the quasi-adjacent pattern, and from 0.4942 to 0.5580 by 12.91% compared to the adjacent pattern. In conclusion, the stable conductivity reconstruction with higherCCin HCSVM-EIT improves the reliability of physiological parameters evaluation for disease detection.

Physiological-induced conductive response evaluation in specific muscle compartments under hybrid of electrical muscle stimulation and voluntary resistance training by electrical impedance tomography

Objective: The physiological-induced conductive response has been visualised for evaluation in specific muscle compartments under hybrid (hybridEMS) of electrical muscle stimulation (EMS) and voluntary resistance training (VRT) by electrical impedance tomography (EIT). Methods: In the experiments, tendency of conductivity distribution images σ over time was clearly detected for three specific muscle compartments, which are called AM 1 compartment composed of biceps brachii muscle, AM 2 compartment composed of triceps brachii muscle, and AM 3 compartment composed of brachialis muscle, under three training modalities. Results: From the experimental results, the tendency of physiological-induced conductive response are increased in all three training modalities with increasing training time. Correspondingly, the spatial-mean conductivity <σ>AM1,AM2,AM3 increased with the conductance value G and extracellular water ratio β of right arm by bio-impedance analysis (BIA) method. In addition, hybridEMS has the greatest effect on physiological-induced conductive response in AM 1, AM 2, and AM 3. Under hybridEMS, the spatial-mean conductivity increased from <σ pre > AM1 = 0.154 to <σ 23mins > AM1 = 0.810 in AM 1 muscle compartment (n = 8, p < 0.001); <σ pre > AM2 = 0.040 to <σ 23mins > AM2 = 0.254 in AM 2 muscle compartment (n = 8, p < 0.05); <σ pre > AM3 = 0.078 to <σ 23mins > AM3 = 0.497 in AM 3 muscle compartment (n = 8, p < 0.05). Conclusion: The paired-samples t-test results of <σ>AM1,AM2,AM3 under all three training modalities suggest hybridEMS has the most efficient elicitation on physiological induced conductive response compared to VRT and EMS. The effect of EMS on deep muscle compartment (AM 3) is slower compared to VRT and hybridEMS, with a significant difference after 15 min of training.

A High Accuracy Voltage Approximation Model Based on Object-oriented Sensitivity Matrix Estimation (OO-SME Model) in Electrical Impedance Tomography

The image reconstruction in electrical impedance tomography (EIT) has low accuracy due to the approximation error between the measured voltage change and the approximated voltage change, from which the object cannot be accurately reconstructed and quantitatively evaluated. A voltage approximation model based on object-oriented sensitivity matrix estimation (OO-SME model) is proposed to reconstruct the image with high accuracy. In the OO-SME model, a sensitivity matrix of the object-field is estimated, and the sensitivity matrix change from the background-field to the object-field is estimated to optimize the approximated voltage change, from which the approximation error is eliminated to improve the reconstruction accuracy. Against the existing linear and nonlinear models, the approximation error in the OO-SME model is eliminated, thus an image with higher accuracy is reconstructed. The simulation shows that the OO-SME model reconstructs a more accurate image than the existing models for quantitative evaluation. The relative accuracy (RA) of reconstructed conductivity is increased up to 83.98% on average. The experiment of lean meat mass evaluation shows that the RA of lean meat mass is increased from 7.70% with the linear model to 54.60% with the OO-SME model. It is concluded that the OO-SME model reconstructs a more accurate image to evaluate the object quantitatively than the existing models.

Electrical-tomographic imaging of physiological-induced conductive response in calf muscle compartments during voltage intensity change of electrical muscle stimulation (vic-EMS)

Objectives. The electrical-tomographic imaging has been achieved for exploring differential tendency of physiological-induced conductive response in calf muscle compartments during voltage intensity change of electrical muscle stimulation (vic-EMS).Approach. In the experiments, the differential tendency of conductivity distribution imagesσduringvic-EMS were clearly imaged as three responsive muscle compartments, which are calledM₁compartment composed of gastrocnemius muscle,M₂compartment composed of tibialis anterior, extensor digitorum longus, and peroneus longus muscles, andM₃compartment composed of soleus muscle.Main results. The differential tendency of spatial-mean conductivity 〈σ〉_M1is the same as the differential tendency of venous blood flow velocityv_bland blood lactate concentrationC_blduringvic-EMS by the increased tendency of spatial-mean conductivity difference Δ〈σ〉_M1, venous blood flow velocity difference Δv_bland blood lactate concentration difference ΔC_bl. The 〈σ〉_M1is increased with the increase of voltage intensity from 〈σ^pre〉_M1 = 0.142 [-] to 〈σ^l14 〉 _M1 = 0.442 [-] (pre: pre-training,l14: voltage level duringvic-EMSl = 14) by Δ〈σ^l14-pre〉_M1 = 204.2% (n = 16,p < 0.01). Correspondingly, thev_blandC_blare increased with the increase of voltage intensity by Δv_bl^l14-pre= 1480.5% (n = 16,p < 0.01) and ΔC_bl^l14-pre= 230.1% (n = 16,p < 0.01) respectively.Significance: The reason for the differential tendency of increase in <σ>_M1suggests an increase in muscle extracellular volumes duringvic-EMS due to the co-effect of venous blood flow velocity and blood lactate metabolism. Based on the conductivity second-order difference images∂2σM1φ∂φ2φand spatial-mean conductivity second-order difference∂2σM1φ∂φ2φ,optimum voltage intensityφ_OVIis discussed among sixteen volunteer subjects, which increased with a thicker subcutaneous fat layer.

In-vivoviscoelastic properties estimation in subcutaneous adipose tissue by integration of poroviscoelastic-mass transport model (pve-MTM) into wearable electrical impedance tomography (w-EIT)

In-vivoviscoelastic properties have been estimated in human subcutaneous adipose tissue (SAT) by integration of poroviscoelastic-mass transport model (pve-MTM) into wearable electrical impedance tomography (w-EIT) under the influence of external compressive pressure-P.Thepve-MTM predicts the ion concentration distributioncmod(t)by coupling the poroviscoelastic and mass transport model to describe the hydrodynamics, rheology, and transport phenomena inside SAT. Thew-EIT measures the time-difference conductivity distribution∆γ(t)in SAT resulted from the ion transport. Based on the integration, the two viscoelastic properties which are viscoelastic shear modulus of SATGvand relaxation time of SATτvare estimated by applying an iterative curve-fitting between the normalized average ion concentration distributioncˆmod(t)predicted frompve-MTM and the experimental normalized average ion concentration distributioncˆexp(t)derived fromw-EIT. Thein-vivoexperiments were conducted by applying external compressive pressure-Pon human calf boundary to induce interstitial fluid flow and ion movement in SAT. As a result, the value ofGvwas range from 4.9-6.3 kPa and the value ofτvwas range from 27.50-38.5 s with the value of average goodness-of-fit curve fittingR² > 0.76. These values ofGvandτvwere compared to the human and animal tissue from the literature in order to verify this method. The results frompve-MTM provide evidence thatGvandτvplay a role in the predicted value ofcˆmod.

Evaluation of the effectiveness of electrical muscle stimulation on human calf muscles via frequency difference electrical impedance tomography

Objectives. The human skeletal muscle responds immediately under electrical muscle stimulation (EMS), and there is an immediate physiological response in human skeletal muscle. Non-invasive quantitative analysis is at the heart of our understanding of the physiological significance of human muscle changes under EMS. Response muscle areas of human calf muscles under EMS have been detected by frequency difference electrical impedance tomography (fd-EIT).Approach. The experimental protocol consists of four parts: pre-training (pre), training (tra), post-training (post), and relaxation (relax) parts. The relaxation part has three relaxation conditions, which are massage relaxation (MR), cold pack relaxation (CR), and hot pack relaxation (HR).Main results. From the experimental results, conductivity distribution imagesσ^p(pmeans protocol = pre,tra,post,or relax) are clearly reconstructed byfd-EIT as response muscle areas, which are called theM₁response area (composed of gastrocnemius muscle) and theM₂response area (composed of the tibialis anterior muscle, extensor digitorum longus muscle, and peroneus longus muscle). A paired samplest-test was conducted to elucidate the statistical significance of spatial-mean conductivities 〈σ^p〉_M1and 〈σ^p〉_M2inM₁andM₂with reference to the conventional extracellular water ratioβ^pby bioelectrical impedance analysis. Significance. From thet-test results, 〈σ^p〉_M1and〈σ^p〉_M2have good correlation withβ^p. In the post-training part, 〈σ^post〉 andβ^postwere significantly higher than in the pre-training part (n = 24,p < 0.001). The relax-pre difference ratios of spatial-mean conductivity Δ〈σ^relax-pre〉 and the relax-pre difference ratios of extracellular water ratio Δβ^relax-prein both MR and CR were lower; on the contrary, the Δ〈σ^relax-pre〉 and Δβ^relax-prein HR were significantly higher than those in post-pre difference ratios of spatial-mean conductivity Δ〈σ^post-pre〉 (n = 8,p < 0.05). The reason for the changes in 〈σ^p〉_M1and 〈σ^p〉_M2are caused by the changes in muscle extracellular volumes. In conclusion,fd-EIT satisfactorily evaluates the effectiveness of human calf muscles under EMS.