Electrical impedance tomography (EIT) is a highly promising brain imaging technique. In brain EIT monitoring, 16 electrodes are commonly used to excite the target field and measure boundary voltages. However, factors such as space constraints, patient comfort and emergency scenarios can restrict the number of deployed electrodes. This reduces the number of voltage measurements available for single-frame image reconstruction and significantly degrades the quality of the reconstructed images.Aiming at this issue, this paper proposes a data augmentation method based on temporal convolutional networks (TCN) and bidirectional gated recurrent units (BiGRU) fused with noise perception for the brain EIT. In the proposed method, to establish an accurate mapping relationship between reduced measurements and complete measurements, the complementary strengths of TCN in capturing hierarchical local features and BiGRU in extracting bidirectional long-sequence dependencies are fully exploited. Moreover, a noise perception module incorporating noise estimation and noise-aware attention gates is designed to enhance data augmentation performance under noisy conditions. The results demonstrate that the augmented voltage data exhibits high consistency with the complete voltage data. With the proposed method, image reconstruction quality for cerebral hemorrhage and cerebral ischemia can be significantly improved under both noise-free and noisy conditions.

Electrical impedance tomography (EIT) is a non-invasive functional imaging technology with the advantages of being radiation-free, rapid, portable and low-cost. It has demonstrated great potential for the clinical monitoring of brain diseases. However, this technology suffers from limited spatial resolution owing to the complex anatomical structure of the head, which leads to inaccurate identification of stroke from reconstructed images. To address the issue, we propose a novel stroke identification approach that integrates GADF, CNN and BiLSTM. The method operates directly on the measured EIT voltage data and the process of image-reconstruction is avoided: raw one-dimensional signals are first encoded into two-dimensional feature maps via GADF, which are then fed into a CNN-BiLSTM network for deep feature extraction and classification. The proposed GADF-CNNBiLSTM model is evaluated under a range of challenging conditions, including noise-free and noisy environments, conductivity variations, contact impedance variation and head shape deviation. The results demonstrate that the proposed method has superior accuracy in identifying stroke types compared to baseline models such as CNN, CNN-BiLSTM, CNN-Transformer and GADF-CNN. This study gives a new, reconstruction-free technique for stroke assessment with the brain EIT.

In the dynamic monitoring with electrical impedance tomography (EIT), some unavoidable factors lead to electrode disconnection. Failure data are measured which greatly affects image reconstruction quality. To enhance the accuracy of lung imaging in the presence of electrode disconnection, this work presents a novel failure data correction approach based on shallow convolutional neural network (sCNN). Electrode disconnection is first identified by calculating the average relative change in the measured voltage. Then the method is applied for failure data correction caused by the disconnected electrode. The performance of the proposed method when the electrode is disconnected is evaluated by comparing the predicted data with the normal data. It is found that mean relative boundary voltage variation when the proposed method is used is very similar to the normal case. Besides, the deviation rate of the predicted voltage data approximates 0. Furthermore, image reconstruction of conductivity distribution is investigated for five different models, and disconnection of one electrode and two electrodes are considered. Also, we have tested the robustness of the proposed method to noise interruption. Both quantitative and qualitative evaluations show that reconstructed images are much better when the voltage data corrected by the proposed method is used for image reconstruction. The shape and size of the reconstructed lung are basically the same with the true object. In addition, there are almost no artifacts. To further estimate the proposed method, a phantom experimental validation is carried out. This work offers a choice for accurate image reconstruction of conductivity distribution under electrode disconnection in the lung EIT.

In the brain electrical impedance tomography (EIT), it is sometimes not possible to attach sixteen electrodes due to space constraints. This causes reduced number of boundary measurements and image reconstruction is seriously affected. To improve brain imaging quality in the case of inadequate measurement, a data replenishment method based on a two-stage neural network is proposed. Four types of electrode-missing models are considered and the objective of the study is to replenish the missing voltage data with the two-stage neural network. The first stage is conducted for electrode-missing identification with residual neural network (ResNet) and the second stage is performed for voltage data replenishment with conditional generative adversarial network (CGAN). The total accuracy rate of classification is nearly 99.7% suggesting that the electrode-missing type can be well identified. The replenished voltage data is compared with the measured voltage of the 16-electrode model and it is found that these two kinds of data are basically the same. The performance of the proposed method is also validated by image reconstruction. The image quality with the inadequate data is poor in the reconstruction of the inclusion near the missing electrodes and the inclusion in the center. Comparatively, a significant improvement is observed when data is replenished by the proposed method. Moreover, the proposed method is tested under noise interference and size variation. The results show that reconstructed images are barely affected. This method offers a new approach for accurate brain imaging under limited measurements.

Cerebral autoregulation refers to the ability of cerebral vasculature to maintain stable blood flow by adjusting vascular resistance in response to changes in perfusion pressure. With advancing age, this regulatory capacity gradually declines, and its early, real-time, and dynamic monitoring holds potential as a promising approach for the prevention and treatment of cerebrovascular diseases. Given the absence of an established “gold standard” for assessing cerebral autoregulation, this study aimed to develop a non-invasive, continuous method for assessing cerebral autoregulation based on bioelectrical impedance technology. Using a wearable headband in combination with a Finapres device, blood pressure and cerebral blood flow were continuously monitored. A novel impedance recovery curve method was developed and, together with systemic blood pressure data, used to construct a hierarchical cerebral autoregulation assessment model via system identification. Moreover, the utility of this method in differentiating autoregulatory capacity across age groups (young adult and middle-aged) was assessed. The results demonstrated that the time constant (τREG), which characterizes the speed of cerebral blood flow recovery, differed significantly between the young adult and middle-aged groups (p < 0.001). These findings suggest the potential of τREG as a quantitative indicator for distinguishing cerebral autoregulatory function between healthy age cohorts.

Introduction:Electrical impedance tomography (EIT) enables non-invasive, continuous, bedside evaluation of ventilation/perfusion (V/Q) match. To avoid the presence of invasive monitoring for cardiac output in relative V/Q ratio calculation, we proposed a novel calibration method based on arterial blood pressure to optimize EIT V/Q match assessments. Methods:We involved 12 mechanically ventilated piglets in three experimental phases: baseline, pulmonary embolism, and atelectasis. After a thorough measurement of EIT signals, arterial blood pressure, cardiac output, and additional physiological parameters, EIT V/Q match was evaluated using existing area limited method (ALM), cardiac output calibrated method (COCM), and our proposed novel blood pressure calibrated method (BPCM). Finally, VD/VT and P/F ratio were calculated and correlated with V/Q match indicators derived from COCM and BPCM. Results:Arterial blood pressure waveform integration demonstrated strong correlation with cardiac output (R 2 = 0.80, p < 0.001), validating its utility for cardiac output estimation and V/Q match calibration. Both COCM and BPCM provided enhanced V/Q match region segmentation compared to ALM, yielding comprehensive diagnostic information with statistically significant differences across all three states (p < 0.05). COCM demonstrates a slightly higher correlation compared to BPCM (r = -0.63 vs. -0.52) between low ventilation index (LVI) and VD/VT, while BPCM demonstrates a slightly higher correlation compared to COCM (r = 0.49 vs. 0.44) between low perfusion index (LQI) and P/F ratio. Conclusion:This study described a novel calibration method for calculating corrected EIT-based V/Q match that utilized arterial blood pressure. Our method exhibited comparable capability in distinguishing V/Q mismatch areas compared to conventional cardiac output-based calibration techniques. With clinical data to establish a linear regression model, our method will ultimately enable us to calculate calibrated EIT V/Q match without cardiac output monitoring.

In electrical impedance tomography (EIT), the commonly used linear reconstruction algorithms are typically suitable for imaging small resistivity changes. However, in many applications of EIT, such as in imaging maximum ventilation of lung with EIT, the resistivity changes can be very large. In such cases, the linear algorithms have reduced accuracy and may affect the image interpretation. To address this issue, a novel iterative decomposition algorithm (IDA) is developed. In IDA, the large resistivity change target is decomposed into small changes so that the final solution can be obtained through multiple linear reconstructions, and the sensitivity matrix is iteratively updated based on the result of each linear reconstruction. To test the performances of IDA, both simulation and in-vivo experiments were conducted. The experimental results demonstrate that, in imaging large resistivity changes in lung ventilation, the traditional linear EIT algorithm caused non-negligible linear approximation errors (LAEs) and location errors (LEs). While for IDA, it could reduce LAEs and LEs by 13.4% and 11.6% respectively. The resistivity changes reconstructed by IDA had a better correlation with the lung volume changes. Therefore, IDA was verified as an efficient method for imaging large resistivity changes.

Electrical Impedance Tomography (EIT) is a noninvasive and real-time medical imaging technique, which means more potential applications for clinical diagnosis and treatments. However, EIT reconstruction problem is a highly nonlinear and ill-posed problem which will lead to poor reconstruction quality. Since neural networks have been proven to fit any nonlinear mapping theoretically, there are significant advantages of neural networks for EIT reconstruction. Convolutional neural networks (CNN), as one of the most famous neural networks, have powerful spatial feature extraction capabilities. Radial basis function (RBF) neural networks represent local approximators to nonlinear mapping, which has the significant advantage of short training time. In this paper, we proposed a two-part solver for the EIT reconstruction problem based on deep learning. RBF neural networks are used to solve the EIT reconstruction problem while CNN are used for feature extraction.

Cerebral hemorrhage is a serious cerebrovascular disease with high morbidity and high mortality, for which timely diagnosis and treatment are crucial. Electrical impedance tomography (EIT) is a functional imaging technique which is able to detect abnormal changes of electrical property of the brain tissue at the early stage of the disease. However, irregular multi-layer structure and different conductivity properties of each layer affect image reconstruction of the brain EIT, resulting in low reconstruction quality. To solve this problem, an image reconstruction method based on an improved densely-connected fully convolutional neural network is proposed in this paper. On the basis of constructing a three-layer cerebral model that approximates the real structure of the human head, the nonlinear mapping between the boundary voltage and the conductivity change is determined by network training, which avoids the error caused by the traditional sensitivity matrix method used for solving inverse problem. The proposed method is also evaluated under the conditions with or without noise, as well as with brain model change. The numerical simulation and phantom experimental results show that conductivity distribution of cerebral hemorrhage can be accurately reconstructed with the proposed method, providing a reliable basis for the diagnosis and treatment of cerebral hemorrhage. Also, it promotes the application of EIT in the diagnosis of brain diseases.

This paper investigates the variation of lung tissue dielectric properties with tidal volume under in vivo conditions to provide reliable and valid a priori information for techniques such as microwave imaging. In this study, the dielectric properties of the lung tissue of 30 rabbits were measured in vivo using the open-end coaxial probe method in the frequency band of 100 MHz to 1 GHz, and 6 different sets of tidal volumes (30, 40, 50, 60, 70, 80 mL) were set up to study the trends of the dielectric properties, and the data at 2 specific frequency points (433 and 915 MHz) were analyzed statistically. It was found that the dielectric coefficient and conductivity of lung tissue tended to decrease with increasing tidal volume in the frequency range of 100 MHz to 1 GHz, and the differences in the dielectric properties of lung tissue for the 6 groups of tidal volumes at 2 specific frequency points were statistically significant. This paper showed that the dielectric properties of lung tissue tend to vary non-linearly with increasing tidal volume. Based on this, more accurate biological tissue parameters can be provided for bioelectromagnetic imaging techniques such as microwave imaging, which could provide a scientific basis and experimental data support for the improvement of diagnostic methods and equipment for lung diseases.

Purpose In the brain imaging based on electrical impedance tomography, it is sometimes not able to attach 16 electrodes due to space restriction caused by craniotomy. As a result of this, the number of boundary measurements decreases, and spatial resolution of reconstructed conductivity distribution is reduced. The purpose of this study is to enhance reconstruction quality in cases of limited measurement. Design/methodology/approach A new data expansion method based on the shallow convolutional neural network is proposed. An eight-electrode model is built from which fewer boundary measurements can be obtained. To improve the imaging quality, shallow convolutional neural network is constructed which maps limited voltage data of the 8-electrode model to expanded voltage data of a quasi-16-electrode model. The predicted data is compared with the quasi-16-electrode data. Besides, image reconstruction based on L1 regularization method is conducted. Findings The results show that the predicted data generally coincides with the quasi-16-electrode data. It is found that images reconstructed with the data of eight-electrode model are the poorest. Nevertheless, imaging results when the limited data is expanded by the proposed method show large improvement, and there is a minor difference with the images recovered with the quasi-16-electrode data. Also, the impact of noise is studied, which shows that the proposed method is robust to noise. Originality/value To enhance reconstruction quality in the case of limited measurement, a new data expansion method based on the shallow convolutional neural network is proposed. Both simulation work and phantom experiments have demonstrated that high-quality images of cerebral hemorrhage and cerebral ischemia can be obtained when the limited measurement is expanded by the proposed method.

Electrical impedance tomography (EIT) is a potential technique for imaging intracranial hemorrhage. Generally, 16 electrodes are used in the brain EIT. In some situations, it may be not possible to attach so many electrodes. This reduces the number of measurement and causes poor reconstruction quality. Besides, the skull has poor conductivity which obstructs the injection of current and degrades the quality of the reconstructed image. Thus, inadequate boundary measurement caused by reduced number of electrodes is a great challenge in brain EIT. To address this problem, a novel approach based on convolutional neural network is proposed for inadequate data augmentation when only 8 electrodes can be equipped. The proposed method enables the establishment of a non-linear relationship between the inadequate voltage data and the desired augmented data. The voltage data augmented for the 8-electrode brain EIT is compared with the measured voltage of the brain EIT equipped with 16 electrodes. It is found that two groups of data are very similar. The effectiveness of the proposed approach is further confirmed by imaging hemorrhage in the brain tissue layer. Images reconstructed for various cases are compared with the results which use directly measured voltage. The qualitative evaluation as well as quantitative analysis demonstrates that the proposed approach greatly enhances the reconstruction quality in the case of inadequate measurement. In summary, the voltage data directly measured from the 8-electrode brain EIT can be well augmented with the proposed method.

A cerebral contrast-enhanced electrical impedance tomography perfusion method is developed for acute ischemic stroke during intravenous thrombolytic therapy. Several clinical contrast agents with stable impedance characteristics and high-conductivity contrast were screened experimentally as electrical impedance contrast agent candidates. The electrical impedance tomography perfusion method was tested on rabbits with focal cerebral infarction, and its capability for early detection was verified based on perfusion images. The experimental results showed that ioversol 350 performed significantly better as an electrical impedance contrast agent than other contrast agents (p < 0.01). Additionally, perfusion images of focal cerebral infarction in rabbits confirmed that the electrical impedance tomography perfusion method could accurately detect the location and area of different cerebral infarction lesions (p < 0.001). Therefore, the cerebral contrast-enhanced electrical impedance tomography perfusion method proposed herein combines traditional, dynamic continuous imaging with rapid detection and could be applied as an early, rapid-detection, auxiliary, bedside imaging method for patients after a suspected ischemic stroke in both prehospital and in-hospital settings.

Acute cerebral ischemia is commonly accompanied by cerebral edema. With brain electrical impedance tomography (EIT), it is possible to monitor conductivity variation during dehydration treatment of the cerebral edema. However, the scalp is also dehydrated when dehydrating brain tissue which exerts a large impact on boundary measurement. Consequently, reconstruction quality of cerebral edema is seriously affected. To reduce the influence of scalp dehydration, a voltage data compensation method based on a fully connected neural network (FCNN) is proposed in this work. With this method, the voltage measured when brain tissue and scalp are simultaneously dehydrated can be compensated. To evaluate the performance of the proposed method, reconstruction of simulated cerebral ischemia in a three-layer head model is conducted. Images reconstructed with the compensated voltage data are compared with the results when there is no compensation. Besides, quantitative comparison is performed. The results indicate that it is almost impossible to identify the simulated cerebral ischemia without voltage compensation. In contrast, cerebral ischemia can be much more accurately reconstructed when the measured voltage is compensated with the proposed method.

Electrical impedance tomography (EIT) has shown its potential in the monitoring of cerebral edema. It is found that scalp is also dehydrated when dehydration of brain tissue is conducted. Simultaneous dehydration of the brain tissue and scalp would greatly affect the image quality. To solve this problem, this work proposes a data compensation method which aims to reduce the influence of scalp dehydration on image reconstruction. Firstly, a rabbit head which is simulated by a three-layer elliptical model is established. The relationship between boundary measurement and dehydration degree in the case of simultaneous dehydration is investigated. Then a prior matrix representing variation of boundary voltage against degree of dehydration when only scalp is dehydrated is established. With the determination of dehydration degree and establishment of the prior information, it is able to reduce the impact of scalp dehydration on measured voltage data. To validate the performance of the proposed method, reconstruction of four different models when scalp and brain are simultaneously dehydrated is conducted. Image reconstruction is performed under noiseless condition and different noise level conditions. Also, reconstructed images under brain dehydration are given for comparison. The performance of the proposed method is also experimentally tested. The results show that the impact of scalp dehydration on image reconstruction can be effectively reduced with the proposed data compensation method.

分析了生物医学工程专业医学仪器原理与应用课程教学的难点,并从理论讲授、仿真演示、学员实践等方面阐述了多物理场仿真的教学过程,指出利用COMSOL软件进行多物理场仿真案例教学,加强了学员对知识的理解与掌握,提高了学员的学习热情,有助于培养学员建模思维、提升教学效果,能为培养具备设备研发能力的生物医学工程人才打下基础.

目的:为了抑制脑部电阻抗断层成像(electrical impedance tomography,EIT)在临床使用过程中因人为或自身体动造成的干扰,提出一种基于小波包分解的抑制方法.方法:首先分析正常与受体动干扰信号的频谱差异,并采用皮尔逊相关系数作为判断受干扰数据的依据.其次基于数据频谱特征,应用小波包分解与重建的方法对体动所造成的偏置性干扰进行抑制.最后选取5名健康志愿者进行临床试验,通过对比处理前后原始电压信号特征以及重构图像结果验证干扰抑制效果.结果:提出的方法可以有效地修正EIT原始电压信号等效时间采样中存在的体动干扰,改善了采样信号的频谱质量,提高了EIT成像效果.结论:提出的方法能初步实现对正常与干扰信号的识别,并对存在干扰的信号进行抑制,对脑部EIT的临床应用具有重要意义.

Electrical impedance tomography (EIT) has attracted research interests in imaging of brain function, detection of breast cancer, monitoring of lung function and other medical fields. For EIT, its image reconstruction is largely affected by variation of electrode placement arranged around the detected object. Note that electrode offset is unavoidable in some medical applications, leading to low-quality reconstructed images or even failure of reconstruction. In this paper, a novel method is proposed to improve reconstruction quality in case of electrode offset. By identifying which electrode moves, mismatched boundary voltage caused by movement of electrode is corrected. After mismatch correction, image reconstruction is conducted with total generalized variation regularization method. To demonstrate performance of the proposed method in suppressing impact of electrode offset on reconstruction, extensive work is carried out. Mismatch corrections for four models under a series of counterclockwise or clockwise movements of a certain electrode are studied. The reconstruction results show that the proposed method can effectively improve the image quality under electrode offset. Even in presence of larger electrode offset and noise, reconstructed images have been greatly improved. Quantitative evaluation is also conducted by calculating relative blur radius (RBR). It is shown that RBR values are much closer to 1 compared with results under mismatch conditions. Image reconstruction when two electrodes move is also investigated which further validate the performance of the proposed method. This method is potential for mismatch correction in some medical applications.

OBJECTIVE:Electrical impedance tomography (EIT) is a bedside tool for lung ventilation and perfusion assessment. However, the ability for long-term monitoring diminished due to interferences from clinical interventions and motion artifacts. The purpose of this study is to investigate the feasibility of the discrete wavelet transform (DWT) to detect and remove the common types of motion artifacts in thoracic EIT.METHODS:Baseline drifting, step-like and spike-like interferences were simulated to mimic three common types of motion artifacts. The discrete wavelet decomposition was employed to characterize those motion artifacts in different frequency levels with different wavelet coefficients, and those motion artifacts were then attenuated by suppressing the relevant wavelet coefficients. Further validation was conducted in two patients when motion artifacts were introduced through pulsating mattress and deliberate body movements. The db8 wavelet was used to decompose the contaminated signals into several sublevels.RESULTS:In the simulation study, it was shown that, after being processed by DWT, the signal consistency improved by 92.98% for baseline drifting, 97.83% for the step-like artifact, and 62.83% for the spike-like artifact; the signal similarity improved by 77.49% for baseline drifting, 73.47% for the step-like artifact, and 2.35% for the spike-like artifact. Results from patient data demonstrated the EIT image errors decreased by 89.24% (baseline drifting), 88.45% (step-like artifact), and 97.80% (spike-like artifact), respectively; the data correlations between EIT images without artifacts and the processed were all > 0.95.CONCLUSION:This study found that DWT is a universal and effective tool to detect and remove these motion artifacts.

Electrical impedance tomography (EIT) is an emerging technique for medical imaging. According to reconstructed images, the human disease can be identified. Note that the reconstruction of EIT is an ill-posed inverse problem. To solve this problem, various sensitivity-based methods have been proposed. Nevertheless, a sensitivity matrix is affected by a number of factors. Therefore, classification reflected by images is sometimes not very reliable. In this work, a novel method based on a residual neural network is proposed for stroke classification. It is realized by directly processing measured difference data from EIT. After training of the residual neural network, the proposed method is quantitatively evaluated when noise is exerted, shape deviation occurs, and conductivity varies in three different layers. Also, impacts of contact impedance and small-sized inclusion are studied. Furthermore, phantom experiments are conducted. Compared with fully connected neural networks and shallow convolution neural networks, the results demonstrate that the proposed method has a better performance in classification for various cases. This article offers an effective alternative for stroke classification in EIT.