Tidal volume monitoring by a set of tetrapolar impedance measurements selected from the 16-electrodes arrangement used in electrical impedance tomography (EIT) technique. Calibration equations in a group of healthy males

High-resolution conductivity reconstruction by electrical impedance tomography using structure-aware hybrid-fusion learning

BACKGROUND

Electrical impedance tomography (EIT) has gained considerable attention in the medical field for the diagnosis of lung-related diseases, owing to its non-invasive and real-time characteristics. However, due to the ill-posedness and underdetermined nature of the inverse problem in EIT, suboptimal reconstruction performance and reduced robustness against the measurement noise and modeling errors are common issues.

OBJECTIVES

This study aims to mine the deep feature information from measurement voltages, acquired from the EIT sensor, to reconstruct the high-resolution conductivity distribution and enhance the robustness against the measurement noise and modeling errors using the deep learning method.

METHODS

A novel data-driven method named the structure-aware hybrid-fusion learning (SA-HFL) is proposed. SA-HFL is composed of three main components: a segmentation branch, a conductivity reconstruction branch, and a feature fusion module. These branches work in tandem to extract different feature information from the measurement voltage, which is then fused to reconstruct the conductivity distribution. The unique aspect of this network is its ability to utilize different features extracted from various branches to accomplish reconstruction objectives. To supervise the training of the network, we generated regular-shaped and lung-shaped EIT datasets through numerical calculations.

RESULTS

The simulations and three experiments demonstrate that the proposed SA-HFL exhibits superior performance in qualitative and quantitative analyses, compared with five cutting-edge deep learning networks and the optical image-guided group sparsity (IGGS) method. The evaluation metrics, relative error (RE), mean structural similarity index (MSSIM), and peak signal-to-noise ratio (PSNR), are improved by implementing the SA-HFL method. For the regular-shaped dataset, the values are 0.119 (RE), 0.9882 (MSSIM), and 31.03 (PSNR). For the lung-shaped dataset, the values are 0.257 (RE), 0.9151 (MSSIM), and 18.67 (PSNR). Furthermore, the proposed network can be executed with appropriate parameters and efficient floating-point operations per second (FLOPs), concerning network complexity and inference speed.

CONCLUSIONS

The reconstruction results indicate that fusing feature information from different branches enhances the accuracy of conductivity reconstruction in the EIT inverse problem. Moreover, the study shows that fusing different modalities of information to reconstruct the EIT conductivity distribution may be a future development direction.

Real-Time Measurements of Relative Tidal Volume and Stroke Volume Using Electrical Impedance Tomography with Spatial Filters: A Feasibility Study in a Swine Model Under Normal and Reduced Ventilation

Continuous monitoring of both hemodynamic and respiratory parameters would be beneficial to patients, e.g., those in intensive care unit. The objective of this exploratory animal study was to test the feasibility of simultaneous measurements of relative tidal volume (rTV) and relative stroke volume (rSV) using an electrical impedance tomography (EIT) device equipped with a new real-time source separation algorithm implemented as two spatial filters. Five pigs were anesthetized and mechanically ventilated. The supplied tidal volume from a mechanical ventilator was reduced to 70, 50 and 30% from the 100% normal volume to simulate hypoventilation. The respiratory volume signal and cardiac volume signal were generated by applying the spatial filters to the acquired EIT data, from which values of rTV and rSV were extracted. The measured rTV values were compared with the TV values from the mechanical ventilator using the four-quadrant concordance analysis method. For changes in TV, the concordance rate in each animal ranged from 81.8% to 100%, while it was 92.5% when the data from all five animals were pooled together. When the measured rTV values for each animal were scaled to the absolute TV_EIT values in mL using the TV_Vent data from the mechanical ventilator, the smallest 95% limits of agreement (LoA) were - 6.04 and 7.44 mL for the 70% ventilation level, and the largest 95% LoA were - 18.1 and 19.4 mL for the 50% ventilation level. The percentage error between TV_EIT and TV_Vent was 10.3%. Although similar statistical analyses on rSV data could not be performed due to limited intra-animal variability, changes in rSV values measured by the EIT device were comparable to those measured by an invasive hemodynamic monitor. In this animal study, we were able to demonstrate the feasibility of an EIT device for noninvasive and simultaneous measurements of rTV and rSV in real time. However, the performance of the real-time source separation method needs to be further validated on animals and human subjects, particularly over a wide range of SV values. Future clinical studies are needed to assess the potential usefulness of the new method in dynamic cardiopulmonary monitoring and explore other clinical applications.