Influence of a Structural Prior Mask on EIT Image Reconstruction.

Electrical Impedance Tomography (EIT) is a low-cost imaging method with promising results in visualizing ventilation distribution within the lungs. However, in clinical settings, the interpretability of EIT images is often limited by blurred anatomical alignment and reconstruction artifacts. Integrating structural priors into the EIT reconstruction process can enhance the interpretability of EIT images. In this contribution, we introduced a patient-specific structural prior mask into the EIT reconstruction process. Such prior mask ensures that only conductivity changes within the lung regions are reconstructed. With the aim to investigate the influence of the structural prior mask on the EIT images, we conducted numerical simulations in terms of four different ventilation status. EIT images were reconstructed with Gauss-Newton algorithm and discrete cosine transform-based EIT algorithm. We carried out quantitative analysis including the reconstruction error and figures of merit for the evaluation. The results show that the morphological structures of the lungs introduced by the prior mask are preserved in the EIT images, and the reconstruction artefacts are also limited. In conclusion, the incorporation of the structural prior mask enhances the interpretability of EIT images in clinical settings.Clinical relevance-The correct interpretation of an EIT image is crucial for a clinical diagnosis. This research demonstrates that a structural prior mask might have the potential to improve the interpretability of an EIT image, which facilitates the clinicians with a better understanding of EIT results.

Sensor Selection for Tidal Volume Determination via Linear Regression-Impact of Lasso versus Ridge Regression.

The measurement of respiratory volume based on upper body movements by means of a smart shirt is increasingly requested in medical applications. This research used upper body surface motions obtained by a motion capture system, and two regression methods to determine the optimal selection and placement of sensors on a smart shirt to recover respiratory parameters from benchmark spirometry values. The results of the two regression methods (Ridge regression and the least absolute shrinkage and selection operator (Lasso)) were compared. This work shows that the Lasso method offers advantages compared to the Ridge regression, as it provides sparse solutions and is more robust to outliers. However, both methods can be used in this application since they lead to a similar sensor subset with lower computational demand (from exponential effort for full exhaustive search down to the order of O (n2)). A smart shirt for respiratory volume estimation could replace spirometry in some cases and would allow for a more convenient measurement of respiratory parameters in home care or hospital settings.

P-CSEM: An Attention Module for Improved Laparoscopic Surgical Tool Detection.

Minimal invasive surgery, more specifically laparoscopic surgery, is an active topic in the field of research. The collaboration between surgeons and new technologies aims to improve operation procedures as well as to ensure the safety of patients. An integral part of operating rooms modernization is the real-time communication between the surgeon and the data gathered using the numerous devices during surgery. A fundamental tool that can aid surgeons during laparoscopic surgery is the recognition of the different phases during an operation. Current research has shown a correlation between the surgical tools utilized and the present phase of surgery. To this end, a robust surgical tool classifier is desired for optimal performance. In this paper, a deep learning framework embedded with a custom attention module, the P-CSEM, has been proposed to refine the spatial features for surgical tool classification in laparoscopic surgery videos. This approach utilizes convolutional neural networks (CNNs) integrated with P-CSEM attention modules at different levels of the architecture for improved feature refinement. The model was trained and tested on the popular, publicly available Cholec80 database. Results showed that the attention integrated model achieved a mean average precision of 93.14%, and visualizations revealed the ability of the model to adhere more towards features of tool relevance. The proposed approach displays the benefits of integrating attention modules into surgical tool classification models for a more robust and precise detection.

Effect of a Patient-Specific Structural Prior Mask on Electrical Impedance Tomography Image Reconstructions.

Electrical Impedance Tomography (EIT) is a low-cost imaging method which reconstructs two-dimensional cross-sectional images, visualising the impedance change within the thorax. However, the reconstruction of an EIT image is an ill-posed inverse problem. In addition, blurring, anatomical alignment, and reconstruction artefacts can hinder the interpretation of EIT images. In this contribution, we introduce a patient-specific structural prior mask into the EIT reconstruction process, with the aim of improving image interpretability. Such a prior mask ensures that only conductivity changes within the lung regions are reconstructed. To evaluate the influence of the introduced structural prior mask, we conducted numerical simulations with two scopes in terms of their different ventilation statuses and varying atelectasis scales. Quantitative analysis, including the reconstruction error and figures of merit, was applied in the evaluation procedure. The results show that the morphological structures of the lungs introduced by the mask are preserved in the EIT reconstructions and the reconstruction artefacts are decreased, reducing the reconstruction error by 25.9% and 17.7%, respectively, in the two EIT algorithms included in this contribution. The use of the structural prior mask conclusively improves the interpretability of the EIT images, which could facilitate better diagnosis and decision-making in clinical settings.

Structural priors represented by discrete cosine transform improve EIT functional imaging.

Structural prior information can improve electrical impedance tomography (EIT) reconstruction. In this contribution, we introduce a discrete cosine transformation-based (DCT-based) EIT reconstruction algorithm to demonstrate a way to incorporate the structural prior with the EIT reconstruction process. Structural prior information is obtained from other available imaging methods, e.g., thorax-CT. The DCT-based approach creates a functional EIT image of regional lung ventilation while preserving the introduced structural information. This leads to an easier interpretation in clinical settings while maintaining the advantages of EIT in terms of bedside monitoring during mechanical ventilation. Structural priors introduced in the DCT-based approach are of two categories in terms of different levels of information included: a contour prior only differentiates lung and non-lung region, while a detail prior includes information, such as atelectasis, within the lung area. To demonstrate the increased interpretability of the EIT image through structural prior in the DCT-based approach, the DCT-based reconstructions were compared with reconstructions from a widely applied one-step Gauss-Newton solver with background prior and from the advanced GREIT algorithm. The comparisons were conducted both on simulation data and retrospective patient data. In the simulation, we used two sets of forward models to simulate different lung conditions. A contour prior and a detail prior were derived from simulation ground truth. With these two structural priors, the reconstructions from the DCT-based approach were compared with the reconstructions from both the one-step Gauss-Newton solver and the GREIT. The difference between the reconstructions and the simulation ground truth is calculated by the ℓ2-norm image difference. In retrospective patient data analysis, datasets from six lung disease patients were included. For each patient, a detail prior was derived from the patient's CT, respectively. The detail prior was used for the reconstructions using the DCT-based approach, which was compared with the reconstructions from the GREIT. The reconstructions from the DCT-based approach are more comprehensive and interpretable in terms of preserving the structure specified by the priors, both in simulation and retrospective patient data analysis. In simulation analysis, the ℓ2-norm image difference of the DCT-based approach with a contour prior decreased on average by 34% from GREIT and 49% from the Gauss-Newton solver with background prior; for reconstructions of the DCT-based approach with detail prior, on average the ℓ2-norm image difference is 53% less than GREIT and 63% less than the reconstruction with background prior. In retrospective patient data analysis, the reconstructions from both the DCT-based approach and GREIT can indicate the current patient status, but the DCT-based approach yields more interpretable results. However, it is worth noting that the preserved structure in the DCT-based approach is derived from another imaging method, not from the EIT measurement. If the structural prior is outdated or wrong, the result might be misleadingly interpreted, which induces false clinical conclusions. Further research in terms of evaluating the validity of the structural prior and detecting the outdated prior is necessary.

Characterisation and Quantification of Upper Body Surface Motions for Tidal Volume Determination in Lung-Healthy Individuals.

Measurement of accurate tidal volumes based on respiration-induced surface movements of the upper body would be valuable in clinical and sports monitoring applications, but most current methods lack the precision, ease of use, or cost effectiveness required for wide-scale uptake. In this paper, the theoretical ability of different sensors, such as inertial measurement units, strain gauges, or circumference measurement devices to determine tidal volumes were investigated, scrutinised and evaluated. Sixteen subjects performed different breathing patterns of different tidal volumes, while using a motion capture system to record surface motions and a spirometer as a reference to obtain tidal volumes. Subsequently, the motion-capture data were used to determine upper-body circumferences, tilt angles, distance changes, movements and accelerations-such data could potentially be measured using optical encoders, inertial measurement units, or strain gauges. From these parameters, the measurement range and correlation with the volume signal of the spirometer were determined. The highest correlations were found between the spirometer volume and upper body circumferences; surface deflection was also well correlated, while accelerations carried minor respiratory information. The ranges of thorax motion parameters measurable with common sensors and the values and correlations to respiratory volume are presented. This article thus provides a novel tool for sensor selection for a smart shirt analysis of respiration.

Laparoscopic Video Analysis Using Temporal, Attention, and Multi-Feature Fusion Based-Approaches.

Adapting intelligent context-aware systems (CAS) to future operating rooms (OR) aims to improve situational awareness and provide surgical decision support systems to medical teams. CAS analyzes data streams from available devices during surgery and communicates real-time knowledge to clinicians. Indeed, recent advances in computer vision and machine learning, particularly deep learning, paved the way for extensive research to develop CAS. In this work, a deep learning approach for analyzing laparoscopic videos for surgical phase recognition, tool classification, and weakly-supervised tool localization in laparoscopic videos was proposed. The ResNet-50 convolutional neural network (CNN) architecture was adapted by adding attention modules and fusing features from multiple stages to generate better-focused, generalized, and well-representative features. Then, a multi-map convolutional layer followed by tool-wise and spatial pooling operations was utilized to perform tool localization and generate tool presence confidences. Finally, the long short-term memory (LSTM) network was employed to model temporal information and perform tool classification and phase recognition. The proposed approach was evaluated on the Cholec80 dataset. The experimental results (i.e., 88.5% and 89.0% mean precision and recall for phase recognition, respectively, 95.6% mean average precision for tool presence detection, and a 70.1% F1-score for tool localization) demonstrated the ability of the model to learn discriminative features for all tasks. The performances revealed the importance of integrating attention modules and multi-stage feature fusion for more robust and precise detection of surgical phases and tools.

Higher order total variation regularization for EIT reconstruction.

Electrical impedance tomography (EIT) attempts to reveal the conductivity distribution of a domain based on the electrical boundary condition. This is an ill-posed inverse problem; its solution is very unstable. Total variation (TV) regularization is one of the techniques commonly employed to stabilize reconstructions. However, it is well known that TV regularization induces staircase effects, which are not realistic in clinical applications. To reduce such artifacts, modified TV regularization terms considering a higher order differential operator were developed in several previous studies. One of them is called total generalized variation (TGV) regularization. TGV regularization has been successively applied in image processing in a regular grid context. In this study, we adapted TGV regularization to the finite element model (FEM) framework for EIT reconstruction. Reconstructions using simulation and clinical data were performed. First results indicate that, in comparison to TV regularization, TGV regularization promotes more realistic images. Graphical abstract Reconstructed conductivity changes located on selected vertical lines. For each of the reconstructed images as well as the ground truth image, conductivity changes located along the selected left and right vertical lines are plotted. In these plots, the notation GT in the legend stands for ground truth, TV stands for total variation method, and TGV stands for total generalized variation method. Reconstructed conductivity distributions from the GREIT algorithm are also demonstrated.

EIT Imaging Regularization Based on Spectral Graph Wavelets.

The objective of electrical impedance tomographic reconstruction is to identify the distribution of tissue conductivity from electrical boundary conditions. This is an ill-posed inverse problem usually solved under the finite-element method framework. In previous studies, standard sparse regularization was used for difference electrical impedance tomography to achieve a sparse solution. However, regarding elementwise sparsity, standard sparse regularization interferes with the smoothness of conductivity distribution between neighboring elements and is sensitive to noise. As an effect, the reconstructed images are spiky and depict a lack of smoothness. Such unexpected artifacts are not realistic and may lead to misinterpretation in clinical applications. To eliminate such artifacts, we present a novel sparse regularization method that uses spectral graph wavelet transforms. Single-scale or multiscale graph wavelet transforms are employed to introduce local smoothness on different scales into the reconstructed images. The proposed approach relies on viewing finite-element meshes as undirected graphs and applying wavelet transforms derived from spectral graph theory. Reconstruction results from simulations, a phantom experiment, and patient data suggest that our algorithm is more robust to noise and produces more reliable images.

EIT based pulsatile impedance monitoring during spontaneous breathing in cystic fibrosis.

OBJECTIVE: Evaluating the lung function in patients with obstructive lung disease by electrical impedance tomography (EIT) usually requires breathing maneuvers containing deep inspirations and forced expirations. Since these maneuvers strongly depend on the patient's co-operation and health status, normal tidal breathing was investigated in an attempt to develop continuous maneuver-free measurements. APPROACH: Ventilation related and pulsatile impedance changes were systematically analyzed during normal tidal breathing in 12 cystic fibrosis (CF) patients and 12 lung-healthy controls (HL). Tidal breaths were subdivided into three inspiratory (In1, In2, In3) and three expiratory (Ex1, Ex2, Ex3) sections of the same amplitude of global impedance change. Maximal changes of the ventilation and the pulsatile impedance signal occurring during these sections were determined (▵I V and ▵I P). Differences in ▵I V and ▵I P among sections were ascertained in relation to the first inspiratory section. In addition, ▵I V/▵I P was calculated for each section. MAIN RESULTS: Medians of changes in ▵I V were <0.05% in all sections for both subject groups. Both groups showed a similar pattern of ▵I P changes during tidal breathing. Changes in ▵I P first decreased during inspiration (In2), then increased towards the end of inspiration (In3) and reached a maximum at the beginning of expiration (Ex1). During the last two sections of expiration (Ex2, Ex3) ▵I P changes decreased. The CF patients showed higher variations in ▵I P changes compared to the controls (CF: -426.5%, HL: -158.1%, coefficient of variation). Furthermore, ▵I V/▵I P significantly differed between expiratory sections for the CF patients (Ex1-Ex2, p < 0.01; Ex1-Ex3, p < 0.001; Ex2-Ex3, p < 0.05), but not for the controls. No significant differences in ▵I V/▵I P between inspiratory sections were determined for both groups. SIGNIFICANCE: Differences in ▵I P changes and in ▵I V/▵I P between both subject groups were speculated to be caused by higher breathing efforts of the CF patients due to airway obstruction leading to higher intrathoracic pressures, and thus to greater changes in lung perfusion.

Lobe based image reconstruction in Electrical Impedance Tomography.

PURPOSE: Electrical Impedance Tomography (EIT) is an imaging modality used to generate two-dimensional cross-sectional images representing impedance change in the thorax. The impedance of lung tissue changes with change in air content of the lungs; hence, EIT can be used to examine regional lung ventilation in patients with abnormal lungs. In lung EIT, electrodes are attached around the circumference of the thorax to inject small alternating currents and measure resulting voltages. In contrast to X-ray computed tomography (CT), EIT images do not depict a thorax slice of well defined thickness, but instead visualize a lens-shaped region around the electrode plane, which results from diffuse current propagation in the thorax. Usually, this is considered a drawback, since image interpretation is impeded if 'off-plane' conductivity changes are projected onto the reconstructed two-dimensional image. In this paper we describe an approach that takes advantage of current propagation below and above the electrode plane. The approach enables estimation of the individual conductivity change in each lung lobe from boundary voltage measurements. This could be used to monitor disease progression in patients with obstructive lung diseases, such as chronic obstructive pulmonary disease (COPD) or cystic fibrosis (CF) and to obtain a more comprehensive insight into the pathophysiology of the lung. METHODS: Electrode voltages resulting from different conductivities in each lung lobe were simulated utilizing a realistic 3D finite element model (FEM) of the human thorax and the lungs. Overall 200 different patterns of conductivity change were simulated. A 'lobe reconstruction' algorithm was developed, applying patient-specific anatomical information in the reconstruction process. A standard EIT image reconstruction algorithm and the proposed 'lobe reconstruction' algorithm were used to estimate conductivity change in the lobes. The agreement between simulated and reconstructed conductivity change in particular lobes were compared using Bland-Altman plots, correlation plots and linear regression. To test the applicability of the approach in a realistic scenario, EIT measurements of a patient suffering from cystic fibrosis (CF) were carried out. RESULTS: Conductivity changes in each lobe generate specific patterns of voltage change. These can be used to estimate the conductivity change in lobes from measured boundary voltage. The correlation coefficient between simulated and reconstructed conductivity change in particular lobes is r > 0.89 for all lobes. Unknown position of the electrode plane leads to over- or underestimation of reconstructed conductivity change. Slight mismatches (± 5% of the forward model height) between the actual position of the electrode plane and the position used in the reconstruction model lead to regression coefficients of 0.7 to 1.3 between simulated and reconstructed conductivity change in the lobes. CONCLUSION: The presented approach enhances common reconstruction methods by providing information about anatomically assignable units and thus facilitates image interpretation, since impedance change and thus ventilation of each lobe is directly determined in the reconstructions.

Multi-layer ventilation inhomogeneity in cystic fibrosis.

Differences in regional lung function between the 3rd and 5th intercostal space (ICS) were explored in 10 cystic fibrosis (CF) patients and compared to 10 lung-healthy controls by electrical impedance tomography (EIT). Regional ratios of impedance changes corresponding to the maximal volume of air exhaled within the first second of a forced expiration (ΔIFEV1) and the forced vital capacity (ΔIFVC) were determined. Regional airway obstruction and ventilation inhomogeneity were assessed by the frequency distribution of these ratios (ΔIFEV1/ΔIFVC) and an inhomogeneity index (GITI). The mean of the frequency distribution of ΔIFEV1/ΔIFVC and the GITI in both thorax planes were significantly different between CF patients and controls (p<0.001). CF patients exhibited a significantly lower mean of ΔIFEV1/ΔIFVC frequency distribution (p<0.05) and a significantly higher degree of ventilation inhomogeneity (p<0.01) in the 3rd ICS compared to the 5th ICS. Results indicated that EIT measurements at more cranial thorax planes may benefit the early diagnosis in CF.

Sparse regularization for EIT reconstruction incorporating structural information derived from medical imaging.

Electrical impedance tomography (EIT) reconstructs the conductivity distribution of a domain using electrical data on its boundary. This is an ill-posed inverse problem usually solved on a finite element mesh. For this article, a special regularization method incorporating structural information of the targeted domain is proposed and evaluated. Structural information was obtained either from computed tomography images or from preliminary EIT reconstructions by a modified k-means clustering. The proposed regularization method integrates this structural information into the reconstruction as a soft constraint preferring sparsity in group level. A first evaluation with Monte Carlo simulations indicated that the proposed solver is more robust to noise and the resulting images show fewer artifacts. This finding is supported by real data analysis. The structure based regularization has the potential to balance structural a priori information with data driven reconstruction. It is robust to noise, reduces artifacts and produces images that reflect anatomy and are thus easier to interpret for physicians.

Structural-functional lung imaging using a combined CT-EIT and a Discrete Cosine Transformation reconstruction method.

Lung EIT is a functional imaging method that utilizes electrical currents to reconstruct images of conductivity changes inside the thorax. This technique is radiation free and applicable at the bedside, but lacks of spatial resolution compared to morphological imaging methods such as X-ray computed tomography (CT). In this article we describe an approach for EIT image reconstruction using morphologic information obtained from other structural imaging modalities. This leads to recon- structed images of lung ventilation that can easily be superimposed with structural CT or MRI images, which facilitates image interpretation. The approach is based on a Discrete Cosine Transformation (DCT) of an image of the considered transversal thorax slice. The use of DCT enables reduction of the dimensionality of the reconstruction and ensures that only conductivity changes of the lungs are reconstructed and displayed. The DCT based approach is well suited to fuse morphological image information with functional lung imaging at low computational costs. Results on simulated data indicate that this approach preserves the morphological structures of the lungs and avoids blurring of the solution. Images from patient measurements reveal the capabilities of the method and demonstrate benefits in possible applications.

Electrical impedance tomography: functional lung imaging on its way to clinical practice?

Electrical impedance tomography (EIT) has the potential to become a bedside tool for monitoring and guiding ventilator therapy as well as tracking the development of chronic lung diseases. This review article summarizes recent publications (from 2011) dealing with the applications of pulmonary EIT. Original papers on EIT lung imaging in clinical settings are analyzed and divided into several categories according to the lung pathology of the study subjects. Studies on children and infants are presented separately from studies on adult patients. Information on the study objectives and main results, the number of studied patients, the performed ventilatory maneuvers or interventions and the analyzed EIT information is given. Limitations that hinder EIT to become a routinely used tool in a clinical setting are also discussed.

Positioning of electrode plane systematically influences EIT imaging.

Up to now, the impact of electrode positioning on electrical impedance tomography (EIT) had not been systematically analyzed due to the lack of a reference method. The aim of the study was to determine the impact of electrode positioning on EIT imaging in spontaneously breathing subjects at different ventilation levels with our novel lung function measurement setup combining EIT and body plethysmography. EIT measurements were conducted in three transverse planes between the 3rd and 4th intercostal space (ICS), at the 5th ICS and between the 6th and 7th ICS (named as cranial, middle and caudal) on 12 healthy subjects. Pulmonary function tests were performed simultaneously by body plethysmography to determine functional residual capacity (FRC), vital capacity (VC), tidal volume (VT), expiratory reserve volume (ERV), and inspiratory reserve volume (IRV). Ratios of impedance changes and body plethysmographic volumes were calculated for every thorax plane (ΔIERV/ERV, ΔIVT/VT and ΔIIRV/IRV). In all measurements of a subject, FRC values and VC values differed ≤5%, which confirmed that subjects were breathing at comparable end-expiratory levels and with similar efforts. In the cranial thorax plane the normalized ΔIERV/ERV ratio in all subjects was significantly higher than the normalized ΔIIRV/IRV ratio whereas the opposite was found in the caudal chest plane. No significant difference between the two normalized ratios was found in the middle thoracic plane. Depending on electrode positioning, impedance to volume ratios may either increase or decrease in the same lung condition, which may lead to opposite clinical decisions.