Alistair Boyle : Publications and Presentations.

Publications

also via Google Scholar, Web of Science, ResearchGate and ORCID and peer-review stats at Publons.

2022

Publications

  • Comparison of machine learning classifiers for differentiating level and sport using movement data,
    Gwyneth B. Ross, Allison L. Clouthier, Alistair Boyle, Steven L. Fischer, Ryan B. Graham; 2022, Journal of Sports Sciences, 40 (19).
    Abstract: The purposes of this study were to determine if 1) recurrent neural networks designed for multivariate, time-series analyses outperform traditional linear and non-linear machine learning classifiers when classifying athletes based on competition level and sport played, and 2) athletes of different sports move differently during non-sport-specific movement screens. Optical-based kinematic data from 542 athletes were used as input data for nine different machine learning algorithms to classify athletes based on competition level and sport played. For the traditional machine learning classifiers, principal component analysis and feature selection were used to reduce the data dimensionality and to determine the best principal components to retain. Across tasks, recurrent neural networks and linear machine learning classifiers tended to outperform the non-linear machine learning classifiers. For all tasks, reservoir computing took the least amount of time to train. Across tasks, reservoir computing had one of the highest classification rates and took the least amount of time to train; however, interpreting the results is more difficult compared to linear classifiers. In addition, athletes were successfully classified based on sport suggesting that athletes competing in different sports move differently during non-sport specific movements. Therefore, movement assessment screens should incorporate sport-specific scoring criteria.

Conferences

  • Vesicoureteral Reflux Imaged in an Animal Model Using EIT
    Alistair Boyle, Hari Aggrawal, Sarah Loughney
    14th International Conference on Bioelectromagnetism, the 18th International Conference on Electrical Bio-Impedance, and the 22nd International Conference on Biomedical Applications of Electrical Impedance Tomography June 28-July 1, 2022, Seoul, South Korea.
    (Paper, Slides, Presentation),
    Abstract: A preclinical study was conducted in July 2021 to detect Vesicoureteral Reflux (VUR) non-invasively using Electrical Impedance Tomography (EIT). VUR involves the backflow of urine from the bladder to the ureters or kidneys and diagnosed predominantly in the paediatric population from birth to 5 years of age. To recreate VUR in three healthy pigs, stents were positioned at the junction between the bladder and ureters – the Ureterovesical Junctions (UVJ). A Foley catheter was introduced into the bladder via the urethra and used to inject solution mimicking the properties of urine. The stents opened the UVJ to allow urine to more freely travel from the bladder into the ureters and the kidneys during bladder infusion. The simulated VUR was confirmed using fluoroscopic imaging. VUR is a long-term condition, whereas this stenting procedure resulted in acute kidney reflux. VUR achieved using this method was approximately Grade II-III (2-3). The refluxed solution was detectable through regional conductivity changes reconstructed from EIT measurements and localised to the correct kidney. Conductivity changes were observable in 73% of cycles without motion (8 of 11) and 44% of cycles with motion (4 of 9).
  • Comparison of Machine Learning Classifiers for Differentiating Level and Sport Using Movement Data
    Gwyneth B. Ross, Allison L. Clouthier, Alistair Boyle, Steven L. Fischer, Ryan B. Graham
    North American Congress of Biomechanics August 21-25, 2022, Ottawa, Canada.
    (Final Draft),
    Abstract: Movement screens are used to identify aberrant movement patterns believed to increase risk of injury and/or impede performance. [...] The purpose of this study was two-fold: 1) to determine if RNNs designed for time-series analyses can outperform the previously used traditional classifiers at classifying athlete skill level, and 2) if athletes can be differentiated based on sport played, and if so, to identify which machine learning algorithm(s) performs the best.

2021

Publications

  • Geophysical ERT book chapter in
    Electrical Impedance Tomography: Methods, History and Applications, 2nd ed.,
    Alistair Boyle, Paul B. Wilkinson; CRC Press, 2021.
    (Final Draft),
    Abstract: In geophysics, electrical measurement techniques to estimate near-surface impedances were initially developed in the context of mineral prospecting by Conrad Schlumberger in 1911 and have been widely used in subsurface investigations ever since. Geoelectric imaging is used world-wide in industry, consultancy and academia, and is the subject of considerable ongoing research and development. Terminology in geophysics has evolved and what was initially referred to as "vertical electrical sounding" (a one-dimensional layered Earth problem) has been refined to Electrical Resistivity Tomography (ERT). Electrical Resistivity Imaging (ERI) is also used in some literature as an alternative to the use of ERT. Resistivity (units: Ωm), the reciprocal of conductivity (units: S/m), is generally the preferred unit when discussing geological phenomena.
    It is important to recognize that, mathematically, biomedical EIT and geophysical ERT solve the same equations; the Calderón Problem. Nonetheless, there are some important distinctions in geophysics for specific applications. In particular, ERT is usually performed at much lower frequencies (1 Hz to 10 kHz) and over greater distances (50–100 m electrode arrays are common), often on an open domain (the Earth’s surface) or in boreholes, and encounters resistivities which can vary over orders of magnitude across nearby regions. In biomedical EIT, time difference EIT is often preferred, but for geophysical settings, a static reconstruction of the actual resistivity distribution is often required. There are also monitoring applications in which, similar to EIT, changes in resistivity are of interest.
  • Development and Validation of a Deep Learning Algorithm and Open-Source Platform for the Automatic Labeling of Motion Capture Markers
    Allison L. Clouthier , Gwyneth B. Ross, Matthew P. Mavor, Isabel Coll , Alistair Boyle, Ryan B. Graham;
    2021, IEEE Access, 9 36444-36454.
    Abstract: Objective: The purpose of this work was to develop an open source deep learning-based algorithm for motion capture marker labelling that can be trained on measured or simulated marker trajectories. Methods: In the proposed algorithm, a deep neural network including recurrent layers is trained on measured or simulated marker trajectories. Labels are assigned to markers using the Hungarian algorithm and a predefined generic marker set is used to identify and correct mislabeled markers. The algorithm was first trained and tested on measured motion capture data. Then, the algorithm was trained on simulated trajectories and tested on data that included movements not contained in the simulated data. The ability to improve accuracy using transfer learning to update the neural network weights based on labelled motion capture data was assessed. The effect of occluded and extraneous markers on labelling accuracy was also examined. Results: Labelling accuracy was 99.6% when trained on measured data and 92.8% when trained on simulated trajectories, but could be improved to 98.8% through transfer learning. Missing or extraneous markers reduced labelling accuracy, but results were comparable to commercial software. Conclusion: The proposed labelling algorithm can be used to accurately label motion capture data in the presence of missing and extraneous markers and accuracy can be improved as data are collected, labelled, and added to the training set. Significance: The algorithm and user interface can reduce the time and manual effort required to label optical motion capture data, particularly for those with limited access to commercial software.

Conferences

Proceedings

2020

Publications

  • Beneficial Techniques for Spatio-Temporal Imaging in Electrical Impedance Tomography
    Alistair Boyle, Kirill Aristovich, Andy Adler; 2020, Physiol. Meas., 41 (6), 16.
    (Final Draft)
    Abstract: Objective: Electrical Impedance Tomography (EIT) typically reconstructs individual images from electrical voltage measurements at pairs of electrodes due to current driven through other electrode pairs on a body. EIT images have low spatial resolution, but excellent temporal resolution. There are four methods for integrating temporal data into an EIT reconstruction: filtering over measurements, filtering over images, combined spatial and temporal (spatio-temporal) regularization, and Kalman filtering. These spatio-temporal methods have not been directly compared, making it difficult to evaluate relative performance and choose an appropriate method for particular use cases.
    Approach: We (1) develop a common framework, (2) develop comparison metrics, (3) perform simulation and tank studies which directly compare algorithms, and (4) report on relative advantages of the different algorithms.
    Main Results: Temporal filtering is well understood, but often not considered as part of the imaging process despite a direct impact on image reconstruction quality. Spatio-temporal regularized techniques are not yet efficient but offer tantalizing advantages. Kalman filtering enables adaptive filtering for time-varying measurement/image noise at the cost of often over-regularized (sub-optimal) images which can now be understood in the same framework as the other techniques. Further research into efficient implementations of Gauss-Newton spatio-temporal regularization will allow temporal and spatial covariance to be explicitly defined for longer time series (n > 10 frames) where temporal regularization can be more effective. For the immediate analysis of temporally varying images, we recommend the use of adaptive (time-varying) temporal filtering of measurements followed by adaptive spatial regularization (hyperparameter selection) as the most computationally efficient and effective approach currently available.
    Significance: The analysis of variation within regions of an EIT image to extract physiological measures (functional imaging), has become an important EIT technique where temporal and spatial aspects of analysis are tightly integrated. This work gives guidance on available methods and suggests directions for future research.

Conferences

  • Machine Learning and Deep Neural Network Architectures for 3D Motion Capture Datasets
    Alistair Boyle, Gwyneth Ross, Ryan Graham;
    42nd Annual International Conferences of the IEEE Engineering in Medicine and Biology Society (EMBC2020), July 20-24, 2020, Montreal, Canada.
    (Final Draft),
    Abstract: Baseline performance for 3D joint centre trajectory classification using a number of traditional machine learning techniques are presented. This framework supports a robust comparison between classifier architectures over a 416 subject dataset of athletes (professional, college, and amateur) from five primary sports and six non-primary sports performing thirteen non-sport specific movements. A variety of deep neural networks specifically intended for time-series data are currently being evaluated.
    Clinical/sports relevance— Patient and athlete movement patterns can be measured by 3D motion capture and evaluated by systematically using machine learning. By providing a distributable “expert”, issues with inter- and intra-rater variability may be reduced. This work explores a variety of machine learning techniques to evaluate which methods are most appropriate for motion capture data.
  • Exploratory Analysis of Ultra-Short-Term Heart Rate Variability Features in Virtual Rehabilitation Sessions
    Roger Selzler, Andrew Smith, François Charih, Alistair Boyle, Janet Holly, Courtney Bridgewater, Markus Besemann, Dorothyann Curran, Adrian D. C. Chan, and James R. Green;
    15th IEEE International Symposium on Medical Measurements and Applications (MeMeA2020), June 1-3, 2020, Bari, Italy.
    Abstract: We are currently collecting multi-modal data from patients undergoing rehabilitation therapy using virtual reality for mild traumatic brain injury (mTBI), post-traumatic stress disorder (PTSD), and complex pain, with the goal of developing novel unobtrusive estimators of SAANS. In this study, we investigate whether heart rate variability (HRV) features extracted from electrocardiogram (ECG) measurements are correlated with clinical estimates of patient SAANS. Although previous studies have shown such correlation, the minimum amount of ECG data typically required for HRV feature extraction preclude their application in the present study; SAANS is expected to change quickly during therapy sessions and a rapid estimator is required. This paper investigates the use of ultra-short-term features extracted from ECG measurements for SAANS estimation with data collected during rehabilitation sessions with a Computer-Assisted Rehabilitation Environment (CAREN). A comparison between different scores and time excerpts were analysed using a single patient (n=1), over fifteen sessions, and sixty-seven activities. Preliminary results show that there was a significant difference between the AFTER and START excerpts (p = 0.045) for the meanNN feature, and between score “four” to scores “zero” (p = 0.003), “two” (p = 0.03), “three” (p = 0.007) and “five” (p = 0.03) for the stdHR feature.
  • In-Shoe Thin-film Pressure Sensor Validation for Gait Biomechanics (submitted, postponed/COVID-19)
    Alistair Boyle and Ryan Graham;
    21st Biennial Meeting of the Canadian Society for Biomechanics (CSB2020), Auguest 11-14, 2020, Montreal, Canada.
    Abstract: Thin-film pressure sensors are used in diagnosing and treating podiatric con- ditions and analyzing footwear performance. The sensors provide continuous information to support gait and foot function analysis outside of lab environ- ments. Tekscan is releasing a new product to update it’s F-scan line of sensors (F-scan-64 “Sensole”). These new sensors have much lighter electronics, fewer sensor cells (64 versus 955 cells), and sensors that match shoe sizes rather than being trimmed to fit. We evaluate the performance of these new sensors in re- lation to their older product (F-scan 3000E) and will validate performance in the lab against 3D motion capture and force plates.
    Both old and new sensors were inserted into a pair of low drop, tight-fitting running shoes. In one shoe the new sensor was placed over the old sensor. The order of the sensors was reversed in the second shoe. New sensors were used. The sensors were not taped into the shoes. Sensors were equilibrated and calibrated. The subject walked on a treadmill at 2.5 km/h: data were collected for 30 seconds, followed by 5 minutes of further walking, and then another 30 seconds of data capture. The procedure was repeated twice.
    Preliminary results show a reasonable agreement in the gait cycle and a peak pressure pattern over the foot regions which roughly agree (see figure). The new sensors consistently estimated greater heel strike and toe-off forces, and reduced mid-stance force. Changing the order of sensor insertion into the shoes, or performing the test with a single sensor, did not noticeably impact performance.
    Further data collection is underway using gold standard motion capture and force plates over a larger cohort. New calibration or equilibration procedures may be required for these new sensors.

2019

Publications

  • 4D Electrical Resistivity Tomography (ERT) for Aquifer Thermal Energy Storage Monitoring
    Nolwenn Lesparre, Robert Tanguy, Frédéric Nguyen, Alistair Boyle, Thomas Hermans; 2019, Geothermics, 77 (1) 368-382.
    (Final Draft),
    Abstract: In the context of aquifer thermal energy storage, we conducted a hydrogeophysical experiment emulating the functioning of a groundwater heat pump for heat storage into an aquifer. This experiment allowed the assessment of surface electrical resistivity tomography (ERT) ability to monitor the 3D development over time of the aquifer thermally affected zone. The resistivity images were converted into temperature. The images reliability was evaluated using synthetic tests and the temperature estimates were compared to direct temperature measurements. Results showed the capacity of surface ERT to characterize the thermal plume and to reveal the spatial variability of the aquifer hydraulic properties, not captured from borehole measurements. A simulation of the experiment was also performed using a groundwater flow and heat transport model calibrated with a larger set-up. Comparisons of the simulation with measurements highlighted the presence of smaller heterogeneities that strongly influenced the groundwater flow and heat transport.
  • Electrical Impedance Tomography book chapter in
    Wiley Encyclopedia of Electrical and Electronics Engineering,
    Andy Adler and Alistair Boyle; 2019, Wiley.
    (Final Draft),
    Abstract: Electrical impedance tomography (EIT) is a medical imaging technique that uses electrical stimulations and measurements at body‐surface electrodes. From these data, images of the distribution of conductivity within the body are calculated by solving an inverse problem. EIT has the advantage of producing high temporal resolution data, while being relatively low cost, noninvasive, small, and not using ionizing radiation. On the other hand, EIT has disadvantages in providing low spatial resolution and being sensitive to changes at the electrodes. EIT is currently being used clinically for monitoring of ventilated patients and is also being actively researched for applications such as cardiovascular flows and pressures, brain and nervous activity, cancer screening, and monitoring of gastrointestinal flows. EIT is similar to the electrical resistance tomography used in geophysical and process monitoring. This article reviews EIT from the point of view of its applications as well as image generation and interpretation.
  • Comparison of bolus- and filtering-based EIT measures of lung perfusion in an animal model,
    Symon Stowe, Alistair Boyle, Michaël Sage, Wendy See, Jean-Paul Praud, Étienne Fortin-Pellerin, Andy Adler; 2019, Physiological Measurement, 40 (5) 9pp.
    (Final Draft),
    Abstract: Two main functional imaging approaches have been used to measure regional lung perfusion using Electrical Impedance Tomography (EIT): venous injection of a hypertonic saline contrast agent and imaging of its passage through the heart and lungs, and digital filtering of heart-frequency impedance changes over sequences of EIT images. This paper systematically compares filtering-based perfusion estimates and bolus injection methods to determine to which degree they are related. Approach: EIT data was recorded on 7 mechanically ventilated newborn lambs in which ventilation distribution was varied through changes in posture between prone, supine, left- and right-lateral positions. Perfusion images were calculated using frequency filtering and ensemble averaging during both ventilation and apnoea time segments for each posture to compare against contrast agent-based methods using Jaccard distance score. Main Results: Using bolus-based EIT measures of lung perfusion as the reference frequency filtering techniques performed better than ensemble averaging and both techniques performed equally well across apnoea and ventilation data segments. Significance: Our results indicate the potential for use of filtering-based EIT measures of heart-frequency activity as a non-invasive proxy for contrast agent injection-based measures of lung perfusion.

Proceedings

Conferences

2018

Publications

  • Jointly reconstructing ground motion and resistivity for ERT-based slope stability monitoring
    Alistair Boyle, Paul B. Wilkinson, Jonathan E. Chambers, Philip I. Meldrum, Sebastian Uhlemann and Andy Adler; February 2018, Geophysical Journal International, 212 (2) 1167-1182.
    (Final Draft),
    Abstract: Electrical Resistivity Tomography (ERT) is a method where current is applied at electrodes on the boundary and voltages are measured at other electrodes. From the collected measurements, a resistivity distribution within the interior may be reconstructed. Movement of electrodes or incorrect placement of electrodes with respect to a simulation model can introduce significant resistivity artifacts into the reconstruction. In this work, we demonstrate a joint resistivity and electrode movement reconstruction algorithm within an iterative Gauss-Newton framework. Results show fewer resistivity artifacts and suggest that electrode movement may be reconstructed at the same time under certain conditions. A new 2.5D formulation for the electrode movement Jacobian was demonstrated to be equivalent to the perturbation method on a 3D model.

Proceedings

Conferences

Technical Reports

  • ERT for Northern Infrastructure Protection: Feasibility Study
    Alistair Boyle and Andy Adler; March 2018
    Abstract: Much critical infrastructure in Canada’s north is vulnerable to changes in due to various factors. These changes are exacerbated by subsiding ground in areas of melting permafrost. We have developed technology based on ERT (Electrical Resistance Tomography), which shows promise for low-cost, long term monitoring of changes in permafrost and the associated ground water movement. This report analyses and discussed the feasibility of this approach for detecting state-changes of permafrost and frozen ground from surface and borehole installations. Overall, our results indicate that detection of the melted liquid in the active layer above permafrost regions is well within the feasibility of ERT monitoring equipment. On the other hand, ERT imaging into the frozen permafrost is much more difficult.

Invited Talks

  • Embedded Systems for Impedance Imaging
    Alistair Boyle; February 2018, invited talk (research overview), Carleton University
    (Presentation),
    Abstract: Low frequency electrical impedance imaging can be used to monitor (and possibly predict) ground stability for railways, mine tailings piles, and remote infrastructure in a changing climatic environment. As an example, changes in permafrost layers in the far north have made it difficult to maintain and update civil infrastructure: roads, airfields, and buildings. Long-term, unsupervised monitoring and movement prediction are important for maintaining remote civil infrastructure where continuous human intervention is impractical and remote sensing alternatives, airborne or satellite-based, do not provide appropriate information. Embedded systems for long-term remote monitoring using impedance imaging may be used to manage slope stability and storage containment risks. These systems will enable active management of risks under changing environmental conditions and enable remedial measures to be undertaken before environmental damages, sometimes into the billions of dollars, befall civil infrastructure owners, rail and mining operators.
    Integrated hardware/software design of cost- and power-constrained remote systems results in highly integrated and application-specific real-time embedded systems which can take years to successfully deploy. Open hardware platforms, much like open software, enable the precipitation of collected wisdom, broad design review, rapid customization, and reduced time-to-market. Application of these open and integrated design principles will enable improved analysis of implementation trade-offs, and extend the addressable market and longevity of many highly specific instruments by enabling rapid variation and adaptation.
  • Electrical Impedance Tomography and Electrical Resistivity Tomography
    Alistair Boyle; March 2018, invited talk (graduate-level medical image processing course), Carleton University
    Lecture Notes (Part 1), Slides (Part 2)
    Abstract: An introductory lecture on Electrical Impedance Tomography (EIT), assuming a passing familiarity with Computed Tomography (CT) imaging. We make use of simple (matrix-based) linear algebra manipulations to arrive at our mathematical formulation.
    In part 1, we review tomographic reconstruction techniques (simple 2D CT), and see how to build a matrix algebra-based Gauss Newton solution, the Maximum Likelihood Estimate. We see how regularization is introduced into this formulation and the algorithmic relationship between CT and EIT.
    In part 2, we see a number of examples of equipment and applications of EIT/ERT to biomedical and geophysics fields. The algebra presented in Part 1 is extended to show reconstructions that incorporate other parameters such as electrode movement.

2017

Publications

  • Methods for Calculating the Electrode Position Jacobian for Impedance Imaging
    Alistair Boyle, Michael G. Crabb, Markus Jehl, William R. B. Lionheart and Andy Adler; January 2017, Physiol. Meas. 38 (3) 555-574.
    (Final Draft),
    Abstract: Electrical Impedance Tomography (EIT) or Electrical Resistivity Tomography (ERT) apply current and measure voltages at the boundary of a domain through electrodes. The movement or incorrect placement of electrodes may lead to modelling errors that result in significant reconstructed image artifacts. These errors may be accounted for by allowing for electrode position estimates in the model. Movement may be reconstructed through a first-order approximation, the electrode position Jacobian. A reconstruction that incorporates electrode position estimates and conductivity can significantly reduce image artifacts. Conversely, if electrode position is ignored it can be difficult to distinguish true conductivity changes from reconstruction artifacts which may increase the risk of a flawed interpretation. In this work, we aim to determine the fastest, most accurate approach for estimating the electrode position Jacobian. Four methods of calculating the electrode position Jacobian were evaluated on a homogeneous halfspace. Results show that Fréchet derivative and rank-one update methods are competitive in computational efficiency but achieve different solutions for certain values of contact impedance and mesh density.
  • Bioimpedance Spectroscopy Processing and Applications book chapter in
    Reference Module in Biomedical Sciences,
    Hershel Caytak, Alistair Boyle, Andy Adler and Miodrag Bolic; Elsevier, 2017.
    (Final Draft),
    Abstract: Bioimpedance spectroscopy (BIS) uses multifrequency impedance measurements of biological tissues to estimate clinically and experimentally relevant parameters. This article reviews the steps involved in measurement, data processing, and applications of BIS data, with an emphasis on managing data quality and sources of errors. Based on a description of error sources, caused by measurement configuration, hardware, and modeling, we describe BIS data denoising. Two classes of modeling, explanatory and descriptive, can be used to reduce data dimensionality to a set of parameters or features. Explanatory models consider the electrical properties of samples and involve fitting data to simplified equivalent electrical circuits. Descriptive models involve reduction of the data to a set of eigenvectors/values which can be studied independently of any assumed electrical characteristics of the sample. Techniques described include fitting and decomposition methods for extraction of explanatory and descriptive model parameters, respectively. Denoising techniques discussed include adjusting measurement configuration, corrective algorithms for removal of artifacts, and use of supervised machine learning for identification of features characteristic of noisy impedance spectra. The article concludes with a discussion of the use of classifiers for labeling BIS data in a range of applications including discrimination of healthy versus pathological tissues.
  • Electrical Impedance Tomography: Tissue Properties to Image Measures
    Andy Adler, Alistair Boyle; November 2017, IEEE Trans. Biomed. Eng. 64 (11) 2494-2504.
    (Final Draft),
    Abstract: Electrical Impedance Tomography (EIT) uses electrical stimulation and measurement at the body surface to image the electrical properties of internal tissues. It has the advantage of non-invasiveness and high temporal resolution but suffers from poor spatial resolution and sensitivity to electrode movement and contact quality. EIT can be useful to applications where there are conductive contrasts between tissues, fluids or gasses, such as imaging of cancerous or ischemic tissue or functional monitoring of breathing, blood flow, gastric motility and neural activity. The past decade has seen clinical application and commercial activity using EIT for ventilation monitoring. Interpretation of EIT-based measures is complex, and this review paper focuses on describing the image interpretation “pathway.” We review this pathway, from Tissue Electrical Properties, EIT Electrodes & Hardware, Sensitivity, Image Reconstruction, Image Processing to EIT Measures. The relationship is discussed between the clinically-relevant parameters and the reconstructed properties. An overview is given of areas of EIT application and of our perspectives for research and development.

Proceedings

Conferences

2016

Publications

  • Geophysical Applications of Electrical Impedance Tomography
    Alistair Boyle, Ph.D. Thesis, Carleton University, Canada, 2016.
    (Presentation & Thesis)
    Abstract: Impedance imaging is a technique where stimulus currents are applied through electrodes to a body or the ground and measurements of the potential at other electrodes are collected. The data, along with any available prior information, are used to reconstruct an image of the conductivity distribution throughout the interior which provides diagnostic, cost effective information upon which decisions can be based for a broad array of geophysics, biomedical and industrial applications. The same technique is known as (biomedical) Electrical Impedance Tomography (EIT) and (geophysics) Electrical Resistivity Tomography (ERT). New geophysical applications have arisen for the automated monitoring of slope stability risks for natural landslides, transport embankments and cuttings, mine tailings dams and piles, and remote infrastructure in changing climatic environments. When impedance imaging is used in challenging scenarios, image quality can suffer unless sources of data error and instability can be addressed. This work develops computational techniques to address the issue of data set stability under adverse measurement conditions and builds practical implementations that demonstrate the effectiveness of our approach. We seek to achieve images with fewer artifacts and better detectability through improved methods for addressing boundary movement which permit the use of this technology on unstable surfaces where the positions of electrodes can change over time. Processes are developed for evaluating the correctness of an implementation and the overall validity of reconstructed images. Results demonstrated by adapting well understood strategies show improved reconstruction quality for simulated and measured geophysics data sets.
  • Electrical resistivity imaging in transmission between surface and underground tunnel for fault characterization
    Nolwenn Lesparre, Alistair Boyle, Bartłomiej Grychtol, Justo Cabrera, J. Marteau and Andy Adler; May 2016, Journal of Applied Geophysics, 128 (1) 163--178.
    Abstract: Electrical resistivity images supply information on sub-surface structures and are classically performed to characterize faults geometry. Here we use the presence of a tunnel intersecting a regional fault to inject electrical currents between surface and the tunnel to improve the image resolution at depth. We apply an original methodology for defining the inversion parametrization based on pilot points to better deal with the heterogeneous sounding of the medium. An increased region of high spatial resolution is shown by analysis of point spread functions as well as inversion of synthetics. Such evaluations highlight the advantages of using transmission measurements by transferring a few electrodes from the main profile to increase the sounding depth. Based on the resulting image we propose a revised structure for the medium surrounding the Cernon fault supported by geological observations and muon flux measurements.

Conferences

  • Modelling with 2.5D Approximations
    Alistair Boyle, Andy Adler,
    17th Conference on Electrical Impedance Tomography, Stockholm, Sweden, June 19-23, 2016. (Presentation & Paper)
    Abstract: In EIT, the 2.5D approximation is a method of reducing a 3D forward modelling problem to a 2D problem. We show that (a) the 2D modelling errors can be important, particularly for half-space like configurations (breast and prostate imaging, for example), and (b) that due to stimulus pattern sensitivity, the finite limit in the z-direction was only relevant out to a dipole spacing beyond the electrodes, at which point truncation errors were negligible. Presented in this work, is a new 2.5D forward solver appropriate for use with the new EIDORS iterative Gauss-Newton solver. We show efficient implementations of adjoint and FEM system matrix-derivative methods.

2015

Conferences

2014

Conferences

  • Slope Stability Monitoring through Impedance Imaging
    Alistair Boyle, Paul Wilkinson, Jonathan Chambers, Nolwenn Lesparre and Andy Adler,
    15th Intl. Conf. on Biomedical Applications of Electrical Impedance Tomography, Gananoque, Canada, 24-26 April, 2014. (Presentation & Paper)
    Abstract: A technique for monitoring slope stability in a geological setting through impedance tomography is demonstrated. An iterative absolute Gauss-Newton solver simultaneously constructs estimates of the underground resistivity distribution and movement of the stimulation and measurement electrodes. The results are a step toward demonstrating that a cost effective and potentially predictive monitoring technology could be practical.
  • Monitoring Steam-Assisted Gravity Drainage (SAGD) with EIT
    Hervé Gagnon, Alistair Boyle, Michal Okoniewski and Andy Adler,
    15th Intl. Conf. on Biomedical Applications of Electrical Impedance Tomography, Gananoque, Canada, 24-26 April, 2014. (Poster & Paper)
    Abstract: Steam-assisted gravity drainage (SAGD) is a technique that has been developed to efficiently extract bitumen from deep reservoirs. We propose using electrical impedance tomography (EIT) for real-time monitoring of SAGD wells to maintain optimal operating conditions. Several electrode configurations along the pipelines and measurement strategies are presented and compared.
  • Open Electrical Impedance Tomography (OEIT) File Format
    Colin Jones, Bartłomiej Grychtol, Hervé Gagnon, Alistair Boyle, Chengbo He, Andy Adler and Pascal O. Gaggero,
    15th Intl. Conf. on Biomedical Applications of Electrical Impedance Tomography, Gananoque, Canada, 24-26 April, 2014. (Poster & Paper)
    Abstract: Electrical impedance tomography (EIT) creates tomographic images from surface electrical stimulation and measurement. Many research and commercial devices have been made, with correspondingly many data formats, which negatively impacts the ability to share data. To address this issue, we have developed the OEIT data format, an XML-based flexible container format for EIT data. We describe its features and structure.
  • Discontinuities detection in low permeability rocks using electrical resistivity imaging
    Nolwenn Lesparre, Justo Cabrera, Alistair Boyle, Bartłomiej Grychtol, Andy Adler
    The challenge of studying low permeability materials: Laboratory, in situ (field) and numerical methods workshop, University Cergy-Pontoise (UCP), Ile de France, 2-3 December, 2014 (Abstract)
    Abstract: In the context of nuclear waste storage, low permeability clays are investigated as potential geological barrier. Discontinuities in such formations might facilitate the radionuclide transport to the environment. The underground experimental platforms of Tournemire (Aveyron, France) and Mont Terri (Jura, Switzerland) present the opportunity to perform in-situ experiments to evaluate in particular the capacity of geophysical methods to detect and characterize the presence of discontinuities. Here we apply electrical resistivity imaging at different scales to detect variations of water saturation, degree of fracturation or presence of voids in the clay medium.

2012

Publications

  • Addressing the Computational Cost of Large EIT Solutions
    Alistair Boyle, Andrea Borsic and Andy Adler; April 2012, Physiol. Meas. 33 (5) 787--800.
    (Final Draft, supplement:profiling_eidors.m, mmwrite.m)
    Abstract: Electrical Impedance Tomography (EIT) is a soft field tomography modality based on the application of electric current to a body and measurement of voltages through electrodes at the boundary. The interior conductivity is reconstructed on a discrete representation of the domain using a FEM mesh and a parametrization of that domain. The reconstruction requires a sequence of numerically intensive calculations. There is strong interest in reducing the cost of these calculations.
    An improvement in the compute time for current problems would encourage further exploration of computationally challenging problems such as the incorporation of time series data, wide-spread adoption of three-dimensional simulations, and correlation of other modalities such as CT and ultrasound. Multicore processors offer an opportunity to reduce EIT computation times but may require some restructuring of the underlying algorithms to maximize the use of available resources.
    This work profiles two EIT software packages (EIDORS and NDRM) to experimentally determine where the computational costs arise in EIT as problems scale. Sparse matrix solvers, a key component for the FEM forward problem and sensitivity estimates in the inverse problem, are shown to take a considerable portion of the total compute time in these packages. A sparse matrix solver performance measurement tool, Meagre-Crowd, is developed to interface with a variety of solvers and compare their performance over a range of two- and three-dimensional problems of increasing node density. Results show that distributed sparse matrix solvers that operate on multiple cores are advantageous up to a limit that increases as the node density increases. We recommend a selection procedure to find a solver and hardware arrangement matched to the problem and provide guidance and tools to perform that selection.
  • Shape Deformation in Two-Dimensional Electrical Impedance Tomography
    Alistair Boyle, Andy Adler and William R. B. Lionheart; December 2012, IEEE Trans. Medical Imaging, 31 (12) 2185--2193. (Final Draft)
    Abstract: Electrical Impedance Tomography (EIT) uses measurements from surface electrodes to reconstruct an image of the conductivity of the contained medium. However, changes in measurements result from both changes in internal conductivity and changes in the shape of the medium relative to the electrode positions. Failure to account for shape changes results in a conductivity image with significant artifacts. Previous work to address shape changes in EIT has shown that in some cases boundary shape and electrode location can be uniquely determined for isotropic conductivities; however, for geometrically conformal changes, this is not possible. This prior work has shown that the shape change problem can be partially addressed. In this paper, we explore the limits of compensation for boundary movement in EIT, using three approaches: first, a theoretical model was developed to separate a deformation vector field into conformal and non-conformal components, from which the reconstruction limits may be determined; next, finite element models were used to simulate EIT measurements from a domain whose boundary has been deformed; finally, an experimental phantom was constructed from which boundary deformation measurements were acquired. Results, both in simulation and with experimental data, suggest that some electrode movement and boundary distortions can be reconstructed based on conductivity changes alone while reducing image artifacts in the process.

Conferences

  • Robust Stimulation and Measurement Patterns in Biomedical EIT
    Alistair Boyle, Yasin Mamatjan and Andy Adler,
    Workshop on 100 Years of Electrical Imaging, Mines ParisTech, Paris, France, 9-10 July, 2012. (Paper & Poster)
    Abstract: Distinguishability criteria incorporating stimulation and measurement patterns as well as an initial conductivity distribution over a specific domain were used in combination with linear programming. The outcomes were stimulation and measurement patterns that maximize the minimum distinguishability to give a robust experimental design. Specific applications for biomedical EIT were demonstrated. The framework is observed to be extensible to arbitrary domains, electrode configurations and geometries.

2011

Publications

  • Impact of Electrode Area, Contact Impedance and Boundary Shape on EIT Images
    Alistair Boyle and Andy Adler, June 2011, Physiol. Meas. 32 (7) 745--754. (Final Draft)
    Abstract: Electrical Impedance Tomography (EIT) measures the conductivity distribution within an object based on the current applied and voltage measured at surface electrodes. Thus, EIT images are sensitive to electrode properties (i.e. contact impedance, electrode area, and boundary shape under the electrode). While some of these electrode properties have been investigated individually, this paper investigates these properties and their interaction using Finite Element Method (FEM) simulations and the Complete Electrode Model (CEM). The ect of conformal deformations on image reconstruction when using the CEM was of specific interest. Observed artifacts were quantified using a measure that compared an ideal image to the reconstructed image, in this case a nonoise reconstruction that isolated the electrodes' ects. For electrode contact impedance and electrode area, uniform reductions to all electrodes resulted in ringing artifacts in the reconstructed images when the CEM was used, while parameter variations that were not correlated amongst electrodes resulted in artifacts distributed throughout the image. When the boundary shape changed under the electrode, as with non-symmetric conformal deformations, using the CEM resulted in structured distortions within the reconstructed image. Mean electrode contact impedance increases, independent of inter-electrode variation, did not result in artifacts in the reconstructed image.

Conferences

  • Scaling the EIT Problem
    Alistair Boyle, Andy Adler, Andrea Borsic,
    12th Intl. Conf. in Electrical Impedance Tomography, University of Bath, Bath, UK, 4-6 May, 2011. (Presentation & Paper [both corrected])
    Abstract: There are a number of interesting problems that could be tackled if the computing capacity of current EIT systems can be improved. We examined the performance of two such systems and found that a noticeable portion of the compute time is spent in finding the solution of sparse matrices. We developed and used a new sparse matrix testbench, Meagre-Crowd, to evaluate a selection of these sparse matrix solvers and found that there are definite performance gains available.

2010

Publications

  • The Effect of Boundary Shape Deformation on Two-Dimensional Electrical Impedance Tomography
    Alistair Boyle, MASc Thesis in Biomedical Engineering, Ottawa-Carleton Institute for Biomedical Engineering, Department of Systems and Computer Engineering, Carleton University, Ottawa, Canada, April 2010.
    (Presentation & Thesis)
    Abstract: Electrical Impedance Tomography (EIT) is a method of obtaining images of interior conductivity from electrode measurements on the boundary. Using EIT, this work focuses upon boundary movement in the two-dimensional reconstruction problem. Investigations were carried out using the tools of the Finite Element Method (FEM), inverse problem theory, and conformal transformations though simulation and tank tests. The limitations of boundary movement reconstruction algorithms that assume isotropic conductivities were explored. Initial testing of the boundary movement reconstruction technique with a deformable phantom showed that the detected boundary movement still had errors. Simulations showed that these errors are likely conformal and do not introduce artifacts into the image but do result in incorrect boundary approximations and subsequent deformations of the reconstructed image. A mathematical exploration of the conformal motions in EIT for two dimensions was carried out. Finally, it was found that, with conformal boundary movements, electrode models could cause various image artifacts.

Conferences

  • Electrode Models under Shape Deformation in Electrical Impedance Tomography
    Alistair Boyle, Andy Adler
    2010 Journal of Physics: Conference Series (J. Phys.: Conf. Ser.) 224 012051, doi: 10.1088/1742-6596/224/1/012051
    11th Conf. Electrical Impedance Tomography, University of Florida, Gainesville, FL, USA, 4-8 April, 2010. (Presentation & Draft Paper)
    Abstract: Electrical Impedance Tomography (EIT) applies current and measures the resulting voltage on the surface of a target. In biomedical applications, this current is applied, and voltage is measured through electrodes attached to the surface. Electrode models represent these connections in the reconstruction, but changes in the contact impedance or boundary relative to the electrode area can introduce artifacts. Using difference imaging, the effects of boundary deformation and contact impedance variation were investigated. The Complete Electrode Model (CEM) was found to be affected by conformal deformations. Contact impedance variability was found to be a significant source of artifacts in some cases.

Presentations

2009

Conferences

  • Artifacts due to Conformal Deformations in Electrical Impedance Tomography
    Alistair Boyle, William R.B. Lionheart, Andy Adler
    10th Conf. Electrical Impedance Tomography, University of Manchester, Manchester, UK, 15-18 June, 2009. (Presentation & Paper)
    Abstract: Artifacts in the images created using Electrical Impedance Tomography (EIT) due to movement of the boundary in difference imaging have been an issue, particularly in pulmonary EIT where chest movement due to breathing and posture change is a regular event.[1] With the recent development of algorithms to detect some types of boundary movement directly from the EIT measurements, it has become possible to correct for many of these boundary distortions by assuming an isotropic medium.[2][3] The further classification of boundary movement into two types, conformal movements and those that are not, provides the opportunity for further refinement of this algorithm.[4] In this paper, we discuss the conformal movements and their properties and show, through the governing conductivity equation for EIT, that conformal movement of an isotropic conductivity domain results in a new isotropic conductivity where the change in conductivity is directly related to the conformal movement.

Presentations

  • An Implementation of the Markov Chain Monte Carlo Technique
    Alistair Boyle, Directed studies course on Inverse Problems, Winter 2009.
    (Presentation 1: Discrete Ill-Posed and Rank-Deficient Problems, Presentation 2: Markov Chain Monte Carlo)
    Abstract: Markov Chain Monte Carlo was used in the solution of regularized Computed Tomography back projection. The effect of reducing the information used in reconstruction was investigated and regions of increased variance around some types of image artifacts were observed.

2008

Conferences

  • Evaluating Deformation Corrections in Electrical Impedance Tomography
    Alistair Boyle, William R.B. Lionheart, Camille Gomez-Laberge, Andy Adler
    9th Conf. Electrical Impedance Tomography, Dartmouth College, Hannover, NH, USA 16-18 June, 2008. (Presentation & Paper)
    Abstract: Electrical Impedance Tomography (EIT) uses the difference in measurements between surface electrodes to reconstruct an image of the conductivity of the contained medium. However, changes in measurements result from changes in internal conductivity and changes in the shape of the medium relative to the electrode positions. Failure to account for shape changes results in a conductivity image with significant artifacts. Previous work to address shape changes in EIT has shown that: a) theoretically, for an infinite number of electrodes, non-conformal changes in boundary shapes and electrode locations can be uniquely determined (Lionheart, 1998); and b) in some cases, conductivity and shape changes can be recovered using a combined image reconstruction model of both conductivity and shape changes (Soleimani et al, 2006). This work has shown that the shape change problem can be partially addressed. In this paper, we explore the limits of compensation for boundary movement in EIT, using three approaches: first, a theoretical model is developed to separate a deformation vector field into conformal and non-conformal components, from which the reconstruction limits may be determined; next, finite element models are constructed from which EIT measurements are simulated; finally, an experimental phantom is constructed using a deformable gasket and stainless steel electrodes in a saline medium, from which boundary deformation measurements are acquired.