Keeley Edwards

Ms. Keeley Edwards

University of Manitoba, Winnipeg, Canada
orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 July 2023 in IEEE Transactions on Antennas and Propagation
Verified
Journal Article
Closed access
BMBen Martin
Ben Martin
Keeley EdwardsKeeley Edwards
Keeley Edwards
IJIan Jeffrey
Ian Jeffrey
Colin GilmoreColin Gilmore
Colin Gilmore
Microwave (or Electromagnetic) Imaging systems seek a quantitative reconstruction of the target permittivity. Such systems require calibration of the raw S -parameter data. This calibration is especially difficult in some systems where known targets cannot be introduced (e.g. grain bin imaging). Herein we present a machine-learning based method of calibrating such systems that does not require measuring a known target inside of the imaging chamber. Using a Cycle-Generative Adversarial Network (Cycle-GAN) machine-learning network, we show that we can calibrate data from a 2-D scalar Microwave Imaging (MWI) system and successfully reconstruct targets. Cycle-GAN makes use of two sets of data: experimental and synthetic. Unlike traditional calibration there is no need for the experimental data to be labeled i.e. the calibration targets do not need to be ‘known’. Results show that with an experimental calibration set of roughly 150 targets the Cycle-GAN approach provides comparable results to known-target calibration for the class of targets we used in this 2D near-field MWI systems..

13 March 2021 in Electronics
Verified
Journal Article
Open access
PDF file
Keeley EdwardsKeeley Edwards
Keeley Edwards
VKVahab Khoshdel
Vahab Khoshdel
MAMohammad Asefi
Mohammad Asefi
JLJoe LoVetri
Joe LoVetri
Colin GilmoreColin Gilmore
Colin Gilmore
IJIan Jeffrey
Ian Jeffrey
A two-stage workflow for detecting and monitoring tumors in the human breast with an inverse scattering-based technique is presented. Stage 1 involves a phaseless bulk-parameter inference neural network that recovers the geometry and permittivity of the breast fibroglandular region. The bulk parameters are used for calibration and as prior information for Stage 2, a full phase contrast source inversion of the measurement data, to detect regions of high relative complex-valued permittivity in the breast based on an assumed known overall tissue geometry. We demonstrate the ability of the workflow to recover the geometry and bulk permittivity of the different sized fibroglandular regions, and to detect and localize tumors of various sizes and locations within the breast model. Preliminary results show promise for a synthetically trained Stage 1 network to be applied to experimental data and provide quality prior information in practical imaging situations.

16 November 2020 in IEEE Access
Verified
Journal Article
Open access
Keeley EdwardsKeeley Edwards
Keeley Edwards
NGNicholas Geddert
Nicholas Geddert
KKKennedy Krakalovich
Kennedy Krakalovich
RKRyan Kruk
Ryan Kruk
MAMohammad Asefi
Mohammad Asefi
IJIan Jeffrey
Ian Jeffrey
We present a neural network architecture to determine the volume and complex permittivity of grain stored in metal bins. The neural networks output the grain height, cone angle and complex permittivity of the grain, using the input of experimental field data ( $S$ -parameters) from an electromagnetic imaging system consisting of 24 transceivers installed in the bin. Key for practical applications, the neural networks are trained on synthetic data sets but generate the parametric information using experimental data as input, without the use of calibration objects or open-short-load measurements. To accomplish this, we formulate a data normalization scheme that enables the use of a loss function that directly compares measured $S$ -parameters and simulation model fields. The normalization strategy and the ability to train on synthetic data means we do not need to collect experimental training data. We demonstrate the applicability of this synthetically trained neural network to experimental data from two different bin geometries, and discuss the ability of these neural networks to successfully infer parameters that can be used for grain inventory management. Our neural-network-based approach enables rapid inference, providing a more cost-effective long-term solution than existing optimization-based parametric inversion methods.

4 October 2023 in IEEE Transactions on Antennas and Propagation
Verified
Journal Article
Closed access
Keeley EdwardsKeeley Edwards
Keeley Edwards
BMBen Martin
Ben Martin
Colin GilmoreColin Gilmore
Colin Gilmore
IJIan Jeffrey
Ian Jeffrey
Uncertainty quantification of machine learning (ML) predictions is of key importance for the wide-spread adoption of ML-enabled electromagnetic imaging. As ML inference is a predictive process, providing a best (most likely) guess, supplementing that prediction with quantitative uncertainty can help to avoid costly errors when interpreting the output of a network. In this work, we present a novel two-output-branch neural network architecture that combines the Monte-Carlo Dropout Bayesian Convolutional Neural Network (BCNN) with an autoencoder (AE) to solve the data-to-image inverse problem. The inclusion of the autoencoder branch complements the predicted uncertainty image from the BCNN with a reconstruction of the network input. The data reconstruction (AE) path provides the user with additional information on the quality of the reconstruction, as a failed data reconstruction may be indicative of an out-of-range input, warranting further investigation.

1 January 2026 in IEEE Transactions on Antennas and Propagation
Verified
Journal Article
Closed access
BMBen Martin
Ben Martin
Keeley EdwardsKeeley Edwards
Keeley Edwards
HJHaylee Janz
Haylee Janz
Colin GilmoreColin Gilmore
Colin Gilmore
IJIan Jeffrey
Ian Jeffrey