A deep learning approach to the inverse problem of modulus identification in elasticity

Ni, Bo; Gao, Huajian

doi:10.1557/mrs.2020.231

Cited by 9 publications

(13 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A conventional deep-learning model is trained on a data set and can be applied to predict an elasticity distribution based on a new strain distribution without retraining the model (35). ElastNet, on the other hand, is not supervised by labeled data, and thus its performance is not limited by the amount, distribution, and accuracy of the data.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Learning hidden elasticity with deep neural networks

Chen

2021

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

Elastography is an imaging technique to reconstruct elasticity distributions of heterogeneous objects. Since cancerous tissues are stiffer than healthy ones, for decades, elastography has been applied to medical imaging for noninvasive cancer diagnosis. Although the conventional strain-based elastography has been deployed on ultrasound diagnostic-imaging devices, the results are prone to inaccuracies. Model-based elastography, which reconstructs elasticity distributions by solving an inverse problem in elasticity, may provide more accurate results but is often unreliable in practice due to the ill-posed nature of the inverse problem. We introduce ElastNet, a de novo elastography method combining the theory of elasticity with a deep-learning approach. With prior knowledge from the laws of physics, ElastNet can escape the performance ceiling imposed by labeled data. ElastNet uses backpropagation to learn the hidden elasticity of objects, resulting in rapid and accurate predictions. We show that ElastNet is robust when dealing with noisy or missing measurements. Moreover, it can learn probable elasticity distributions for areas even without measurements and generate elasticity images of arbitrary resolution. When both strain and elasticity distributions are given, the hidden physics in elasticity—the conditions for equilibrium—can be learned by ElastNet.

show abstract

Section: Resultsmentioning

confidence: 99%

“…a few inclusions and constrain the possible shapes, sizes, locations, and elastic moduli of these inclusions (35). However, when an ML model is trained in this manner, adding such artificial constraints limits the application of the model in practice.…”

Section: Significancementioning

confidence: 99%

Learning hidden elasticity with deep neural networks

Chen

2021

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

show abstract

“…Existing model-based approaches in elasticity imaging [1], [2] typically assume fixed regularization terms for various tissue types in elastography tasks while advanced priors might be required to mitigate the corresponding corrupted incomplete measurements and also to capture complex spatial information about the underlying tissues. On the other hand, end-to-end deep learning methods [3][4][5][6] require large datasets which conflict with fast and timeefficient image reconstruction essentials by real-time medical applications. Moreover, as the forward measurement model is not explicitly used in such methods, the estimated solution may not be consistent with the physics governing the imaging problem.…”

Section: Introductionmentioning

confidence: 99%

Ultrasound Elasticity Imaging Using Physics-Based Models and Learning-Based Plug-and-Play Priors

Mohammadi

Doyley

Çetin

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Existing physical model-based imaging methods for ultrasound elasticity reconstruction utilize fixed variational regularizers that may not be appropriate for the application of interest or may not capture complex spatial prior information about the underlying tissues. On the other hand, end-to-end learning-based methods count solely on the training data, not taking advantage of the governing physical laws of the imaging system. Integrating learning-based priors with physical forward models for ultrasound elasticity imaging, we present a joint reconstruction framework which guarantees that learning driven reconstructions are consistent with the underlying physics. For solving the elasticity inverse problem as a regularized optimization problem, we propose a plug-and-play (PnP) reconstruction approach in which each iteration of the elasticity image estimation process involves separate updates incorporating data fidelity and learningbased regularization. In this methodology, the data fidelity term is developed using a statistical linear algebraic model of quasi-static equilibrium equation revealing the relationship of the observed displacement fields to the unobserved elastic modulus. The regularizer comprises a convolutional neural network (CNN) based denoiser that captures the learned prior structure of the underlying tissues. Preliminary simulation results demonstrate the robustness and effectiveness of the proposed approach with limited training datasets and noisy displacement measurements.

show abstract

“…One major concern in existing model-based approaches [1], [2] is how to capture the appropriate prior information about the complex structure of the underlying tissues and how to incorporate this prior knowledge into the image reconstruction scheme. By the advent of deep neural networks (DNNs) [3], various end-to-end learning-based methods [4][5][6][7][8] have been proposed which try to learn both the physical model and the prior information about the underlying tissues. These methods lead to many shortcomings such as very large number of training pairs requirements and no-guaranteed solutions consistent with the true physical models.…”

Section: Introductionmentioning

confidence: 99%

Regularization by Adversarial Learning for Ultrasound Elasticity Imaging

Mohammadi¹,

Doyley²,

Çetin³

2021

Preprint

View full text Add to dashboard Cite

Classical model-based imaging methods for ultrasound elasticity inverse problem require prior constraints about the underlying elasticity patterns, while finding the appropriate hand-crafted prior for each tissue type is a challenge. In contrast, standard data-driven methods count solely on supervised learning on the training data pairs leading to massive network parameters for unnecessary physical model relearning which might not be consistent with the governing physical models of the imaging system. Fusing the physical forward model and noise statistics with data-adaptive priors leads to a united reconstruction framework that guarantees the learned reconstruction agrees with the physical models while coping with the limited training data. In this paper, we propose a new methodology for estimating the elasticity image by solving a regularized optimization problem which benefits from the physics-based modeling via a data-fidelity term and adversarially learned priors via a regularization term. In this method, the regularizer is trained based on the Wasserstein Generative Adversarial Network (WGAN) objective function which tries to distinguish the distribution of clean and noisy images. Leveraging such an adversarial regularizer for parameterizing the distribution of latent images and using gradient descent (GD) for solving the corresponding regularized optimization task leads to stability and convergence of the reconstruction compared to pixel-wise supervised learning schemes. Our simulation results verify the effectiveness and robustness of the proposed methodology with limited training datasets.

show abstract

A deep learning approach to the inverse problem of modulus identification in elasticity

Abstract: Abstract

Cited by 9 publications

References 23 publications

Learning hidden elasticity with deep neural networks

Learning hidden elasticity with deep neural networks

Ultrasound Elasticity Imaging Using Physics-Based Models and Learning-Based Plug-and-Play Priors

Regularization by Adversarial Learning for Ultrasound Elasticity Imaging

Contact Info

Product

Resources

About