Circumventing the solution of inverse problems in mechanics through deep learning: Application to elasticity imaging

Patel, Dhruv; Tibrewala, Raghav; Vega, Adriana; Dong, Li; Hugenberg, Nicholas; Oberai, Assad A.

doi:10.1016/j.cma.2019.04.045

Cited by 33 publications

(15 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AI techniques in the field of medicine are gaining more interest recently, but it should be recognized that initial efforts had started as early as in 1970s (12). Due to enhanced algorithms and computing power, AI has made great strides in many areas -classification of photos, person recognition, self-driving vehicles, natural language processing, and data mining, just to name a few and has generated great enthusiasm for streamlining cancer screening to improve early detection and diagnosis of cancer and personalize treatment and outcome prediction (72)(73)(74)(75)(76)(77)(78). The future of AI in lung cancer screening lies in the integration of algorithms that detect and diagnose all diseases visible in a LDCT, not only lung cancer but emphysema, interstitial lung disease, cardiovascular disease, and liver disease to automatically produce a report of everything that is visible on a LDCT chest scan.…”

Section: Integration Of Ai Approaches For Ldct Interpretationmentioning

confidence: 99%

Review of the future of lung cancer screening

Yip¹,

Jirapatnakul²,

Li³

et al. 2022

Curr Chall Thorac Surg

View full text Add to dashboard Cite

Lung cancer screening has progressed greatly in the last two decades since the initial publication of ELCAP in 1999. Since then, large randomized trials, the NLST and NELSON studies, have also shown the benefit of lung cancer screening with CT, and it is now widely available. There are many areas for improving the benefit of screening by refining the eligibility criteria, continued optimization of the screening regimen, identifying other related diseases, and improved treatment. As artificial intelligence (AI) techniques continue to improve, an automated report of all diseases can be produced from the low-dose CT scan. New screening tests are being developed for lung cancer including blood-based tests. There is growing recognition that use of risk factor-based criteria which dichotomize age and pack-years of smoking to select eligible individuals for screening may not be the most efficient and may also lead to increased lung cancer disparity. Continued research will focus on optimal approaches for selecting high risk individuals for lung cancer screening and thus hopefully expand eligibility criteria beyond the current criteria which include only 20% of lung cancer patients. With the increase in early stage lung cancers identified from lung cancer screening, a key focus of ongoing research is the treatment of these cancers.

show abstract

Section: Integration Of Ai Approaches For Ldct Interpretationmentioning

confidence: 99%

Review of the future of lung cancer screening

Yip¹,

Jirapatnakul²,

Li³

et al. 2022

Curr Chall Thorac Surg

View full text Add to dashboard Cite

show abstract

“…Given the success achieved by neural networks on the wide variety of applications cited in the previous section, it is natural to explore the application of neural networks to the inverse problem of elasticity imaging and several recent efforts [11,13,15] have done so.…”

Section: Neural Network and Elasticity Imagingmentioning

confidence: 99%

Solving an elastic inverse problem using Convolutional Neural Networks

Gokhale¹

2021

Preprint

View full text Add to dashboard Cite

We explore the application of a Convolutional Neural Network (CNN) to image the shear modulus field of an almost incompressible, isotropic, linear elastic medium in plane strain using displacement or strain field data. This problem is important in medicine because the shear modulus of suspicious and potentially cancerous growths in soft tissue is elevated by about an order of magnitude as compared to the background of normal tissue. Imaging the shear modulus field therefore can lead to high-contrast medical images. Our imaging problem is: Given a displacement or strain field (or its components), predict the corresponding shear modulus field. Our CNN is trained using 6000 training examples consisting of a displacement or strain field and a corresponding shear modulus field. We observe encouraging results which warrant further research and show the promise of this methodology.

show abstract

“…The work [4] used neural networks to identify the shape of the aberration prism that distorts the ultrasonic signal. The work [5] applied convolutional networks to elasticity imaging to distinguish benign tumors from their malignant counterparts based on measured displacement fields on the boundary of the domain.…”

Section: Introductionmentioning

confidence: 99%

Learning velocity model for complex media with deep convolutional neural networks

Stankevich,

Nechepurenko,

Shevchenko

et al. 2021

Preprint

View full text Add to dashboard Cite

The paper considers the problem of velocity model acquisition for a complex media based on boundary measurements. The acoustic model is used to describe the media. We used an open-source dataset of velocity distributions to compare the presented results with the previous works directly. Forward modeling is performed using the grid-characteristic numerical method. The inverse problem is solved using deep convolutional neural networks. Modifications for a baseline UNet architecture are proposed to improve both structural similarity index measure quantitative correspondence of the velocity profiles with the ground truth. We evaluate our enhancements and demonstrate the statistical significance of the results.

show abstract

Circumventing the solution of inverse problems in mechanics through deep learning: Application to elasticity imaging

Cited by 33 publications

References 24 publications

Review of the future of lung cancer screening

Review of the future of lung cancer screening

Solving an elastic inverse problem using Convolutional Neural Networks

Learning velocity model for complex media with deep convolutional neural networks

Contact Info

Product

Resources

About