Displacement Estimation in Ultrasound Elastography Using Pyramidal Convolutional Neural Network

Tehrani, Ali K. Z.; Rivaz, Hassan

doi:10.1109/tuffc.2020.2973047

Cited by 66 publications

(42 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We first performed a quantitative comparison on numerical simulation of both USENet and ReUSENet together with two state-of-the-art elastography methods, namely RF modified pyramid, warping and cost volume network (RFMPWC-Net) (Tehrani and Rivaz 2020 ) and global ultrasound elastography (GLUE) (Hashemi and Rivaz 2017 ). GLUE is an optimisation-based approach that relies on a regularised cost function to perform displacement estimation.…”

Section: Methodsmentioning

confidence: 99%

“…We used the publicly available demo code and trained weights of the RFMPWC-Net for comparison. The network’s weights have been fine-tuned in a supervised way using an ultrasound simulation database the authors made publicly available, ‘ultrasound simulation database for deep learning’ (Tehrani and Rivaz 2020 ) 6 6 The ultrasound simulation database, GLUE and RFMPWC-Net are available at https://users.encs.concordia.ca/~impact/ .…”

Section: Methodsmentioning

confidence: 99%

“…Recent methods have adopted the use of deep neural networks for USE, and have demonstrated high accuracy and robustness in displacement estimation. Most of these methods share the same general training strategy, which minimises a supervised loss function between the network’s displacement estimates and their respective ground truth labels, generated from numerical ultrasound phantoms via finite element methods (FEM) (Kibria and Rivaz 2018 , Wu et al 2018 , Gao et al 2019 , Peng et al 2020 , Tehrani and Rivaz 2020 ). This learning strategy prevents the model from training on real-world ultrasound data because ground truth displacement fields are not possible to obtain when the magnitude of applied stress is unknown.…”

Section: Introductionmentioning

confidence: 99%

“…We compare the performance of ReUSENet with a standard feed-forward neural network architecture, named here unsupervised strain elastography network (USENet). We validated our two models on numerical simulation and in vivo data, and compared our results to state-of-the-art deep learning-based and optimisation-based algorithms (Hashemi and Rivaz 2017 , Tehrani and Rivaz 2020 ). Both networks can be run in real-time at a speed of about 20 frames per second with a standard 12 GB GPU.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

An unsupervised learning approach to ultrasound strain elastography with spatio-temporal consistency

Delaunay

Vercauteren

2021

Phys. Med. Biol.

View full text Add to dashboard Cite

Quasi-static ultrasound elastography (USE) is an imaging modality that measures deformation (i.e. strain) of tissue in response to an applied mechanical force. In USE, the strain modulus is traditionally obtained by deriving the displacement field estimated between a pair of radio-frequency data. In this work we propose a recurrent network architecture with convolutional long-short-term memory decoder blocks to improve displacement estimation and spatio-temporal continuity between time series ultrasound frames. The network is trained in an unsupervised way, by optimising a similarity metric between the reference and compressed image. Our training loss is also composed of a regularisation term that preserves displacement continuity by directly optimising the strain smoothness, and a temporal continuity term that enforces consistency between successive strain predictions. In addition, we propose an open-access in vivo database for quasi-static USE, which consists of radio-frequency data sequences captured on the arm of a human volunteer. Our results from numerical simulation and in vivo data suggest that our recurrent neural network can account for larger deformations, as compared with two other feed-forward neural networks. In all experiments, our recurrent network outperformed the state-of-the-art for both learning-based and optimisation-based methods, in terms of elastographic signal-to-noise ratio, strain consistency, and image similarity. Finally, our open-source code provides a 3D-slicer visualisation module that can be used to process ultrasound RF frames in real-time, at a rate of up to 20 frames per second, using a standard GPU.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An unsupervised learning approach to ultrasound strain elastography with spatio-temporal consistency

Delaunay

Vercauteren

2021

Phys. Med. Biol.

View full text Add to dashboard Cite

show abstract

“…In the ultrasound elastography, Peng et al [23] adopted a multistack Flownet network (FlownetCSS) as the speckle tracking method. Tehrani et al [24] used modified pyramidal CNNs to exploit information in RF data for displacement estimation. They also utilized an unsupervised fine-tuning method for the same goal [25].…”

Section: Introductionmentioning

confidence: 99%

Phase Aberration Correction: A Convolutional Neural Network Approach

2020

Self Cite

View full text Add to dashboard Cite

One of the main sources of image degradation in ultrasound imaging is the phase aberration effect, which imposes limitations to both data acquisition and reconstruction. Phase aberration is induced by spatial changes in sound velocity compared to the default values and degrades the quality of beam focusing. In addition, it prevents received channel signals to be summed coherently. In this paper, for the first time, we propose a method to estimate the aberrator profile from an ultrasound B-mode image using a deep convolutional neural network (CNN) in order to compensate for the phase aberration effect. In contrast to traditional methods, which mostly apply time-consuming processing techniques on channel RF signals and need several iterations for a reasonable accuracy, the proposed approach is computationally efficient, and utilizes only the B-mode image to estimate the aberrator profile in one shot with a high accuracy. We experimentally investigate main characteristics of the proposed approach and present a quantitative evaluation of the estimated aberrator profile. The proposed method is compared with the conventional delay-and-sum (DAS) method and a method based on nearest-neighbor cross-correlation (NNCC). Results demonstrate that the proposed CNN method substantially outperforms other methods. INDEX TERMS Convolutional neural networks, deep learning, phase aberration, ultrasound.

show abstract

Deep learning in ultrasound elastography imaging: A review

Bhatt

et al. 2022

Medical Physics

View full text Add to dashboard Cite

It is known that changes in the mechanical properties of tissues are associated with the onset and progression of certain diseases. Ultrasound elastography is a technique to characterize tissue stiffness using ultrasound imaging either by measuring tissue strain using quasi‐static elastography or natural organ pulsation elastography, or by tracing a propagated shear wave induced by a source or a natural vibration using dynamic elastography. In recent years, deep learning has begun to emerge in ultrasound elastography research. In this review, several common deep learning frameworks in the computer vision community, such as multilayered perceptron, convolutional neural network, and recurrent neural network, are described. Then, recent advances in ultrasound elastography using such deep learning techniques are revisited in terms of algorithm development and clinical diagnosis. Finally, the current challenges and future developments of deep learning in ultrasound elastography are prospected.

show abstract

Displacement Estimation in Ultrasound Elastography Using Pyramidal Convolutional Neural Network

Cited by 66 publications

References 34 publications

An unsupervised learning approach to ultrasound strain elastography with spatio-temporal consistency

An unsupervised learning approach to ultrasound strain elastography with spatio-temporal consistency

Phase Aberration Correction: A Convolutional Neural Network Approach

Deep learning in ultrasound elastography imaging: A review

Contact Info

Product

Resources

About