Abstract
Purpose
The purpose of this study was to detect two dimensional and sub-pixel displacement with high spatial resolution using an ultrasonic diagnostic apparatus. Conventional displacement detection methods assume neighborhood uniformity and cannot achieve both high spatial resolution and sub-pixel displacement detection.
Methods
A deep-learning network that utilizes ultrasound images and output displacement distribution was developed. The network structure was constructed by modifying FlowNet2, a widely used network for optical flow estimation, and a training dataset was developed using ultrasound image simulation. Detection accuracy and spatial resolution were evaluated via simulated ultrasound images, and the clinical usefulness was evaluated with ultrasound images of the liver exposed to high-intensity-focused ultrasound (HIFU). These results were compared to the Lucas–Kanade method, a conventional sub-pixel displacement detection method.
Results
For a displacement within ± 40 µm (± 0.6 pixels), a pixel size of 67 µm, and signal noise of 1%, the accuracy was above 0.5 µm and 0.2 µm, the precision was above 0.4 µm and 0.3 µm, and the spatial resolution was 1.1 mm and 0.8 mm for the lateral and axial displacements, respectively. These improvements were also observed in the experimental data. Visualization of the lateral displacement distribution, which determines the edge of the treated lesion using HIFU, was also realized.
Conclusion
Two-dimensional and sub-pixel displacement detection with high spatial resolution was realized using a deep-learning methodology. The proposed method enabled the monitoring of small and local tissue deformations induced by HIFU exposure.
Similar content being viewed by others
References
Szabo T. Diagnostic ultrasound imaging: inside out. Cambridge: Academic Press; 2013.
Lucas BD, Kanade T. An iterative image registration technique with an application to stereo vision. Proceedings of imaging understanding workshop. 1981;121–30
Yoshikawa H, Yoshizawa S, Umemura S, et al. Ultrasound sub-pixel motion-tracking method with out-of-plane motion detection for precise vascular imaging. Ultrasound Med Biol. 2019. https://doi.org/10.1016/j.ultrasmedbio.2019.11.005.
Yoshikawa H. Study on precise vascular imaging with ultrasound speckle tracking. Doctral dissertation, Tohoku University, Sendai, Japan. 2017, p. 143. Accessed 1st Mar 2021. Available from: http://hdl.handle.net/10097/00125122.
Kanai H, Sato M, Chubachi N, et al. Transcutaneous measurement and spectrum analysis of heart wall vibrations. IEEE Trans Ultrason Ferroelectr Freq Control. 1996;43:791–810.
Shiina T, Nitta N, Bamber JC, et al. Real time tissue elasticity imaging using the combined autocorrelation method. J Med Ultrason. 2002;29:119–28.
Goodfellow I, Gengio Y, Courville A. Deep learning. Cambridge: The MIT Press; 2016.
Evan E, Faraz K, Grenier T, Garcia D, et al. A pilot study on convolutional neural networks for motion estimation from ultrasound images. IEEE Trans Ultrason Ferroelectr Freq Control. 2020;67:2565–73.
Xiao C, Li Z, Lu J, et al. A new deep learning method for displacement tracking from ultrasound RF signals of vascular walls. Comput Med Imaging Graph. 2021. https://doi.org/10.1016/j.compmedimag.2020.101819.
Tehrani AKZ, Rivaz H. Displacement estimation in ultrasound elastography using pyramidal convolutional neural network. IEEE Trans Ultrason Ferroelectr Freq Control. 2020;67:2629–39.
Peng B, Xian Y, Jiang J. A convolution neural network-based speckle tracking method for ultrasound elastography. IEEE Int Ultrason Symp (IUS). 2018; Doi: https://doi.org/10.1109/ULTSYM.2018.8580034
Kibria MG, Rivaz H. Global ultrasound elastography using convolutional neural network. 2018, p. 4. Accessed on 1 Mar 2021. Available from: https://arxiv.org/pdf/1805.07493.pdf
Gao A, Wu S, Liu Z, et al. Learning the implicit strain reconstruction in ultrasound elastography using privileged information. Med Image Anal. 2019. https://doi.org/10.1016/j.media.2019.101534.
Wu S, Gao Z, Lui J et al. Direct reconstruction of ultrasound elastography using an end-to-end deep neural network. International conference on medical image computing and computer assisted intervention (MICCAI). 2018;374–82
Ilg E, Mayer N, Saikia T et al. FlowNet2.0: evolution of optical flow estimation with deep networks. 2016, p.16. Accessed on 1 Mar 2021. Available from: https://arxiv.org/pdf/1612.01925.pdf
Sum D, Yang X, Liu MY et al. PWC-Net: CNNs for optical flow using pyramid, warping, and cost volume. 2018, p. 18. Accessed on 1 Mar 2021. Available from: https://arxiv.org/pdf/1709.02371.pdf
Jensen JA. Field: a program for simulating ultrasound systems. The 10th Nordic-Baltic conference on biomedical imaging published in medical and biological engineering and computing. 1996;34:351–3
Giomore GR. Practival gamma-ray spectrometry. New York: Wiley; 2008.
Yamamoto M, Yoshizawa S. Analysis of tissue displacement induced by high-intensity focused ultrasound exposure for coagulation monitoring. Jpn J Appl Phys. 2021;60:040903.
Montaldo G, Tanter M, Bercoff J, et al. Coherent plane-wave compounding for very high frame rate ultrasonography and transient elastography. IEEE Trans Ultrason Ferroelectr Freq Control. 2009;56:489–506.
Sasaki S, Takagi R, Matsuura K, et al. Feasibility of real-time treatment feedback using novel filter for eliminating therapeutic ultrasound noise with high-speed ultrasonic imaging in ultrasound-guided high-intensity focused ultrasound treatment. Jpn J Appl Phys. 2014;53:07KF10.
Berson M, Roncin A, Pourcelot L. Compound scanning with an electrically steered beam. Ultrason Imaging. 1981. https://doi.org/10.1177/016173468100300306.
Luo J, Chen CW, et al. Artifact reduction in low bit rate DCT-based image compression. IEEE Trans Image Proc. 1996;9:1363–8.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
There are no financial or other relations that could lead to a conflict of interest.
Ethical approval
The samples used in this study were obtained via commercial distribution and were not associated with animal experiments.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Yamamoto, M., Yoshizawa, S. Displacement detection with sub-pixel accuracy and high spatial resolution using deep learning. J Med Ultrasonics 49, 3–15 (2022). https://doi.org/10.1007/s10396-021-01162-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10396-021-01162-7