Fast inter-frame motion correction in contrast-free ultrasound quantitative microvasculature imaging using deep learning.
Contrast-free ultrasound microvessel imaging
Deep learning
High frame rate
Inter-frame motion correction
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
30 10 2024
30 10 2024
Historique:
received:
03
07
2024
accepted:
23
10
2024
medline:
31
10
2024
pubmed:
31
10
2024
entrez:
31
10
2024
Statut:
epublish
Résumé
Contrast-free ultrasound quantitative microvasculature imaging shows promise in several applications, including the assessment of benign and malignant lesions. However, motion represents one of the major challenges in imaging tumor microvessels in organs that are prone to physiological motions. This study aims at addressing potential microvessel image degradation in in vivo human thyroid due to its proximity to carotid artery. The pulsation of the carotid artery induces inter-frame motion that significantly degrades microvasculature images, resulting in diagnostic errors. The main objective of this study is to reduce inter-frame motion artifacts in high-frame-rate ultrasound imaging to achieve a more accurate visualization of tumor microvessel features. We propose a low-complex deep learning network comprising depth-wise separable convolutional layers and hybrid adaptive and squeeze-and-excite attention mechanisms to correct inter-frame motion in high-frame-rate images. Rigorous validation using phantom and in-vivo data with simulated inter-frame motion indicates average improvements of 35% in Pearson correlation coefficients (PCCs) between motion corrected and reference data with respect to that of motion corrupted data. Further, reconstruction of microvasculature images using motion-corrected frames demonstrates PCC improvement from 31 to 35%. Another thorough validation using in-vivo thyroid data with physiological inter-frame motion demonstrates average improvement of 20% in PCC and 40% in mean inter-frame correlation. Finally, comparison with the conventional image registration method indicates the suitability of proposed network for real-time inter-frame motion correction with 5000 times reduction in motion corrected frame prediction latency.
Identifiants
pubmed: 39478021
doi: 10.1038/s41598-024-77610-4
pii: 10.1038/s41598-024-77610-4
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
26161Subventions
Organisme : NIH HHS
ID : R01CA239548
Pays : United States
Informations de copyright
© 2024. The Author(s).
Références
Zhao, S. X. et al. A local and global feature disentangled network: Toward classification of benign-malignant thyroid nodules from US image. IEEE Trans. Med. Imaging 41(6), 1497–1509 (2022).
doi: 10.1109/TMI.2022.3140797
pubmed: 34990353
Evain, E. et al. Motion estimation by deep learning in 2D echocardiography: Synthetic dataset and validation. IEEE Trans. Med. Imaging 41(8), 1911–1924 (2022).
doi: 10.1109/TMI.2022.3151606
pubmed: 35157582
Ternifi, R. et al. Quantitative biomarkers for cancer detection using contrast-free US high-definition microvessel imaging: Fractal dimension, Murray’s deviation, bifurcation angle and spatial vascularity pattern. IEEE Trans. Med. Imaging 40(12), 3891–3900 (2021).
doi: 10.1109/TMI.2021.3101669
pubmed: 34329160
pmcid: 8668387
Bayat, M. et al. Background removal and vessel filtering of non-contrast US images of microvasculature. IEEE Trans. Biomed. Eng. 66(3), 831–842 (2019).
doi: 10.1109/TBME.2018.2858205
pubmed: 30040621
Ghavami, S., Bayat, M., Fatemi, M. & Alizad, A. Quantification of morphological features in non-contrast-enhanced US microvasculature imaging. IEEE Access. 8, 18925–18937 (2020).
doi: 10.1109/ACCESS.2020.2968292
pubmed: 32328394
pmcid: 7179329
Ternifi, R. et al. Ultrasound high-definition microvasculature imaging with novel quantitative biomarkers improves breast cancer detection accuracy. Eur. Radiol. 32, 7448–7462 (2022).
doi: 10.1007/s00330-022-08815-2
pubmed: 35486168
pmcid: 9616967
Sabeti, S. et al. Morphometric analysis of tumor microvessels for detection of hepatocellular carcinoma using contrast-free US imaging: A feasibility study. Front. Oncol. 13, 1121664 (2023).
doi: 10.3389/fonc.2023.1121664
pubmed: 37124492
pmcid: 10134399
Gu, J. et al. Volumetric imaging and morphometric analysis of breast tumor angiogenesis using a new contrast-free US technique: a feasibility study. Breast Cancer Res. 24(1), 1–5 (2022).
doi: 10.1186/s13058-022-01583-3
pubmed: 34983617
pmcid: 8725284
Kurti, M. et al. Quantitative biomarkers derived from a novel contrast-free US high-definition microvessel imaging for distinguishing thyroid nodules. Cancers 15(6), 1888 (2023).
doi: 10.3390/cancers15061888
pubmed: 36980774
pmcid: 10046818
Nayak, R. et al. Non-invasive small vessel imaging of human thyroid using motion-corrected spatiotemporal clutter filtering. Ultrasound Med. Biol. 45(4), 1010–1018 (2019).
doi: 10.1016/j.ultrasmedbio.2018.10.028
pubmed: 30718145
pmcid: 6391182
Harput, S. et al. Two stage sub-wavelength motion correction in human microvasculature for CEUS imaging. in 2017 IEEE International Ultrasonics Symposium (IUS) IEEE. 1–4 (2017).
Harput, S. et al. Two-stage motion correction for super-resolution US imaging in human lower limb. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 65(5), 803–814 (2018).
doi: 10.1109/TUFFC.2018.2824846
pubmed: 29733283
Ta, C. N. et al. 2-tier in-plane motion correction and out-of-plane motion filtering for contrast-enhanced US. Invest. Radiol. 49(11), 707–719 (2014).
doi: 10.1097/RLI.0000000000000074
pubmed: 24901545
pmcid: 4184933
Oezdemir, I. et al. Faster motion correction of clinical contrast-enhanced US imaging using deep learning. in 2020 IEEE International Ultrasonics Symposium (IUS) (2020).
Barrois, G. et al. New reference-free, simultaneous motion-correction and quantification in dynamic contrast-enhanced US. in2014 IEEE International Ultrasonics Symposium (2014).
Hingot, V. et al. Subwavelength motion-correction for ultrafast US localization microscopy. Ultrasonics 77, 17–21 (2017).
doi: 10.1016/j.ultras.2017.01.008
pubmed: 28167316
Hao, Y. et al. Non-rigid motion correction for US localization microscopy of the liver in vivo. in2019 IEEE International Ultrasonics Symposium (IUS), 2263–2266 (2019).
Nayak, R. et al. Non-contrast agent based small vessel imaging of human thyroid using motion corrected power Doppler imaging. Sci. Rep. 8(1), 15318 (2018).
doi: 10.1038/s41598-018-33602-9
pubmed: 30333509
pmcid: 6193022
Taghavi, I. et al. In vivo motion correction in super-resolution imaging of rat kidneys. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 68(10), 3082–3093 (2021).
doi: 10.1109/TUFFC.2021.3086983
pubmed: 34097608
Stanziola, A. et al. Motion correction in contrast-enhanced US scans of carotid atherosclerotic plaques. in 2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI) (2015).
Stanziola, A. et al. Motion artifacts and correction in multipulse high-frame rate contrast-enhanced US. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 66(2), 417–420 (2018).
doi: 10.1109/TUFFC.2018.2887164
pubmed: 30571621
Zhang, J. et al. Respiratory motion correction for liver contrast enhanced US by automatic selection of a reference image. Med. Phys. 46(11), 4992–5001 (2019).
doi: 10.1002/mp.13776
pubmed: 31444798
Brown, K. G. et al. Deep learning of spatiotemporal filtering for fast super-resolution US imaging. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 67(9), 1820–1829 (2020).
doi: 10.1109/TUFFC.2020.2988164
pubmed: 32305911
pmcid: 7523282
Zheng, S. et al. A deep learning method for motion artifact correction in intravascular photoacoustic image sequence. IEEE Trans. Med. Imaging 42(1), 66–78 (2022).
doi: 10.1109/TMI.2022.3202910
pubmed: 36037455
Shi, L. et al. Automatic inter-frame patient motion correction for dynamic cardiac PET using deep learning. IEEE Trans. Med. Imaging 40(12), 3293–3304 (2021).
doi: 10.1109/TMI.2021.3082578
pubmed: 34018932
pmcid: 8670362
Chen, G. et al. AAU-Net: An adaptive attention U-Net for breast lesions segmentation in ultrasound images. IEEE Trans. Med. Imaging 42(5), 1289–1300 (2023).
doi: 10.1109/TMI.2022.3226268
pubmed: 36455083
Nayak, R. et al. Quantitative assessment of ensemble coherency in contrast-free US microvasculature imaging. Med. Phys. 48(7), 3540–3558 (2021).
doi: 10.1002/mp.14918
pubmed: 33942320
Adusei, S. et al. Custom-made flow phantoms for quantitative US microvessel imaging. Ultrasonics 134, 107092 (2023).
doi: 10.1016/j.ultras.2023.107092
pubmed: 37364357
pmcid: 10530522
Goodfellow, I. et al. Deep Learning (MIT Press, 2016).
Bai, L., Zhao, Y. & Huang, X. A CNN accelerator on FPGA using depthwise separable convolution. IEEE Trans. Circuits Syst. II Express Briefs 65(10), 1415–1419 (2018).
Hendrycks, D. & Gimpel, K. Gaussian error linear units (gelus). arXiv preprint arXiv:1606.08415 (2016).
Saini, M., Satija, U. & Upadhayay, M. D. DSCNN-CAU: Deep-learning-based mental activity classification for IoT implementation toward portable BCI. IEEE Internet Things J. 10(10), 8944–8957 (2023).
doi: 10.1109/JIOT.2022.3232481
Chollet, F. Xception: Deep learning with depthwise separable convolutions. in In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 1251–1258 (2017).
Howard, A. G. et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861 . (2017).
Saini, M. & Satija, U. On-Device implementation for deep-learning-based cognitive activity prediction. IEEE Sens. Lett. 6(4), 1–4 (2022).
doi: 10.1109/LSENS.2022.3156158
Hu, J., Shen, L. & Sun, G. Squeeze-and-excitation networks. in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition 7132–7141 (2018).
Neto, A. M. et al. Image processing using Pearson’s correlation coefficient: Applications on autonomous robotics. in 2013 13th IEEE International Conference on Autonomous Robot Systems (2013).
Mason, A. et al. Comparison of objective image quality metrics to expert radiologists’ scoring of diagnostic quality of MR images. IEEE Trans. Med. Imaging 39(4), 1064–1072 (2020).
doi: 10.1109/TMI.2019.2930338
pubmed: 31535985
Akasaka, T. et al. Optimization of regularization parameters in compressed sensing of magnetic resonance angiography: Can statistical image metrics mimic radiologists’ perception?. PLoS ONE 11(1), e0146548 (2016).
doi: 10.1371/journal.pone.0146548
pubmed: 26744843
pmcid: 4706324
Mudeng, V. et al. Prospects of structural similarity index for medical image analysis. Appl. Sci. 12(8), 3754 (2022).
doi: 10.3390/app12083754
Wang, Z. & Bovik, A. C. Mean squared error: Love it or leave it? A new look at signal fidelity measures. IEEE Sign. Process. Mag. 26(1), 98–117 (2009).
doi: 10.1109/MSP.2008.930649