Preconditioned P-ULA for Joint Deconvolution-Segmentation of Ultrasound Images.
Corbineau, M., Kouamé, D., Chouzenoux, E., Tourneret, J., & Pesquet, J.
IEEE Signal Processing Letters, 26(10): (on line). 2019.
link
bibtex
@Article{ CoKoChToPe2019.1,
author = {Corbineau, Marie-Caroline and Kouam\'e, Denis and Chouzenoux, Emilie and Tourneret, Jean-Yves and Pesquet, Jean-Christophe},
title = "{Preconditioned P-ULA for Joint Deconvolution-Segmentation of Ultrasound Images}",
journal = {IEEE Signal Processing Letters},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
year = {2019},
volume = {26},
number = {10},
pages = {(on line)},
language = {anglais}
}
A Tensor Factorization Method for 3D Super-Resolution with Application to Dental CT.
Hatvani, J., Basarab, A., Tourneret, J., Gyongy, M., & Kouamé, D.
IEEE Transactions on Medical Imaging, 38(6): 1524–1531. April 2019.
link
bibtex
@Article{ HaBaToGyKo2019.1,
author = {Hatvani, Janka and Basarab, Adrian and Tourneret, Jean-Yves and Gyongy, Miklos and Kouam\'e, Denis},
title = "{A Tensor Factorization Method for 3D Super-Resolution with Application to Dental CT}",
journal = {IEEE Transactions on Medical Imaging},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
year = {2019},
month = {April},
volume = {38},
number = {6},
pages = {1524--1531},
language = {anglais}
}
Deep Learning-Based Super-Resolution Applied to Dental Computed Tomography.
Hatvani, J., Horvath, A., Michetti, J., Basarab, A., Kouamé, D., & Gyongy, M.
IEEE Transactions on Radiation and Plasma Medical Sciences, Machine learning in radiation based medical sciences, 3(2): 120–128. March 2019.
Paper
link
bibtex
abstract
@Article{ HaHoMiBaKoGy2019.1,
author = {Hatvani, Janka and Horvath, Andras and Michetti, J\'er�me and Basarab, Adrian and Kouam\'e, Denis and Gyongy, Miklos},
title = "{Deep Learning-Based Super-Resolution Applied to Dental Computed Tomography}",
journal = {IEEE Transactions on Radiation and Plasma Medical Sciences, Machine learning in radiation based medical sciences},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
year = {2019},
month = {March},
volume = {3},
number = {2},
pages = {120--128},
language = {anglais},
URL = {https://doi.org/10.1109/TRPMS.2018.2827239 - https://oatao.univ-toulouse.fr/24705/},
abstract = {The resolution of dental computed tomography (CT) images is limited by detector geometry, sensitivity, patient movement, the reconstruction technique and the need to minimize radiation dose. Recently, the use of
convolutional neural network (CNN) architectures has shown promise as a resolution enhancement method. In the current work, two CNN architectures?a subpixel network and the so called U-net?have been considered for the resolution
enhancement of 2-D cone-beam CT image slices of ex vivo teeth. To do so, a training set of 5680 cross-sectional slices of 13 teeth and a test set of 1824 slices of 4 structurally different teeth were used. Two existing
reconstruction-based super-resolution methods using l2-norm and total variation regularization were used for comparison. The results were evaluated with different metrics (peak signal-to-noise ratio, structure similarity index,
and other objective measures estimating human perception) and subsequent image-segmentation-based analysis. In the evaluation, micro-CT images were used as ground truth.The results suggest the superiority of the proposed CNN-based
approaches over reconstruction-based methods in the case of dental CT images, allowing better detection of medically salient features, such as the size, shape, or curvature of the root canal.}
}
The resolution of dental computed tomography (CT) images is limited by detector geometry, sensitivity, patient movement, the reconstruction technique and the need to minimize radiation dose. Recently, the use of convolutional neural network (CNN) architectures has shown promise as a resolution enhancement method. In the current work, two CNN architectures?a subpixel network and the so called U-net?have been considered for the resolution enhancement of 2-D cone-beam CT image slices of ex vivo teeth. To do so, a training set of 5680 cross-sectional slices of 13 teeth and a test set of 1824 slices of 4 structurally different teeth were used. Two existing reconstruction-based super-resolution methods using l2-norm and total variation regularization were used for comparison. The results were evaluated with different metrics (peak signal-to-noise ratio, structure similarity index, and other objective measures estimating human perception) and subsequent image-segmentation-based analysis. In the evaluation, micro-CT images were used as ground truth.The results suggest the superiority of the proposed CNN-based approaches over reconstruction-based methods in the case of dental CT images, allowing better detection of medically salient features, such as the size, shape, or curvature of the root canal.
Brain Segmentation from Super-Resolved Magnetic Resonance Images (regular paper).
Bazzi, F., Rodriguez-Callejas, J. D., Fonta, C., Diab, A., Amoud, H., Falou, O., Mescam, M., Basarab, A., & Kouamé, D.
In
International Conference on Advances in Biomedical Engineering, Tripoli, Lebanon, 17/10/2019-19/10/2019, 2019.
link
bibtex
@InProceedings{ BaRoDiAmFaMeBaKo2019.1,
author = {Bazzi, Farah and Rodriguez-Callejas, Juan Dios and Fonta, Caroline and Diab, Ahmad and Amoud, Hassan and Falou, Omar and Mescam, Muriel and Basarab, Adrian and Kouam\'e, Denis},
title = "{Brain Segmentation from Super-Resolved Magnetic Resonance Images (regular paper)}",
booktitle = "{International Conference on Advances in Biomedical Engineering, Tripoli, Lebanon, 17/10/2019-19/10/2019}",
year = {2019},
publisher = {},
pages = {},
language = {anglais}
}
High-resolution and high-sensitivity blood flow estimation using deconvolution and optimization approaches with application to thyroid vascularization imaging (regular paper).
Shen, H., Barthélémy, C., Khoury, E., Zemmoura, I., Reménieras, J., Basarab, A., & Kouamé, D.
In
IEEE International Ultrasonics Symposium, Glasgow, 06/10/2019-09/10/2019, pages (electronic medium), 2019.
Paper
link
bibtex
@InProceedings{ ShBaKhZeReBaKo2019.1,
author = {Shen, Hong and Barth\'el\'emy, Chlo\'e and Khoury, Elise and Zemmoura, Ilyess and Rem\'enieras, Jean-Pierre and Basarab, Adrian and Kouam\'e, Denis},
title = "{High-resolution and high-sensitivity blood flow estimation using deconvolution and optimization approaches with application to thyroid vascularization imaging (regular paper)}",
booktitle = "{IEEE International Ultrasonics Symposium, Glasgow, 06/10/2019-09/10/2019}",
year = {2019},
publisher = {},
pages = {(electronic medium)},
language = {anglais},
URL = {https://oatao.univ-toulouse.fr/26249/}
}
Development of ultrasensitive Doppler imaging method for the surgical management of open-brain tumors (regular paper).
Barthélémy, C., Khoury, E., Beuve, S., Zemmoura, I., Gennisson, J., Basarab, A., Kouamé, D., & Reménieras, J.
In
IEEE International Ultrasonics Symposium, Glasgow, 06/10/2019-09/10/2019, pages (on line), http://www.ieee.org/, 2019. IEEE : Institute of Electrical and Electronics Engineers
Paper
link
bibtex
abstract
@InProceedings{ BaKhBeZeGeBaKoRe2019.1,
author = {Barth\'el\'emy, Chlo\'e and Khoury, Elise and Beuve, Steve and Zemmoura, Ilyess and Gennisson, Jean-Luc and Basarab, Adrian and Kouam\'e, Denis and Rem\'enieras, Jean-Pierre},
title = "{Development of ultrasensitive Doppler imaging method for the surgical management of open-brain tumors (regular paper)}",
booktitle = "{IEEE International Ultrasonics Symposium, Glasgow, 06/10/2019-09/10/2019}",
year = {2019},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
pages = {(on line)},
language = {anglais},
URL = {https://doi.org/10.1109/ULTSYM.2019.8925955 - https://oatao.univ-toulouse.fr/26248/},
abstract = {Gliomas are infiltrating tumors in the healthy brain parenchyma with no clear boundaries and can be located near or within "functional" brain zones driving as an example, motor skills, sensitivity, cognition or vision.
Two conflicting objectives must be achieved during cerebral glioma surgery: (1) to obtain a tumor excision as complete as possible, the oncological prognosis being improved by surgery; and (2) to limit the risk of definitive
neurological deficit by respecting the brain areas infiltrated by the tumor reMayning functional. The objective of our work is to develop an intraoperative biomecanical analysis and a micro vascularization imaging method and
validate the interest of these techniques for the diagnosis of tumor neo-angiogenesis and ultimately to target the surgical procedure during surgery.}
}
Gliomas are infiltrating tumors in the healthy brain parenchyma with no clear boundaries and can be located near or within "functional" brain zones driving as an example, motor skills, sensitivity, cognition or vision. Two conflicting objectives must be achieved during cerebral glioma surgery: (1) to obtain a tumor excision as complete as possible, the oncological prognosis being improved by surgery; and (2) to limit the risk of definitive neurological deficit by respecting the brain areas infiltrated by the tumor reMayning functional. The objective of our work is to develop an intraoperative biomecanical analysis and a micro vascularization imaging method and validate the interest of these techniques for the diagnosis of tumor neo-angiogenesis and ultimately to target the surgical procedure during surgery.
Spatio-temporal compressed quantitative acoustic microscopy (regular paper).
Kim, J., Mamou, J., Kouamé, D., Achim, A., & Basarab, A.
In
IEEE International Ultrasonics Symposium, Glasgow, 06/10/2019-09/10/2019, pages (on line), http://www.ieee.org/, 2019. IEEE : Institute of Electrical and Electronics Engineers
Paper
link
bibtex
abstract
@InProceedings{ KiMaKoAcBa2019.1,
author = {Kim, Jong-Hoon and Mamou, Jonathan and Kouam\'e, Denis and Achim, Alin and Basarab, Adrian},
title = "{Spatio-temporal compressed quantitative acoustic microscopy (regular paper)}",
booktitle = "{IEEE International Ultrasonics Symposium, Glasgow, 06/10/2019-09/10/2019}",
year = {2019},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
pages = {(on line)},
language = {anglais},
URL = {https://doi.org/10.1109/ULTSYM.2019.8925562 - https://oatao.univ-toulouse.fr/26246/},
abstract = {This study proposes an elegant spatio-temporal compressed sensing scheme to significantly reduce the amount of data required to form quantitative acoustic microscopy (QAM) images. QAM systems form two-dimensional
acoustic parameter maps of thin section of soft tissues. QAM data collection consists in raster scanning a sample in 2D and digitizing backscattered RF signals at each scan location. Therefore, the raw QAM data is
three-dimensional and when using this conventional data acquisition process, data sets can be large causing processing and storage limitations. Our previous work demonstrated that the amount of QAM data can be remarkably reduced
either spatially or temporally by using compressive sampling (CS) or finite rate of innovation (FRI) approaches, respectively. These approaches take advantage of the properties of QAM data, i.e., the sparsity of 2D maps and the
parametric representation of RF signals. Therefore, in this study both approaches were combined into a single spatio-temporal solution. Results yielded a new data volume size of only 2.6% of the data originated by classical
sampling techniques without significant deterioration of the 2D maps.}
}
This study proposes an elegant spatio-temporal compressed sensing scheme to significantly reduce the amount of data required to form quantitative acoustic microscopy (QAM) images. QAM systems form two-dimensional acoustic parameter maps of thin section of soft tissues. QAM data collection consists in raster scanning a sample in 2D and digitizing backscattered RF signals at each scan location. Therefore, the raw QAM data is three-dimensional and when using this conventional data acquisition process, data sets can be large causing processing and storage limitations. Our previous work demonstrated that the amount of QAM data can be remarkably reduced either spatially or temporally by using compressive sampling (CS) or finite rate of innovation (FRI) approaches, respectively. These approaches take advantage of the properties of QAM data, i.e., the sparsity of 2D maps and the parametric representation of RF signals. Therefore, in this study both approaches were combined into a single spatio-temporal solution. Results yielded a new data volume size of only 2.6% of the data originated by classical sampling techniques without significant deterioration of the 2D maps.
Joint Deconvolution of Fundamental and Harmonic Ultrasound Images (regular paper).
Hourani, M., Basarab, A., Kouamé, D., & Tourneret, J.
In
IEEE International Ultrasonics Symposium, Glasgow, 06/10/2019-09/10/2019, pages (on line), http://www.ieee.org/, 2019. IEEE : Institute of Electrical and Electronics Engineers
Paper
link
bibtex
abstract
@InProceedings{ HoBaKoTo2019.1,
author = {Hourani, Mohamad and Basarab, Adrian and Kouam\'e, Denis and Tourneret, Jean-Yves},
title = "{Joint Deconvolution of Fundamental and Harmonic Ultrasound Images (regular paper)}",
booktitle = "{IEEE International Ultrasonics Symposium, Glasgow, 06/10/2019-09/10/2019}",
year = {2019},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
pages = {(on line)},
language = {anglais},
URL = {https://doi.org/10.1109/ULTSYM.2019.8925657 - https://oatao.univ-toulouse.fr/26245/},
abstract = {This paper studies the interest of using harmonic ultrasound (US) images in the process of tissue reflectivity function restoration from RF data. To this end, two direct models (one for fundamental and another for
harmonique images) derived from the equation of US wave propagation are proposed. In particular, an axially varying attenuation matrix is used within the harmonic image model in order to account for the attenuation of harmonic
echoes. Based on these two image formation models, a joint deconvolution problem is investigated. The solution of this problem is obtained by minimizing a cost function composed of two data fidelity terms representing the linear
and non-linear model components,regularized by an l 1 -norm regularization. The tissue reflectivity function minimizing this function is finally determined using an alternating direction method of multipliers. The performance of
the proposed algorithm is quantitatively and qualitatively evaluated on synthetic data, and compared with a classical restoration method used for US images.}
}
This paper studies the interest of using harmonic ultrasound (US) images in the process of tissue reflectivity function restoration from RF data. To this end, two direct models (one for fundamental and another for harmonique images) derived from the equation of US wave propagation are proposed. In particular, an axially varying attenuation matrix is used within the harmonic image model in order to account for the attenuation of harmonic echoes. Based on these two image formation models, a joint deconvolution problem is investigated. The solution of this problem is obtained by minimizing a cost function composed of two data fidelity terms representing the linear and non-linear model components,regularized by an l 1 -norm regularization. The tissue reflectivity function minimizing this function is finally determined using an alternating direction method of multipliers. The performance of the proposed algorithm is quantitatively and qualitatively evaluated on synthetic data, and compared with a classical restoration method used for US images.
Magnetic Resonance and Ultrasound Image Fusion Using a PALM Algorithm (regular paper).
El Mansouri, O., Basarab, A., Vidal, F., Kouamé, D., & Tourneret, J.
In
Workshop on Signal Processing with Adaptative Sparse Structured Representations (SPARS 2019), Toulouse, 01/07/2019-04/07/2019, pages (on line), http://www.inp-toulouse.fr, 2019. INPT : Institut National Polytechnique de Toulouse
Paper
link
bibtex
abstract
@InProceedings{ ElBaViKoTo2019.1,
author = {El Mansouri, OuMayma and Basarab, Adrian and Vidal, Fabien and Kouam\'e, Denis and Tourneret, Jean-Yves},
title = "{Magnetic Resonance and Ultrasound Image Fusion Using a PALM Algorithm (regular paper)}",
booktitle = "{Workshop on Signal Processing with Adaptative Sparse Structured Representations (SPARS 2019), Toulouse, 01/07/2019-04/07/2019}",
year = {2019},
publisher = {INPT : Institut National Polytechnique de Toulouse},
address = {http://www.inp-toulouse.fr},
pages = {(on line)},
language = {anglais},
URL = {https://www.irit.fr/~Adrian.Basarab/img/Spars2019_OeM.pdf - https://oatao.univ-toulouse.fr/26243/},
abstract = {This paper studies a new fusion algorithm for magnetic resonance (MR) and ultrasound (US) images combining two inverse problems for MR image super-resolution and US image despeckling. A polynomial function is used to
link the gray levels of the two imaging modalities. Qualitative and quantitative evaluations on experimental phantom data show the interest of the proposed algorithm. The fused image is shown to take advantage of both the good
contrast and high signal to noise ratio of the MR image and the good spatial resolution of the US image.}
}
This paper studies a new fusion algorithm for magnetic resonance (MR) and ultrasound (US) images combining two inverse problems for MR image super-resolution and US image despeckling. A polynomial function is used to link the gray levels of the two imaging modalities. Qualitative and quantitative evaluations on experimental phantom data show the interest of the proposed algorithm. The fused image is shown to take advantage of both the good contrast and high signal to noise ratio of the MR image and the good spatial resolution of the US image.
Tissue Reflectivity Function Restoration from Fundamental and Harmonic Ultrasound Images (regular paper).
Hourani, M., Basarab, A., Kouamé, D., & Tourneret, J.
In
Workshop on Signal Processing with Adaptative Sparse Structured Representations (SPARS 2019), Toulouse, 01/07/2019-04/07/2019, pages (on line), http://www.inp-toulouse.fr, 2019. INPT : Institut National Polytechnique de Toulouse
Paper
link
bibtex
abstract
@InProceedings{ Ho2019.1,
author = {Hourani, Mohamad and Basarab, Adrian and Kouam\'e, Denis and Tourneret, Jean-Yves},
title = "{Tissue Reflectivity Function Restoration from Fundamental and Harmonic Ultrasound Images (regular paper)}",
booktitle = "{Workshop on Signal Processing with Adaptative Sparse Structured Representations (SPARS 2019), Toulouse, 01/07/2019-04/07/2019}",
year = {2019},
publisher = {INPT : Institut National Polytechnique de Toulouse},
address = {http://www.inp-toulouse.fr},
pages = {(on line)},
language = {anglais},
URL = {https://www.irit.fr/~Adrian.Basarab/img/Spars2019_MH.pdf - https://oatao.univ-toulouse.fr/26242/},
abstract = {This paper addresses the problem of ultrasound (US) image restoration. In contrast to most of the existing approaches that only take into account fundamental radiofrequency (RF) data, the proposed method also considers
harmonic US images. An algorithm based on the alternating direction of multipliers method (ADMM) is proposed to solve the joint deconvolution problem. Simulation results show the interest of the proposed approach when compared to
classical US image restoration schemes based only on fundamental data.}
}
This paper addresses the problem of ultrasound (US) image restoration. In contrast to most of the existing approaches that only take into account fundamental radiofrequency (RF) data, the proposed method also considers harmonic US images. An algorithm based on the alternating direction of multipliers method (ADMM) is proposed to solve the joint deconvolution problem. Simulation results show the interest of the proposed approach when compared to classical US image restoration schemes based only on fundamental data.
Tensor-Factorization-Based 3D Single Image Super-Resolution with Semi-Blind Point Spread Function Estimation (regular paper).
Hatvani, J., Basarab, A., Michetti, J., Gyongy, M., & Kouamé, D.
In
IEEE International Conference on Image Processing (ICIP 2019), Taipei, Taiwan, 22/09/19-25/09/19, pages (on line), http://www.ieee.org/, 2019. IEEE : Institute of Electrical and Electronics Engineers
Paper
link
bibtex
abstract
@InProceedings{ HaBaMiGyKo2019.1,
author = {Hatvani, Janka and Basarab, Adrian and Michetti, J\'er�me and Gyongy, Mikl�s and Kouam\'e, Denis},
title = "{Tensor-Factorization-Based 3D Single Image Super-Resolution with Semi-Blind Point Spread Function Estimation (regular paper)}",
booktitle = "{IEEE International Conference on Image Processing (ICIP 2019), Taipei, Taiwan, 22/09/19-25/09/19}",
year = {2019},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
pages = {(on line)},
language = {anglais},
URL = {https://doi.org/10.1109/ICIP.2019.8803354 - https://oatao.univ-toulouse.fr/26210/},
abstract = {A volumetric non-blind single image super-resolution technique using tensor factorization has been recently introduced by our group. That method allowed a 2-order-of-magnitude faster high-resolution image
reconstruction with equivalent image quality compared to state-of-the-art algorithms. In this work a joint alternating recovery of the high-resolution image and of the unknown point spread function parameters is proposed. The
method is evaluated on dental computed tomography images. The algorithm was compared to an existing 3D super-resolution method using low-rank and total variation regularization, combined with the same alternating PSF-optimization.
The two algorithms have shown similar improvement in PSNR, but our method converged roughly 40 times faster, under 6 minutes both in simulation and on experimental dental computed tomography data.}
}
A volumetric non-blind single image super-resolution technique using tensor factorization has been recently introduced by our group. That method allowed a 2-order-of-magnitude faster high-resolution image reconstruction with equivalent image quality compared to state-of-the-art algorithms. In this work a joint alternating recovery of the high-resolution image and of the unknown point spread function parameters is proposed. The method is evaluated on dental computed tomography images. The algorithm was compared to an existing 3D super-resolution method using low-rank and total variation regularization, combined with the same alternating PSF-optimization. The two algorithms have shown similar improvement in PSNR, but our method converged roughly 40 times faster, under 6 minutes both in simulation and on experimental dental computed tomography data.
On Single-Image Super-Resolution in 3D Brain Magnetic Resonance Imaging (regular paper).
Bazzi, F., Mescam, M., Basarab, A., & Kouamé, D.
In
IEEE International Engineering in Medicine and Biology Conference (EMBC 2019), Berlin, 23/07/2019-27/07/2019, pages (on line), http://www.ieee.org/, July 2019. IEEE : Institute of Electrical and Electronics Engineers
Paper
link
bibtex
abstract
@InProceedings{ Ba2019.5,
author = {Bazzi, Farah and Mescam, Muriel and Basarab, Adrian and Kouam\'e, Denis},
title = "{On Single-Image Super-Resolution in 3D Brain Magnetic Resonance Imaging (regular paper)}",
booktitle = "{IEEE International Engineering in Medicine and Biology Conference (EMBC 2019), Berlin, 23/07/2019-27/07/2019}",
year = {2019},
month = {July},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
pages = {(on line)},
language = {anglais},
URL = {https://doi.org/10.1109/EMBC.2019.8857959 - https://oatao.univ-toulouse.fr/26199/},
abstract = {The objective of this work is to apply 3D super resolution (SR) techniques to brain magnetic resonance (MR) image restoration. Two 3D SR methods are considered following different trends: one recently proposed
tensor-based approach and one inverse problem algorithm based on total variation and low rank regularization. The evaluation of their effectiveness is assessed through the segmentation of brain compartments: gray matter, white
matter and cerebrospinal fluid. The two algorithms are qualitatively and quantitatively evaluated on simulated images with ground truth available and on experimental data. The originality of this work is to consider the SR methods
as an initial step towards the final segmentation task. The results show the ability of both methods to overcome the loss of spatial resolution and to facilitate the segmentation of brain structures with improved accuracy compared
to native low-resolution MR images. Both algorithms achieved almost equivalent results with a highly reduced computational time cost for the tensor-based approach.}
}
The objective of this work is to apply 3D super resolution (SR) techniques to brain magnetic resonance (MR) image restoration. Two 3D SR methods are considered following different trends: one recently proposed tensor-based approach and one inverse problem algorithm based on total variation and low rank regularization. The evaluation of their effectiveness is assessed through the segmentation of brain compartments: gray matter, white matter and cerebrospinal fluid. The two algorithms are qualitatively and quantitatively evaluated on simulated images with ground truth available and on experimental data. The originality of this work is to consider the SR methods as an initial step towards the final segmentation task. The results show the ability of both methods to overcome the loss of spatial resolution and to facilitate the segmentation of brain structures with improved accuracy compared to native low-resolution MR images. Both algorithms achieved almost equivalent results with a highly reduced computational time cost for the tensor-based approach.
A convolutional neural network for 250-MHz quantitative acoustic-microscopy resolution enhancement (regular paper).
Mamou, J., Pellegrini, T., Kouamé, D., & Basarab, A.
In
IEEE International Engineering in Medicine and Biology Conference (EMBC 2019), Berlin, 23/07/2019-27/07/2019, pages (on line), http://www.ieee.org/, July 2019. IEEE : Institute of Electrical and Electronics Engineers
Paper
link
bibtex
abstract
@InProceedings{ Ma2019.1,
author = {Mamou, Jonathan and Pellegrini, Thomas and Kouam\'e, Denis and Basarab, Adrian},
title = "{A convolutional neural network for 250-MHz quantitative acoustic-microscopy resolution enhancement (regular paper)}",
booktitle = "{IEEE International Engineering in Medicine and Biology Conference (EMBC 2019), Berlin, 23/07/2019-27/07/2019}",
year = {2019},
month = {July},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
pages = {(on line)},
language = {anglais},
URL = {https://doi.org/10.1109/EMBC.2019.8857865 - https://oatao.univ-toulouse.fr/26198/},
abstract = {Quantitative acoustic microscopy (QAM) permits the formation of quantitative two-dimensional (2D) maps of acoustic and mechanical properties of soft tissues at microscopic resolution. The 2D maps formed using our
custom SAM systems employing a 250-MHz and a 500-MHz single-element transducer have a nominal resolution of 7$\mu$m and 4$\mu$m, respectively. In a previous study, the potential of single-image super-resolution (SR) image post-processing
to enhance the spatial resolution of 2D SAM maps was demonstrated using a forward model accounting for blur, decimation, and noise. However, results obtained when the SR method was applied to soft tissue data were not entirely
satisfactory because of the limitation of the convolution model considered and by the difficulty of estimating the system point spread function and designing the appropriate regularization term. Therefore, in this study, a machine
learning approach based on convolutional neural networks was implemented. For training, data acquired on the same samples at 250 and 500 MHz were used. The resulting trained network was tested on 2D impedance maps (2DZMs) of human
lymph nodes acquired from breast-cancer patients. Visual inspection of the reconstructed enhanced 2DZMs were found similar to the 2DZMs obtained at 500 MHz which were used as ground truth. In addition, the enhanced 250-MHz 2DZMs
obtained from the proposed method yielded better peak signal to noise ratio and normalized mean square error than those obtained with the previous SR method. This improvement was also demonstrated by the statistical analyses. This
pioneering work could significantly reduce challenges and costs associated with current very high-frequency SAM systems while providing enhanced spatial resolution.}
}
Quantitative acoustic microscopy (QAM) permits the formation of quantitative two-dimensional (2D) maps of acoustic and mechanical properties of soft tissues at microscopic resolution. The 2D maps formed using our custom SAM systems employing a 250-MHz and a 500-MHz single-element transducer have a nominal resolution of 7$μ$m and 4$μ$m, respectively. In a previous study, the potential of single-image super-resolution (SR) image post-processing to enhance the spatial resolution of 2D SAM maps was demonstrated using a forward model accounting for blur, decimation, and noise. However, results obtained when the SR method was applied to soft tissue data were not entirely satisfactory because of the limitation of the convolution model considered and by the difficulty of estimating the system point spread function and designing the appropriate regularization term. Therefore, in this study, a machine learning approach based on convolutional neural networks was implemented. For training, data acquired on the same samples at 250 and 500 MHz were used. The resulting trained network was tested on 2D impedance maps (2DZMs) of human lymph nodes acquired from breast-cancer patients. Visual inspection of the reconstructed enhanced 2DZMs were found similar to the 2DZMs obtained at 500 MHz which were used as ground truth. In addition, the enhanced 250-MHz 2DZMs obtained from the proposed method yielded better peak signal to noise ratio and normalized mean square error than those obtained with the previous SR method. This improvement was also demonstrated by the statistical analyses. This pioneering work could significantly reduce challenges and costs associated with current very high-frequency SAM systems while providing enhanced spatial resolution.
On Multifractal Tissue Characterization in Ultrasound Imaging (regular paper).
Villain, E., Wendt, H., Basarab, A., & Kouamé, D.
In
IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Venice, Italy, 08/04/2019-11/04/2019, pages (electronic medium), http://www.ieee.org/, April 2019. IEEE : Institute of Electrical and Electronics Engineers
Paper
link
bibtex
@InProceedings{ ViWeBaKo2019.1,
author = {Villain, Edouard and Wendt, Herwig and Basarab, Adrian and Kouam\'e, Denis},
title = "{On Multifractal Tissue Characterization in Ultrasound Imaging (regular paper)}",
booktitle = "{IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Venice, Italy, 08/04/2019-11/04/2019}",
year = {2019},
month = {April},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
pages = {(electronic medium)},
language = {anglais},
URL = {https://oatao.univ-toulouse.fr/24731/}
}
Iterative Reconstruction of Medical Ultrasound Images Using Spectrally Constrained Phase Updates (regular paper).
Michailovich, O., Basarab, A., & Kouamé, D.
In
IEEE International Symposium on Biomedical Imaging: From Nano to Macro (ISBI 2019), Venice, Italy, 08/04/2019-11/04/2019, pages (electronic medium), http://www.ieee.org/, 2019. IEEE : Institute of Electrical and Electronics Engineers
Paper
link
bibtex
abstract
@InProceedings{ MiBaKo2019.1,
author = {Michailovich, Oleg and Basarab, Adrian and Kouam\'e, Denis},
title = "{Iterative Reconstruction of Medical Ultrasound Images Using Spectrally Constrained Phase Updates (regular paper)}",
booktitle = "{IEEE International Symposium on Biomedical Imaging: From Nano to Macro (ISBI 2019), Venice, Italy, 08/04/2019-11/04/2019}",
year = {2019},
publisher = {IEEE : Institute of Electrical and Electronics Engineers},
address = {http://www.ieee.org/},
pages = {(electronic medium)},
language = {anglais},
URL = {https://doi.org/10.1109/ISBI.2019.8759245 - https://oatao.univ-toulouse.fr/26162/},
abstract = {Image deconvolution is a standard numerical procedure used in medical ultrasound imaging for improving the resolution and contrast of diagnostic sonograms. However, due to the intrinsic bandlimitedness of ultrasound
scanners and the adverse effect of measurement noises, image deconvolution is known to be exceedingly sensitive to the errors incurred during inference of the point spread function (PSF) that characterizes the imaging system in
use. In this case, even the slightest errors in specification of the PSF are likely to result in significant artifacts, rendering the reconstructed images worthless. To address the aforementioned problem, this paper describes a
new method for blind deconvolution of ultrasound images, in which the errors due to inaccuracies in specification of the PSF are eliminated concurrently with estimation of tissue reflectivity directly from its associated
radio-frequency data. A derivation and justification of the proposed method are supported by experimental results which demonstrate the effectiveness and viability of the new technique.}
}
Image deconvolution is a standard numerical procedure used in medical ultrasound imaging for improving the resolution and contrast of diagnostic sonograms. However, due to the intrinsic bandlimitedness of ultrasound scanners and the adverse effect of measurement noises, image deconvolution is known to be exceedingly sensitive to the errors incurred during inference of the point spread function (PSF) that characterizes the imaging system in use. In this case, even the slightest errors in specification of the PSF are likely to result in significant artifacts, rendering the reconstructed images worthless. To address the aforementioned problem, this paper describes a new method for blind deconvolution of ultrasound images, in which the errors due to inaccuracies in specification of the PSF are eliminated concurrently with estimation of tissue reflectivity directly from its associated radio-frequency data. A derivation and justification of the proposed method are supported by experimental results which demonstrate the effectiveness and viability of the new technique.
Fusion of Magnetic Resonance and Ultrasound Images: a Preliminar Study on Simulated Data (regular paper).
El Mansouri, O., Basarab, A., Vidal, F., Kouamé, D., & Tourneret, J.
In
IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Venice, Italy, 08/04/19-11/04/19, 2019.
link
bibtex
@InProceedings{ El2019.1,
author = {El Mansouri, OuMayma and Basarab, Adrian and Vidal, Fabien and Kouam\'e, Denis and Tourneret, Jean-Yves},
title = "{Fusion of Magnetic Resonance and Ultrasound Images: a Preliminar Study on Simulated Data (regular paper)}",
booktitle = "{IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Venice, Italy, 08/04/19-11/04/19}",
year = {2019},
to_appear = {to appear},
publisher = {},
pages = {},
language = {anglais}
}