Unrolled Optimization via Physics-Assisted Convolutional Neural Network for MR-Based Electrical Properties Tomography: A Numerical Investigation
Magnetic Resonance imaging based Electrical Properties Tomography (MR-EPT) is a non-invasive technique that measures the electrical properties (EPs) of biological tissues. In this work, we present and numerically investigate the performance of an unrolled, physics-assisted method for 2D MR-EPT recon...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Open Journal of Engineering in Medicine and Biology |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10534835/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832582330134298624 |
|---|---|
| author | Sabrina Zumbo Stefano Mandija Ettore F. Meliado Peter Stijnman Thierry G. Meerbothe Cornelis A.T. van den Berg Tommaso Isernia Martina T. Bevacqua |
| author_facet | Sabrina Zumbo Stefano Mandija Ettore F. Meliado Peter Stijnman Thierry G. Meerbothe Cornelis A.T. van den Berg Tommaso Isernia Martina T. Bevacqua |
| author_sort | Sabrina Zumbo |
| collection | DOAJ |
| description | Magnetic Resonance imaging based Electrical Properties Tomography (MR-EPT) is a non-invasive technique that measures the electrical properties (EPs) of biological tissues. In this work, we present and numerically investigate the performance of an unrolled, physics-assisted method for 2D MR-EPT reconstructions, where a cascade of Convolutional Neural Networks is used to compute the contrast update. Each network takes in input the EPs and the gradient descent direction (encoding the physics underlying the adopted scattering model) and returns as output the updated contrast function. The network is trained and tested in silico using 2D slices of realistic brain models at 128 MHz. Results show the capability of the proposed procedure to reconstruct EPs maps with quality comparable to that of the popular Contrast Source Inversion-EPT, while significantly reducing the computational time. |
| format | Article |
| id | doaj-art-b2bccd4ee1b74e24a32c9dbc86bac947 |
| institution | Kabale University |
| issn | 2644-1276 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Engineering in Medicine and Biology |
| spelling | doaj-art-b2bccd4ee1b74e24a32c9dbc86bac9472025-01-30T00:03:53ZengIEEEIEEE Open Journal of Engineering in Medicine and Biology2644-12762024-01-01550551310.1109/OJEMB.2024.340299810534835Unrolled Optimization via Physics-Assisted Convolutional Neural Network for MR-Based Electrical Properties Tomography: A Numerical InvestigationSabrina Zumbo0Stefano Mandija1https://orcid.org/0000-0002-4612-5509Ettore F. Meliado2https://orcid.org/0000-0003-1240-3141Peter Stijnman3https://orcid.org/0000-0001-8277-1420Thierry G. Meerbothe4https://orcid.org/0009-0009-5736-1038Cornelis A.T. van den Berg5https://orcid.org/0000-0002-5565-6889Tommaso Isernia6https://orcid.org/0000-0003-3830-9540Martina T. Bevacqua7https://orcid.org/0000-0001-6557-1283Department DIIES, Università Mediterranea di Reggio Calabria, Reggio Calabria, ItalyDepartment of Radiotherapy, Division of Imaging & Oncology, University Medical Center Utrecht, Utrecht, The NetherlandsComputational Imaging Group for MR Diagnostics & Therapy, Center for Image Sciences, Utrecht University, Utrecht, The NetherlandsDepartment of Radiotherapy, Division of Imaging & Oncology, University Medical Center Utrecht, Utrecht, The NetherlandsDepartment of Radiotherapy, Division of Imaging & Oncology, University Medical Center Utrecht, Utrecht, The NetherlandsDepartment of Radiotherapy, Division of Imaging & Oncology, University Medical Center Utrecht, Utrecht, The NetherlandsDepartment DIIES, Università Mediterranea di Reggio Calabria, Reggio Calabria, ItalyDepartment DIIES, Università Mediterranea di Reggio Calabria, Reggio Calabria, ItalyMagnetic Resonance imaging based Electrical Properties Tomography (MR-EPT) is a non-invasive technique that measures the electrical properties (EPs) of biological tissues. In this work, we present and numerically investigate the performance of an unrolled, physics-assisted method for 2D MR-EPT reconstructions, where a cascade of Convolutional Neural Networks is used to compute the contrast update. Each network takes in input the EPs and the gradient descent direction (encoding the physics underlying the adopted scattering model) and returns as output the updated contrast function. The network is trained and tested in silico using 2D slices of realistic brain models at 128 MHz. Results show the capability of the proposed procedure to reconstruct EPs maps with quality comparable to that of the popular Contrast Source Inversion-EPT, while significantly reducing the computational time.https://ieeexplore.ieee.org/document/10534835/Convolutional neural networkelectrical propertiesinverse scattering problemslearning methodsmagnetic resonance imaging |
| spellingShingle | Sabrina Zumbo Stefano Mandija Ettore F. Meliado Peter Stijnman Thierry G. Meerbothe Cornelis A.T. van den Berg Tommaso Isernia Martina T. Bevacqua Unrolled Optimization via Physics-Assisted Convolutional Neural Network for MR-Based Electrical Properties Tomography: A Numerical Investigation IEEE Open Journal of Engineering in Medicine and Biology Convolutional neural network electrical properties inverse scattering problems learning methods magnetic resonance imaging |
| title | Unrolled Optimization via Physics-Assisted Convolutional Neural Network for MR-Based Electrical Properties Tomography: A Numerical Investigation |
| title_full | Unrolled Optimization via Physics-Assisted Convolutional Neural Network for MR-Based Electrical Properties Tomography: A Numerical Investigation |
| title_fullStr | Unrolled Optimization via Physics-Assisted Convolutional Neural Network for MR-Based Electrical Properties Tomography: A Numerical Investigation |
| title_full_unstemmed | Unrolled Optimization via Physics-Assisted Convolutional Neural Network for MR-Based Electrical Properties Tomography: A Numerical Investigation |
| title_short | Unrolled Optimization via Physics-Assisted Convolutional Neural Network for MR-Based Electrical Properties Tomography: A Numerical Investigation |
| title_sort | unrolled optimization via physics assisted convolutional neural network for mr based electrical properties tomography a numerical investigation |
| topic | Convolutional neural network electrical properties inverse scattering problems learning methods magnetic resonance imaging |
| url | https://ieeexplore.ieee.org/document/10534835/ |
| work_keys_str_mv | AT sabrinazumbo unrolledoptimizationviaphysicsassistedconvolutionalneuralnetworkformrbasedelectricalpropertiestomographyanumericalinvestigation AT stefanomandija unrolledoptimizationviaphysicsassistedconvolutionalneuralnetworkformrbasedelectricalpropertiestomographyanumericalinvestigation AT ettorefmeliado unrolledoptimizationviaphysicsassistedconvolutionalneuralnetworkformrbasedelectricalpropertiestomographyanumericalinvestigation AT peterstijnman unrolledoptimizationviaphysicsassistedconvolutionalneuralnetworkformrbasedelectricalpropertiestomographyanumericalinvestigation AT thierrygmeerbothe unrolledoptimizationviaphysicsassistedconvolutionalneuralnetworkformrbasedelectricalpropertiestomographyanumericalinvestigation AT cornelisatvandenberg unrolledoptimizationviaphysicsassistedconvolutionalneuralnetworkformrbasedelectricalpropertiestomographyanumericalinvestigation AT tommasoisernia unrolledoptimizationviaphysicsassistedconvolutionalneuralnetworkformrbasedelectricalpropertiestomographyanumericalinvestigation AT martinatbevacqua unrolledoptimizationviaphysicsassistedconvolutionalneuralnetworkformrbasedelectricalpropertiestomographyanumericalinvestigation |