A joint three-plane physics-constrained deep learning based polynomial fitting approach for MR electrical properties tomography
Magnetic resonance electrical properties tomography can extract the electrical properties of in-vivo tissue. To estimate tissue electrical properties, various reconstruction algorithms have been proposed. However, physics-based reconstructions are prone to various artifacts such as noise amplificati...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-02-01
|
| Series: | NeuroImage |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1053811925000564 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832087836267905024 |
|---|---|
| author | Kyu-Jin Jung Thierry G. Meerbothe Chuanjiang Cui Mina Park Cornelis A.T. van den Berg Stefano Mandija Dong-Hyun Kim |
| author_facet | Kyu-Jin Jung Thierry G. Meerbothe Chuanjiang Cui Mina Park Cornelis A.T. van den Berg Stefano Mandija Dong-Hyun Kim |
| author_sort | Kyu-Jin Jung |
| collection | DOAJ |
| description | Magnetic resonance electrical properties tomography can extract the electrical properties of in-vivo tissue. To estimate tissue electrical properties, various reconstruction algorithms have been proposed. However, physics-based reconstructions are prone to various artifacts such as noise amplification and boundary artifact. Deep learning-based approaches are robust to these artifacts but need extensive training datasets and suffer from generalization to unseen data. To address these issues, we introduce a joint three-plane physics-constrained deep learning framework for polynomial fitting MR-EPT by merging physics-based weighted polynomial fitting with deep learning. Within this framework, deep learning is used to discern the optimal polynomial fitting weights for a physics based polynomial fitting reconstruction on the complex B1+ data. For the prediction of optimal fitting coefficients, three neural networks were separately trained on simulated heterogeneous brain models to predict optimal polynomial weighting parameters in three orthogonal planes. Then, the network weights were jointly optimized to estimate the polynomial weights in each plane for a combined conductivity reconstruction. Based on this physics-constrained deep learning approach, we achieved an improvement of conductivity estimation accuracy in comparison to a single plane estimation and a reduction of computational load. The results demonstrate that the proposed method based on 3D data exhibits superior performance in comparison to conventional polynomial fitting methods in terms of capturing anatomical detail and homogeneity. Crucially, in-vivo application of the proposed method showed that the method generalizes well to in-vivo data, without introducing significant errors or artifacts. This generalization makes the presented method a promising candidate for use in clinical applications. |
| format | Article |
| id | doaj-art-ce7cc0c65a1545d28311843904eaa02f |
| institution | Kabale University |
| issn | 1095-9572 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | NeuroImage |
| spelling | doaj-art-ce7cc0c65a1545d28311843904eaa02f2025-02-06T05:11:09ZengElsevierNeuroImage1095-95722025-02-01307121054A joint three-plane physics-constrained deep learning based polynomial fitting approach for MR electrical properties tomographyKyu-Jin Jung0Thierry G. Meerbothe1Chuanjiang Cui2Mina Park3Cornelis A.T. van den Berg4Stefano Mandija5Dong-Hyun Kim6Department of Electrical and Electronic Engineering, Yonsei University, Seoul, Republic of KoreaComputational Imaging Group for MR Therapy and Diagnostics, University Medical Center Utrecht, Utrecht, the Netherlands; Department of Radiotherapy, University Medical Center Utrecht, Utrecht, the NetherlandsDepartment of Electrical and Electronic Engineering, Yonsei University, Seoul, Republic of KoreaDepartment of Radiology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of KoreaComputational Imaging Group for MR Therapy and Diagnostics, University Medical Center Utrecht, Utrecht, the Netherlands; Department of Radiotherapy, University Medical Center Utrecht, Utrecht, the NetherlandsComputational Imaging Group for MR Therapy and Diagnostics, University Medical Center Utrecht, Utrecht, the Netherlands; Department of Radiotherapy, University Medical Center Utrecht, Utrecht, the Netherlands; Correspondence author at: Computational Imaging Group for MR Therapy and Diagnostics, University Medical Center Utrecht, Utrecht, the Netherlands.Department of Electrical and Electronic Engineering, Yonsei University, Seoul, Republic of Korea; Correspondence author at: Department of Electrical and Electronic Engineering, Yonsei University, Seoul, Republic of Korea.Magnetic resonance electrical properties tomography can extract the electrical properties of in-vivo tissue. To estimate tissue electrical properties, various reconstruction algorithms have been proposed. However, physics-based reconstructions are prone to various artifacts such as noise amplification and boundary artifact. Deep learning-based approaches are robust to these artifacts but need extensive training datasets and suffer from generalization to unseen data. To address these issues, we introduce a joint three-plane physics-constrained deep learning framework for polynomial fitting MR-EPT by merging physics-based weighted polynomial fitting with deep learning. Within this framework, deep learning is used to discern the optimal polynomial fitting weights for a physics based polynomial fitting reconstruction on the complex B1+ data. For the prediction of optimal fitting coefficients, three neural networks were separately trained on simulated heterogeneous brain models to predict optimal polynomial weighting parameters in three orthogonal planes. Then, the network weights were jointly optimized to estimate the polynomial weights in each plane for a combined conductivity reconstruction. Based on this physics-constrained deep learning approach, we achieved an improvement of conductivity estimation accuracy in comparison to a single plane estimation and a reduction of computational load. The results demonstrate that the proposed method based on 3D data exhibits superior performance in comparison to conventional polynomial fitting methods in terms of capturing anatomical detail and homogeneity. Crucially, in-vivo application of the proposed method showed that the method generalizes well to in-vivo data, without introducing significant errors or artifacts. This generalization makes the presented method a promising candidate for use in clinical applications.http://www.sciencedirect.com/science/article/pii/S1053811925000564Electrical properties tomographyConductivity neuroimagingPhase-based conductivity reconstructionMR image synthetizationPhysics-constrained neural networkDeep learning |
| spellingShingle | Kyu-Jin Jung Thierry G. Meerbothe Chuanjiang Cui Mina Park Cornelis A.T. van den Berg Stefano Mandija Dong-Hyun Kim A joint three-plane physics-constrained deep learning based polynomial fitting approach for MR electrical properties tomography NeuroImage Electrical properties tomography Conductivity neuroimaging Phase-based conductivity reconstruction MR image synthetization Physics-constrained neural network Deep learning |
| title | A joint three-plane physics-constrained deep learning based polynomial fitting approach for MR electrical properties tomography |
| title_full | A joint three-plane physics-constrained deep learning based polynomial fitting approach for MR electrical properties tomography |
| title_fullStr | A joint three-plane physics-constrained deep learning based polynomial fitting approach for MR electrical properties tomography |
| title_full_unstemmed | A joint three-plane physics-constrained deep learning based polynomial fitting approach for MR electrical properties tomography |
| title_short | A joint three-plane physics-constrained deep learning based polynomial fitting approach for MR electrical properties tomography |
| title_sort | joint three plane physics constrained deep learning based polynomial fitting approach for mr electrical properties tomography |
| topic | Electrical properties tomography Conductivity neuroimaging Phase-based conductivity reconstruction MR image synthetization Physics-constrained neural network Deep learning |
| url | http://www.sciencedirect.com/science/article/pii/S1053811925000564 |
| work_keys_str_mv | AT kyujinjung ajointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT thierrygmeerbothe ajointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT chuanjiangcui ajointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT minapark ajointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT cornelisatvandenberg ajointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT stefanomandija ajointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT donghyunkim ajointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT kyujinjung jointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT thierrygmeerbothe jointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT chuanjiangcui jointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT minapark jointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT cornelisatvandenberg jointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT stefanomandija jointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography AT donghyunkim jointthreeplanephysicsconstraineddeeplearningbasedpolynomialfittingapproachformrelectricalpropertiestomography |