Improving MRI Resolution: A Cycle Consistent Generative Adversarial Network-Based Approach for 3T to 7T Translation
<p dir="ltr">Brain magnetic resonance imaging (MRI) offers intricate soft tissue contrasts that are essential for diagnosing diseases and conducting neuroscience research. At 7 Tesla (7T) magnetic field intensity, MRI enables increased resolution, enhanced tissue contrast, and improv...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , , , , , , , , |
| Published: |
2024
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1864513541084545024 |
|---|---|
| author | Zakaria Shams Siam (22048001) |
| author2 | Rubyat Tasnuva Hasan (22048004) Moajjem Hossain Chowdhury (21842429) Md. Shaheenur Islam Sumon (22048007) Mamun Bin Ibne Reaz (641768) Sawal Hamid Bin Md Ali (22048010) Adam Mushtak (22048013) Israa Al-Hashimi (18131374) Sohaib Bassam Zoghoul (22048016) Muhammad E. H. Chowdhury (14150526) |
| author2_role | author author author author author author author author author |
| author_facet | Zakaria Shams Siam (22048001) Rubyat Tasnuva Hasan (22048004) Moajjem Hossain Chowdhury (21842429) Md. Shaheenur Islam Sumon (22048007) Mamun Bin Ibne Reaz (641768) Sawal Hamid Bin Md Ali (22048010) Adam Mushtak (22048013) Israa Al-Hashimi (18131374) Sohaib Bassam Zoghoul (22048016) Muhammad E. H. Chowdhury (14150526) |
| author_role | author |
| dc.creator.none.fl_str_mv | Zakaria Shams Siam (22048001) Rubyat Tasnuva Hasan (22048004) Moajjem Hossain Chowdhury (21842429) Md. Shaheenur Islam Sumon (22048007) Mamun Bin Ibne Reaz (641768) Sawal Hamid Bin Md Ali (22048010) Adam Mushtak (22048013) Israa Al-Hashimi (18131374) Sohaib Bassam Zoghoul (22048016) Muhammad E. H. Chowdhury (14150526) |
| dc.date.none.fl_str_mv | 2024-08-30T12:00:00Z |
| dc.identifier.none.fl_str_mv | 10.1109/access.2024.3430968 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/journal_contribution/Improving_MRI_Resolution_A_Cycle_Consistent_Generative_Adversarial_Network-Based_Approach_for_3T_to_7T_Translation/29900759 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Biomedical and clinical sciences Neurosciences Health sciences Health services and systems Information and computing sciences Artificial intelligence Cycle consistent generative adversarial network image-to-image translation magnetic resonance imaging paired dataset T1-weighted MRI Magnetic resonance imaging Training Image processing Spatial resolution Image quality Generative adversarial networks Superresolution Data models |
| dc.title.none.fl_str_mv | Improving MRI Resolution: A Cycle Consistent Generative Adversarial Network-Based Approach for 3T to 7T Translation |
| dc.type.none.fl_str_mv | Text Journal contribution info:eu-repo/semantics/publishedVersion text contribution to journal |
| description | <p dir="ltr">Brain magnetic resonance imaging (MRI) offers intricate soft tissue contrasts that are essential for diagnosing diseases and conducting neuroscience research. At 7 Tesla (7T) magnetic field intensity, MRI enables increased resolution, enhanced tissue contrast, and improved SNR, compared to MRI collected from the commonly employed 3 Tesla (3T) MRI scanners. However, the exorbitant expenses associated with 7T MRI scanners hinder their broad use in research and clinical facilities. Efforts are underway to develop algorithms that can generate 7T MRI from 3T MRI to achieve better image quality without the need for 7T MRI machines. In this study, we have adopted a cycle consistent generative adversarial network (CycleGAN)-based approach for 3T MRI to 7T MRI translation, and vice versa, using a recently published dataset of paired T1-weighted MR images collected at 3T and 7T from a total of ten subjects. Various CycleGAN architectures were experimented with and compared on this dataset. The best performing CycleGAN architecture successfully produced the reconstructed images with a high level of accuracy based on different quantitative and qualitative evaluation criteria. Utilizing a post-processing technique, the best performing model generated 7T MRI from 3T MRI with a structural similarity index measure (SSIM) of 83.80%, peak SNR (PSNR) of 26.25, normalized mean squared error (NMSE) of 0.0088 and normalized mean absolute error (NMAE) of 0.0630. Utilizing CycleGAN to convert images from 3T to 7T MRI has shown a substantial improvement in MRI resolution, setting the stage for advancements in more informative and precise diagnostic imaging.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/deed.en" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/access.2024.3430968" target="_blank">https://dx.doi.org/10.1109/access.2024.3430968</a></p> |
| eu_rights_str_mv | openAccess |
| id | Manara2_1b59cdcb891aa52d7943b5484ee021e0 |
| identifier_str_mv | 10.1109/access.2024.3430968 |
| network_acronym_str | Manara2 |
| network_name_str | Manara2 |
| oai_identifier_str | oai:figshare.com:article/29900759 |
| publishDate | 2024 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Improving MRI Resolution: A Cycle Consistent Generative Adversarial Network-Based Approach for 3T to 7T TranslationZakaria Shams Siam (22048001)Rubyat Tasnuva Hasan (22048004)Moajjem Hossain Chowdhury (21842429)Md. Shaheenur Islam Sumon (22048007)Mamun Bin Ibne Reaz (641768)Sawal Hamid Bin Md Ali (22048010)Adam Mushtak (22048013)Israa Al-Hashimi (18131374)Sohaib Bassam Zoghoul (22048016)Muhammad E. H. Chowdhury (14150526)Biomedical and clinical sciencesNeurosciencesHealth sciencesHealth services and systemsInformation and computing sciencesArtificial intelligenceCycle consistent generative adversarial networkimage-to-image translationmagnetic resonance imagingpaired datasetT1-weighted MRIMagnetic resonance imagingTrainingImage processingSpatial resolutionImage qualityGenerative adversarial networksSuperresolutionData models<p dir="ltr">Brain magnetic resonance imaging (MRI) offers intricate soft tissue contrasts that are essential for diagnosing diseases and conducting neuroscience research. At 7 Tesla (7T) magnetic field intensity, MRI enables increased resolution, enhanced tissue contrast, and improved SNR, compared to MRI collected from the commonly employed 3 Tesla (3T) MRI scanners. However, the exorbitant expenses associated with 7T MRI scanners hinder their broad use in research and clinical facilities. Efforts are underway to develop algorithms that can generate 7T MRI from 3T MRI to achieve better image quality without the need for 7T MRI machines. In this study, we have adopted a cycle consistent generative adversarial network (CycleGAN)-based approach for 3T MRI to 7T MRI translation, and vice versa, using a recently published dataset of paired T1-weighted MR images collected at 3T and 7T from a total of ten subjects. Various CycleGAN architectures were experimented with and compared on this dataset. The best performing CycleGAN architecture successfully produced the reconstructed images with a high level of accuracy based on different quantitative and qualitative evaluation criteria. Utilizing a post-processing technique, the best performing model generated 7T MRI from 3T MRI with a structural similarity index measure (SSIM) of 83.80%, peak SNR (PSNR) of 26.25, normalized mean squared error (NMSE) of 0.0088 and normalized mean absolute error (NMAE) of 0.0630. Utilizing CycleGAN to convert images from 3T to 7T MRI has shown a substantial improvement in MRI resolution, setting the stage for advancements in more informative and precise diagnostic imaging.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/deed.en" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/access.2024.3430968" target="_blank">https://dx.doi.org/10.1109/access.2024.3430968</a></p>2024-08-30T12:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/access.2024.3430968https://figshare.com/articles/journal_contribution/Improving_MRI_Resolution_A_Cycle_Consistent_Generative_Adversarial_Network-Based_Approach_for_3T_to_7T_Translation/29900759CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/299007592024-08-30T12:00:00Z |
| spellingShingle | Improving MRI Resolution: A Cycle Consistent Generative Adversarial Network-Based Approach for 3T to 7T Translation Zakaria Shams Siam (22048001) Biomedical and clinical sciences Neurosciences Health sciences Health services and systems Information and computing sciences Artificial intelligence Cycle consistent generative adversarial network image-to-image translation magnetic resonance imaging paired dataset T1-weighted MRI Magnetic resonance imaging Training Image processing Spatial resolution Image quality Generative adversarial networks Superresolution Data models |
| status_str | publishedVersion |
| title | Improving MRI Resolution: A Cycle Consistent Generative Adversarial Network-Based Approach for 3T to 7T Translation |
| title_full | Improving MRI Resolution: A Cycle Consistent Generative Adversarial Network-Based Approach for 3T to 7T Translation |
| title_fullStr | Improving MRI Resolution: A Cycle Consistent Generative Adversarial Network-Based Approach for 3T to 7T Translation |
| title_full_unstemmed | Improving MRI Resolution: A Cycle Consistent Generative Adversarial Network-Based Approach for 3T to 7T Translation |
| title_short | Improving MRI Resolution: A Cycle Consistent Generative Adversarial Network-Based Approach for 3T to 7T Translation |
| title_sort | Improving MRI Resolution: A Cycle Consistent Generative Adversarial Network-Based Approach for 3T to 7T Translation |
| topic | Biomedical and clinical sciences Neurosciences Health sciences Health services and systems Information and computing sciences Artificial intelligence Cycle consistent generative adversarial network image-to-image translation magnetic resonance imaging paired dataset T1-weighted MRI Magnetic resonance imaging Training Image processing Spatial resolution Image quality Generative adversarial networks Superresolution Data models |