MMAR-Net: A Multi-Stride and Multi-Resolution Affine Registration Network for CT Images
The evolution of lung lesions can be assessed by examining multiple CT screenings, which needs to align two CT images accurately. In this study, we propose a multi-stride and multi-resolution affine registration network, called MMAR-net, for 3D affine registration of medical images, which works in a...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tsinghua University Press
2024-12-01
|
| Series: | Big Data Mining and Analytics |
| Subjects: | |
| Online Access: | https://www.sciopen.com/article/10.26599/BDMA.2024.9020005 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846100743713456128 |
|---|---|
| author | Fu Zhou Fei Luo Ruoshan Kong Yi-Ping Phoebe Chen Feng Liu |
| author_facet | Fu Zhou Fei Luo Ruoshan Kong Yi-Ping Phoebe Chen Feng Liu |
| author_sort | Fu Zhou |
| collection | DOAJ |
| description | The evolution of lung lesions can be assessed by examining multiple CT screenings, which needs to align two CT images accurately. In this study, we propose a multi-stride and multi-resolution affine registration network, called MMAR-net, for 3D affine registration of medical images, which works in an unsupervised way by optimizing the similarity loss. In order to extract more extensive image features, we use a multi-stride module to replace the conventional convolution module. Furthermore, we make use of the image features at multiple scales by dot product between two feature vectors, which could enhance the robustness of image representation. We conduct comprehensive comparison experiments between our model and the existing affine registration methods on two publicly available datasets, DIR-Lab and Learn2Reg, which are both relevant to lung CT image registration. Quantitative and qualitative comparison results demonstrate that our model outperforms existing single-step affine registration networks. Our method improves the key metric of dice similarity coefficient on DIR-Lab and Learn2Reg to 90.57% and 95.51%, respectively. |
| format | Article |
| id | doaj-art-3c2427ae2df14e93bb5e2f0aa29f1e20 |
| institution | Kabale University |
| issn | 2096-0654 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Tsinghua University Press |
| record_format | Article |
| series | Big Data Mining and Analytics |
| spelling | doaj-art-3c2427ae2df14e93bb5e2f0aa29f1e202024-12-29T15:36:22ZengTsinghua University PressBig Data Mining and Analytics2096-06542024-12-01741287130010.26599/BDMA.2024.9020005MMAR-Net: A Multi-Stride and Multi-Resolution Affine Registration Network for CT ImagesFu Zhou0Fei Luo1Ruoshan Kong2Yi-Ping Phoebe Chen3Feng Liu4School of Computer Science, Wuhan University, Wuhan 430000, ChinaSchool of Computer Science, Wuhan University, Wuhan 430000, ChinaSchool of Computer Science, Wuhan University, Wuhan 430000, ChinaDepartment of Computer Science and Information Technology, La Trobe University, Melbourne 3083, AustraliaSchool of Computer Science, Wuhan University, Wuhan 430000, ChinaThe evolution of lung lesions can be assessed by examining multiple CT screenings, which needs to align two CT images accurately. In this study, we propose a multi-stride and multi-resolution affine registration network, called MMAR-net, for 3D affine registration of medical images, which works in an unsupervised way by optimizing the similarity loss. In order to extract more extensive image features, we use a multi-stride module to replace the conventional convolution module. Furthermore, we make use of the image features at multiple scales by dot product between two feature vectors, which could enhance the robustness of image representation. We conduct comprehensive comparison experiments between our model and the existing affine registration methods on two publicly available datasets, DIR-Lab and Learn2Reg, which are both relevant to lung CT image registration. Quantitative and qualitative comparison results demonstrate that our model outperforms existing single-step affine registration networks. Our method improves the key metric of dice similarity coefficient on DIR-Lab and Learn2Reg to 90.57% and 95.51%, respectively.https://www.sciopen.com/article/10.26599/BDMA.2024.9020005lung lesionaffine registrationmulti-stridemulti-resolution |
| spellingShingle | Fu Zhou Fei Luo Ruoshan Kong Yi-Ping Phoebe Chen Feng Liu MMAR-Net: A Multi-Stride and Multi-Resolution Affine Registration Network for CT Images Big Data Mining and Analytics lung lesion affine registration multi-stride multi-resolution |
| title | MMAR-Net: A Multi-Stride and Multi-Resolution Affine Registration Network for CT Images |
| title_full | MMAR-Net: A Multi-Stride and Multi-Resolution Affine Registration Network for CT Images |
| title_fullStr | MMAR-Net: A Multi-Stride and Multi-Resolution Affine Registration Network for CT Images |
| title_full_unstemmed | MMAR-Net: A Multi-Stride and Multi-Resolution Affine Registration Network for CT Images |
| title_short | MMAR-Net: A Multi-Stride and Multi-Resolution Affine Registration Network for CT Images |
| title_sort | mmar net a multi stride and multi resolution affine registration network for ct images |
| topic | lung lesion affine registration multi-stride multi-resolution |
| url | https://www.sciopen.com/article/10.26599/BDMA.2024.9020005 |
| work_keys_str_mv | AT fuzhou mmarnetamultistrideandmultiresolutionaffineregistrationnetworkforctimages AT feiluo mmarnetamultistrideandmultiresolutionaffineregistrationnetworkforctimages AT ruoshankong mmarnetamultistrideandmultiresolutionaffineregistrationnetworkforctimages AT yipingphoebechen mmarnetamultistrideandmultiresolutionaffineregistrationnetworkforctimages AT fengliu mmarnetamultistrideandmultiresolutionaffineregistrationnetworkforctimages |