Fighting MRI Anisotropy: Learning Multiple Cardiac Shapes From a Single Implicit Neural Representation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911442453659648 |
|---|---|
| author | Brás, Carolina Haddou, Soufiane Ben Kuipers, Thijs P. Alvarez-Florez, Laura Planken, R. Nils Tjong, Fleur V. Y. Bezzina, Connie Išgum, Ivana |
| author_facet | Brás, Carolina Haddou, Soufiane Ben Kuipers, Thijs P. Alvarez-Florez, Laura Planken, R. Nils Tjong, Fleur V. Y. Bezzina, Connie Išgum, Ivana |
| contents | The anisotropic nature of short-axis (SAX) cardiovascular magnetic resonance imaging (CMRI) limits cardiac shape analysis. To address this, we propose to leverage near-isotropic, higher resolution computed tomography angiography (CTA) data of the heart. We use this data to train a single neural implicit function to jointly represent cardiac shapes from CMRI at any resolution. We evaluate the method for the reconstruction of right ventricle (RV) and myocardium (MYO), where MYO simultaneously models endocardial and epicardial left-ventricle surfaces. Since high-resolution SAX reference segmentations are unavailable, we evaluate performance by extracting a 4-chamber (4CH) slice of RV and MYO from their reconstructed shapes. When compared with the reference 4CH segmentation masks from CMRI, our method achieved a Dice similarity coefficient of 0.91 $\pm$ 0.07 and 0.75 $\pm$ 0.13, and a Hausdorff distance of 6.21 $\pm$ 3.97 mm and 7.53 $\pm$ 5.13 mm for RV and MYO, respectively. Quantitative and qualitative assessment demonstrate the model's ability to reconstruct accurate, smooth and anatomically plausible shapes, supporting improvements in cardiac shape analysis. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_11436 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Fighting MRI Anisotropy: Learning Multiple Cardiac Shapes From a Single Implicit Neural Representation Brás, Carolina Haddou, Soufiane Ben Kuipers, Thijs P. Alvarez-Florez, Laura Planken, R. Nils Tjong, Fleur V. Y. Bezzina, Connie Išgum, Ivana Computer Vision and Pattern Recognition Artificial Intelligence The anisotropic nature of short-axis (SAX) cardiovascular magnetic resonance imaging (CMRI) limits cardiac shape analysis. To address this, we propose to leverage near-isotropic, higher resolution computed tomography angiography (CTA) data of the heart. We use this data to train a single neural implicit function to jointly represent cardiac shapes from CMRI at any resolution. We evaluate the method for the reconstruction of right ventricle (RV) and myocardium (MYO), where MYO simultaneously models endocardial and epicardial left-ventricle surfaces. Since high-resolution SAX reference segmentations are unavailable, we evaluate performance by extracting a 4-chamber (4CH) slice of RV and MYO from their reconstructed shapes. When compared with the reference 4CH segmentation masks from CMRI, our method achieved a Dice similarity coefficient of 0.91 $\pm$ 0.07 and 0.75 $\pm$ 0.13, and a Hausdorff distance of 6.21 $\pm$ 3.97 mm and 7.53 $\pm$ 5.13 mm for RV and MYO, respectively. Quantitative and qualitative assessment demonstrate the model's ability to reconstruct accurate, smooth and anatomically plausible shapes, supporting improvements in cardiac shape analysis. |
| title | Fighting MRI Anisotropy: Learning Multiple Cardiac Shapes From a Single Implicit Neural Representation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2602.11436 |