TY - GEN
T1 - Unsupervised super-resolution: Creating high-resolution medical images from low-resolution anisotropic examples
AU - Sander, Jorg
AU - de Vos, Bob D.
AU - Išgum, Ivana
N1 - Funding Information: This study was performed within the DLMedIA program (P15-26) funded by Dutch Technology Foundation with participation of PIE Medical Imaging. Publisher Copyright: © 2021 SPIE.
PY - 2021
Y1 - 2021
N2 - Although high resolution isotropic 3D medical images are desired in clinical practice, their acquisition is not always feasible. Instead, lower resolution images are upsampled to higher resolution using conventional interpolation methods. Sophisticated learning-based super-resolution approaches are frequently unavailable in clinical setting, because such methods require training with high-resolution isotropic examples. To address this issue, we propose a learning-based super-resolution approach that can be trained using solely anisotropic images, i.e. without high-resolution ground truth data. The method exploits the latent space, generated by autoencoders trained on anisotropic images, to increase spatial resolution in low-resolution images. The method was trained and evaluated using 100 publicly available cardiac cine MR scans from the Automated Cardiac Diagnosis Challenge (ACDC). The quantitative results show that the proposed method performs better than conventional interpolation methods. Furthermore, the qualitative results indicate that especially finer cardiac structures are synthesized with high quality. The method has the potential to be applied to other anatomies and modalities and can be easily applied to any 3D anisotropic medical image dataset.
AB - Although high resolution isotropic 3D medical images are desired in clinical practice, their acquisition is not always feasible. Instead, lower resolution images are upsampled to higher resolution using conventional interpolation methods. Sophisticated learning-based super-resolution approaches are frequently unavailable in clinical setting, because such methods require training with high-resolution isotropic examples. To address this issue, we propose a learning-based super-resolution approach that can be trained using solely anisotropic images, i.e. without high-resolution ground truth data. The method exploits the latent space, generated by autoencoders trained on anisotropic images, to increase spatial resolution in low-resolution images. The method was trained and evaluated using 100 publicly available cardiac cine MR scans from the Automated Cardiac Diagnosis Challenge (ACDC). The quantitative results show that the proposed method performs better than conventional interpolation methods. Furthermore, the qualitative results indicate that especially finer cardiac structures are synthesized with high quality. The method has the potential to be applied to other anatomies and modalities and can be easily applied to any 3D anisotropic medical image dataset.
KW - Autoencoder
KW - Cardiac MRI
KW - Image Super-Resolution
KW - Latent Space Interpolation
UR - http://www.scopus.com/inward/record.url?scp=85103625469&partnerID=8YFLogxK
U2 - https://doi.org/10.1117/12.2580412
DO - https://doi.org/10.1117/12.2580412
M3 - Conference contribution
SN - 9781510640214
VL - 11596
T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE
BT - Medical Imaging 2021
A2 - Isgum, Ivana
A2 - Landman, Bennett A.
A2 - Išgum, I.
A2 - Landman, B.A.
PB - SPIE
CY - Bellingham, WA
T2 - Medical Imaging 2021: Image Processing
Y2 - 15 February 2021 through 19 February 2021
ER -