Segmenting Two-Dimensional Structures with Strided Tensor Networks
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Standard
Segmenting Two-Dimensional Structures with Strided Tensor Networks. / Selvan, Raghavendra; Dam, Erik B.; Petersen, Jens.
Information Processing in Medical Imaging - 27th International Conference, IPMI 2021, Proceedings. red. / Aasa Feragen; Stefan Sommer; Julia Schnabel; Mads Nielsen. Springer, 2021. s. 401-414 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Bind 12729 LNCS).Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Segmenting Two-Dimensional Structures with Strided Tensor Networks
AU - Selvan, Raghavendra
AU - Dam, Erik B.
AU - Petersen, Jens
N1 - Publisher Copyright: © 2021, Springer Nature Switzerland AG.
PY - 2021
Y1 - 2021
N2 - Tensor networks provide an efficient approximation of operations involving high dimensional tensors and have been extensively used in modelling quantum many-body systems. More recently, supervised learning has been attempted with tensor networks, primarily focused on tasks such as image classification. In this work, we propose a novel formulation of tensor networks for supervised image segmentation which allows them to operate on high resolution medical images. We use the matrix product state (MPS) tensor network on non-overlapping patches of a given input image to predict the segmentation mask by learning a pixel-wise linear classification rule in a high dimensional space. The proposed model is end-to-end trainable using backpropagation. It is implemented as a strided tensor network to reduce the parameter complexity. The performance of the proposed method is evaluated on two public medical imaging datasets and compared to relevant baselines. The evaluation shows that the strided tensor network yields competitive performance compared to CNN-based models while using fewer resources. Additionally, based on the experiments we discuss the feasibility of using fully linear models for segmentation tasks.(Source code: https://github.com/raghavian/strided-tenet )
AB - Tensor networks provide an efficient approximation of operations involving high dimensional tensors and have been extensively used in modelling quantum many-body systems. More recently, supervised learning has been attempted with tensor networks, primarily focused on tasks such as image classification. In this work, we propose a novel formulation of tensor networks for supervised image segmentation which allows them to operate on high resolution medical images. We use the matrix product state (MPS) tensor network on non-overlapping patches of a given input image to predict the segmentation mask by learning a pixel-wise linear classification rule in a high dimensional space. The proposed model is end-to-end trainable using backpropagation. It is implemented as a strided tensor network to reduce the parameter complexity. The performance of the proposed method is evaluated on two public medical imaging datasets and compared to relevant baselines. The evaluation shows that the strided tensor network yields competitive performance compared to CNN-based models while using fewer resources. Additionally, based on the experiments we discuss the feasibility of using fully linear models for segmentation tasks.(Source code: https://github.com/raghavian/strided-tenet )
KW - Image segmentation
KW - Linear models
KW - Tensor networks
U2 - 10.1007/978-3-030-78191-0_31
DO - 10.1007/978-3-030-78191-0_31
M3 - Article in proceedings
AN - SCOPUS:85111469288
SN - 9783030781903
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 401
EP - 414
BT - Information Processing in Medical Imaging - 27th International Conference, IPMI 2021, Proceedings
A2 - Feragen, Aasa
A2 - Sommer, Stefan
A2 - Schnabel, Julia
A2 - Nielsen, Mads
PB - Springer
T2 - 27th International Conference on Information Processing in Medical Imaging, IPMI 2021
Y2 - 28 June 2021 through 30 June 2021
ER -
ID: 282748194