Semantic similarity metrics for image registration
Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Standard
Semantic similarity metrics for image registration. / Czolbe, Steffen; Pegios, Paraskevas; Krause, Oswin; Feragen, Aasa.
I: Medical Image Analysis, Bind 87, 102830, 2023.Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Semantic similarity metrics for image registration
AU - Czolbe, Steffen
AU - Pegios, Paraskevas
AU - Krause, Oswin
AU - Feragen, Aasa
N1 - Publisher Copyright: © 2023 The Author(s)
PY - 2023
Y1 - 2023
N2 - Image registration aims to find geometric transformations that align images. Most algorithmic and deep learning-based methods solve the registration problem by minimizing a loss function, consisting of a similarity metric comparing the aligned images, and a regularization term ensuring smoothness of the transformation. Existing similarity metrics like Euclidean Distance or Normalized Cross-Correlation focus on aligning pixel intensity values or correlations, giving difficulties with low intensity contrast, noise, and ambiguous matching. We propose a semantic similarity metric for image registration, focusing on aligning image areas based on semantic correspondence instead. Our approach learns dataset-specific features that drive the optimization of a learning-based registration model. We train both an unsupervised approach extracting features with an auto-encoder, and a semi-supervised approach using supplemental segmentation data. We validate the semantic similarity metric using both deep-learning-based and algorithmic image registration methods. Compared to existing methods across four different image modalities and applications, the method achieves consistently high registration accuracy and smooth transformation fields.
AB - Image registration aims to find geometric transformations that align images. Most algorithmic and deep learning-based methods solve the registration problem by minimizing a loss function, consisting of a similarity metric comparing the aligned images, and a regularization term ensuring smoothness of the transformation. Existing similarity metrics like Euclidean Distance or Normalized Cross-Correlation focus on aligning pixel intensity values or correlations, giving difficulties with low intensity contrast, noise, and ambiguous matching. We propose a semantic similarity metric for image registration, focusing on aligning image areas based on semantic correspondence instead. Our approach learns dataset-specific features that drive the optimization of a learning-based registration model. We train both an unsupervised approach extracting features with an auto-encoder, and a semi-supervised approach using supplemental segmentation data. We validate the semantic similarity metric using both deep-learning-based and algorithmic image registration methods. Compared to existing methods across four different image modalities and applications, the method achieves consistently high registration accuracy and smooth transformation fields.
KW - Deep learning
KW - Image registration
KW - Representation learning
UR - http://www.scopus.com/inward/record.url?scp=85159141451&partnerID=8YFLogxK
U2 - 10.1016/j.media.2023.102830
DO - 10.1016/j.media.2023.102830
M3 - Journal article
C2 - 37172390
AN - SCOPUS:85159141451
VL - 87
JO - Medical Image Analysis
JF - Medical Image Analysis
SN - 1361-8415
M1 - 102830
ER -
ID: 347484754