TY - JOUR
T1 - High-level prior-based loss functions for medical image segmentation
T2 - A survey
AU - El Jurdi, Rosana
AU - Petitjean, Caroline
AU - Honeine, Paul
AU - Cheplygina, Veronika
AU - Abdallah, Fahed
N1 - Funding Information:
The authors would like to acknowledge the National Council for Scientific Research of Lebanon (CNRS-L) and the Agence Fran?aise de la Francophonie (AUF) for granting a doctoral fellowship to Rosana El Jurdi, as well as the ANR (project APi, grant ANR-18-CE23-0014). This work is part of the DAISI project, co-financed by the European Union with the European Regional Development Fund (ERDF) and by the Normandy Region. This research was conducted as part of a collaboration with the Eindhoven university of Technology under the support of the PHC Van Gogh project WeSmile.
Funding Information:
The authors would like to acknowledge the National Council for Scientific Research of Lebanon (CNRS-L) and the Agence Française de la Francophonie (AUF) for granting a doctoral fellowship to Rosana El Jurdi, as well as the ANR (project APi, grant ANR-18-CE23-0014). This work is part of the DAISI project, co-financed by the European Union with the European Regional Development Fund (ERDF) and by the Normandy Region. This research was conducted as part of a collaboration with the Eindhoven university of Technology under the support of the PHC Van Gogh project WeSmile.
Publisher Copyright:
© 2021 Elsevier Inc.
PY - 2021/9
Y1 - 2021/9
N2 - Today, deep convolutional neural networks (CNNs) have demonstrated state of the art performance for supervised medical image segmentation, across various imaging modalities and tasks. Despite early success, segmentation networks may still generate anatomically aberrant segmentations, with holes or inaccuracies near the object boundaries. To mitigate this effect, recent research works have focused on incorporating spatial information or prior knowledge to enforce anatomically plausible segmentation. If the integration of prior knowledge in image segmentation is not a new topic in classical optimization approaches, it is today an increasing trend in CNN based image segmentation, as shown by the growing literature on the topic. In this survey, we focus on high level prior, embedded at the loss function level. We categorize the articles according to the nature of the prior: the object shape, size, topology, and the inter-regions constraints. We highlight strengths and limitations of current approaches, discuss the challenge related to the design and the integration of prior-based losses, and the optimization strategies, and draw future research directions.
AB - Today, deep convolutional neural networks (CNNs) have demonstrated state of the art performance for supervised medical image segmentation, across various imaging modalities and tasks. Despite early success, segmentation networks may still generate anatomically aberrant segmentations, with holes or inaccuracies near the object boundaries. To mitigate this effect, recent research works have focused on incorporating spatial information or prior knowledge to enforce anatomically plausible segmentation. If the integration of prior knowledge in image segmentation is not a new topic in classical optimization approaches, it is today an increasing trend in CNN based image segmentation, as shown by the growing literature on the topic. In this survey, we focus on high level prior, embedded at the loss function level. We categorize the articles according to the nature of the prior: the object shape, size, topology, and the inter-regions constraints. We highlight strengths and limitations of current approaches, discuss the challenge related to the design and the integration of prior-based losses, and the optimization strategies, and draw future research directions.
KW - Anatomical constraint losses
KW - Convolutional neural networks
KW - Deep learning
KW - Medical image segmentation
KW - Prior-based loss functions
UR - http://www.scopus.com/inward/record.url?scp=85111911511&partnerID=8YFLogxK
U2 - 10.1016/j.cviu.2021.103248
DO - 10.1016/j.cviu.2021.103248
M3 - Article
AN - SCOPUS:85111911511
SN - 1077-3142
VL - 210
JO - Computer Vision and Image Understanding
JF - Computer Vision and Image Understanding
M1 - 103248
ER -