Cross-Domain Carotid Artery Segmentation Using Folding Fan ResNet and Quadratic Mapping Loss
Vessel-wall volume and localized three-dimensional ultrasound (3DUS) metrics are sensitive indicators of carotid atherosclerosis changes in response to therapies. Deep-learning segmentation models have been proposed to segment the media-adventitia boundary (MAB) and lumen-intima boundary (LIB) neces...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11062868/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849234469456707584 |
|---|---|
| author | Zhaozheng Chen Zhiyin Liu Bernard Chiu |
| author_facet | Zhaozheng Chen Zhiyin Liu Bernard Chiu |
| author_sort | Zhaozheng Chen |
| collection | DOAJ |
| description | Vessel-wall volume and localized three-dimensional ultrasound (3DUS) metrics are sensitive indicators of carotid atherosclerosis changes in response to therapies. Deep-learning segmentation models have been proposed to segment the media-adventitia boundary (MAB) and lumen-intima boundary (LIB) necessary to obtain these measurements. However, the performance of these models declines substantially when applied to datasets with different properties from their training data, highlighting a need for more generalizable models. We introduce Folding Fan ResNet (FFRN), a deep-learning model designed to enhance cross-domain generalizability. FFRN incorporates a novel skip connection pathway, the Folding Fan Connection (FFC), which improves encoder-decoder feature fusion through multiple addition and convolutional operations. This design reduces parameter redundancy while maintaining efficient feature propagation, leading to increased generalizability compared to existing models. Additionally, the quadratic mapping loss (QML) function is proposed to improve model fine-tuning. QML addresses arterial mislocalization by deferring the optimization of incorrectly localized boundaries while prioritizing adjustments to correctly segmented regions. This strategy prevents the reinforcement of erroneous features and allows for better modeling of carotid structures. FFRN demonstrated higher generalizability than four state-of-the-art segmentation models, improving Dice similarity coefficients (DSCs) of 78.9±24.1% and 79.0±23.2% attained by the U-Net for LIB and MAB segmentation, respectively, to 84.0±18.2% and 84.6±17.1% in MAB and LIB segmentation, respectively. With QML fine-tuning, the DSCs increased to 88.9±8.2% (LIB) and 90.6±8.1% (MAB), outperforming all competing approaches, even after their performances were enhanced by QML. Notably, the fine-tuned model is the first to achieve a zero incorrect localization rate (ILR) in the target dataset. The improved generalizability afforded by the model enhances efficiency for carotid disease monitoring across clinical centers. |
| format | Article |
| id | doaj-art-a2d2312f30284edc9173458331b6fdbd |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-a2d2312f30284edc9173458331b6fdbd2025-08-20T04:03:07ZengIEEEIEEE Access2169-35362025-01-011311737411738410.1109/ACCESS.2025.358506411062868Cross-Domain Carotid Artery Segmentation Using Folding Fan ResNet and Quadratic Mapping LossZhaozheng Chen0https://orcid.org/0000-0002-7946-1111Zhiyin Liu1https://orcid.org/0009-0004-4125-4565Bernard Chiu2https://orcid.org/0000-0001-5237-2410Department of Electrical Engineering, City University of Hong Kong, Hong Kong, Special Administrative Region, ChinaDivision of Life Science, The Hong Kong University of Science and Technology, Hong Kong, Special Administrative Region, ChinaDepartment of Physics & Computer Science, Wilfrid Laurier University, Waterloo, ON, CanadaVessel-wall volume and localized three-dimensional ultrasound (3DUS) metrics are sensitive indicators of carotid atherosclerosis changes in response to therapies. Deep-learning segmentation models have been proposed to segment the media-adventitia boundary (MAB) and lumen-intima boundary (LIB) necessary to obtain these measurements. However, the performance of these models declines substantially when applied to datasets with different properties from their training data, highlighting a need for more generalizable models. We introduce Folding Fan ResNet (FFRN), a deep-learning model designed to enhance cross-domain generalizability. FFRN incorporates a novel skip connection pathway, the Folding Fan Connection (FFC), which improves encoder-decoder feature fusion through multiple addition and convolutional operations. This design reduces parameter redundancy while maintaining efficient feature propagation, leading to increased generalizability compared to existing models. Additionally, the quadratic mapping loss (QML) function is proposed to improve model fine-tuning. QML addresses arterial mislocalization by deferring the optimization of incorrectly localized boundaries while prioritizing adjustments to correctly segmented regions. This strategy prevents the reinforcement of erroneous features and allows for better modeling of carotid structures. FFRN demonstrated higher generalizability than four state-of-the-art segmentation models, improving Dice similarity coefficients (DSCs) of 78.9±24.1% and 79.0±23.2% attained by the U-Net for LIB and MAB segmentation, respectively, to 84.0±18.2% and 84.6±17.1% in MAB and LIB segmentation, respectively. With QML fine-tuning, the DSCs increased to 88.9±8.2% (LIB) and 90.6±8.1% (MAB), outperforming all competing approaches, even after their performances were enhanced by QML. Notably, the fine-tuned model is the first to achieve a zero incorrect localization rate (ILR) in the target dataset. The improved generalizability afforded by the model enhances efficiency for carotid disease monitoring across clinical centers.https://ieeexplore.ieee.org/document/11062868/Carotid artery segmentationthree-dimensional ultrasound (3DUS) imagesfolding fan ResNet (FFRN)quadratic mapping loss (QML) |
| spellingShingle | Zhaozheng Chen Zhiyin Liu Bernard Chiu Cross-Domain Carotid Artery Segmentation Using Folding Fan ResNet and Quadratic Mapping Loss IEEE Access Carotid artery segmentation three-dimensional ultrasound (3DUS) images folding fan ResNet (FFRN) quadratic mapping loss (QML) |
| title | Cross-Domain Carotid Artery Segmentation Using Folding Fan ResNet and Quadratic Mapping Loss |
| title_full | Cross-Domain Carotid Artery Segmentation Using Folding Fan ResNet and Quadratic Mapping Loss |
| title_fullStr | Cross-Domain Carotid Artery Segmentation Using Folding Fan ResNet and Quadratic Mapping Loss |
| title_full_unstemmed | Cross-Domain Carotid Artery Segmentation Using Folding Fan ResNet and Quadratic Mapping Loss |
| title_short | Cross-Domain Carotid Artery Segmentation Using Folding Fan ResNet and Quadratic Mapping Loss |
| title_sort | cross domain carotid artery segmentation using folding fan resnet and quadratic mapping loss |
| topic | Carotid artery segmentation three-dimensional ultrasound (3DUS) images folding fan ResNet (FFRN) quadratic mapping loss (QML) |
| url | https://ieeexplore.ieee.org/document/11062868/ |
| work_keys_str_mv | AT zhaozhengchen crossdomaincarotidarterysegmentationusingfoldingfanresnetandquadraticmappingloss AT zhiyinliu crossdomaincarotidarterysegmentationusingfoldingfanresnetandquadraticmappingloss AT bernardchiu crossdomaincarotidarterysegmentationusingfoldingfanresnetandquadraticmappingloss |