Source-Free Domain Adaptation for Cross-Modality Abdominal Multi-Organ Segmentation Challenges
Abdominal organ segmentation in CT images is crucial for accurate diagnosis, treatment planning, and condition monitoring. However, the annotation process is often hindered by challenges such as low contrast, artifacts, and complex organ structures. While unsupervised domain adaptation (UDA) has sho...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2078-2489/16/6/460 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850167869433708544 |
|---|---|
| author | Xiyu Zhang Xu Chen Yang Wang Dongliang Liu Yifeng Hong |
| author_facet | Xiyu Zhang Xu Chen Yang Wang Dongliang Liu Yifeng Hong |
| author_sort | Xiyu Zhang |
| collection | DOAJ |
| description | Abdominal organ segmentation in CT images is crucial for accurate diagnosis, treatment planning, and condition monitoring. However, the annotation process is often hindered by challenges such as low contrast, artifacts, and complex organ structures. While unsupervised domain adaptation (UDA) has shown promise in addressing these issues by transferring knowledge from a different modality (source domain), its reliance on both source and target data during training presents a practical challenge in many clinical settings due to data privacy concerns. This study aims to develop a cross-modality abdominal multi-organ segmentation model for label-free CT (target domain) data, leveraging knowledge solely from a pre-trained source domain (MRI) model without accessing the source data. To achieve this, we generate source-like images from target-domain images using a one-way image translation approach with the pre-trained model. These synthesized images preserve the anatomical structure of the target, enabling segmentation predictions from the pre-trained model. To further enhance segmentation accuracy, particularly for organ boundaries and small contours, we introduce an auxiliary translation module with an image decoder and multi-level discriminator. The results demonstrate significant improvements across several performance metrics, including the Dice similarity coefficient (DSC) and average symmetric surface distance (ASSD), highlighting the effectiveness of the proposed method. |
| format | Article |
| id | doaj-art-39ec2c0b9827416cbd3fb0b9c95e8968 |
| institution | OA Journals |
| issn | 2078-2489 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Information |
| spelling | doaj-art-39ec2c0b9827416cbd3fb0b9c95e89682025-08-20T02:21:07ZengMDPI AGInformation2078-24892025-05-0116646010.3390/info16060460Source-Free Domain Adaptation for Cross-Modality Abdominal Multi-Organ Segmentation ChallengesXiyu Zhang0Xu Chen1Yang Wang2Dongliang Liu3Yifeng Hong4College of Computer Science and Technology, Huaqiao University, Xiamen 361021, ChinaCollege of Computer Science and Technology, Huaqiao University, Xiamen 361021, ChinaCollege of Computer Science and Technology, Huaqiao University, Xiamen 361021, ChinaCollege of Computer Science and Technology, Huaqiao University, Xiamen 361021, ChinaCollege of Computer Science and Technology, Huaqiao University, Xiamen 361021, ChinaAbdominal organ segmentation in CT images is crucial for accurate diagnosis, treatment planning, and condition monitoring. However, the annotation process is often hindered by challenges such as low contrast, artifacts, and complex organ structures. While unsupervised domain adaptation (UDA) has shown promise in addressing these issues by transferring knowledge from a different modality (source domain), its reliance on both source and target data during training presents a practical challenge in many clinical settings due to data privacy concerns. This study aims to develop a cross-modality abdominal multi-organ segmentation model for label-free CT (target domain) data, leveraging knowledge solely from a pre-trained source domain (MRI) model without accessing the source data. To achieve this, we generate source-like images from target-domain images using a one-way image translation approach with the pre-trained model. These synthesized images preserve the anatomical structure of the target, enabling segmentation predictions from the pre-trained model. To further enhance segmentation accuracy, particularly for organ boundaries and small contours, we introduce an auxiliary translation module with an image decoder and multi-level discriminator. The results demonstrate significant improvements across several performance metrics, including the Dice similarity coefficient (DSC) and average symmetric surface distance (ASSD), highlighting the effectiveness of the proposed method.https://www.mdpi.com/2078-2489/16/6/460source-free domain adaptationcontrastive learningmedical image segmentation |
| spellingShingle | Xiyu Zhang Xu Chen Yang Wang Dongliang Liu Yifeng Hong Source-Free Domain Adaptation for Cross-Modality Abdominal Multi-Organ Segmentation Challenges Information source-free domain adaptation contrastive learning medical image segmentation |
| title | Source-Free Domain Adaptation for Cross-Modality Abdominal Multi-Organ Segmentation Challenges |
| title_full | Source-Free Domain Adaptation for Cross-Modality Abdominal Multi-Organ Segmentation Challenges |
| title_fullStr | Source-Free Domain Adaptation for Cross-Modality Abdominal Multi-Organ Segmentation Challenges |
| title_full_unstemmed | Source-Free Domain Adaptation for Cross-Modality Abdominal Multi-Organ Segmentation Challenges |
| title_short | Source-Free Domain Adaptation for Cross-Modality Abdominal Multi-Organ Segmentation Challenges |
| title_sort | source free domain adaptation for cross modality abdominal multi organ segmentation challenges |
| topic | source-free domain adaptation contrastive learning medical image segmentation |
| url | https://www.mdpi.com/2078-2489/16/6/460 |
| work_keys_str_mv | AT xiyuzhang sourcefreedomainadaptationforcrossmodalityabdominalmultiorgansegmentationchallenges AT xuchen sourcefreedomainadaptationforcrossmodalityabdominalmultiorgansegmentationchallenges AT yangwang sourcefreedomainadaptationforcrossmodalityabdominalmultiorgansegmentationchallenges AT dongliangliu sourcefreedomainadaptationforcrossmodalityabdominalmultiorgansegmentationchallenges AT yifenghong sourcefreedomainadaptationforcrossmodalityabdominalmultiorgansegmentationchallenges |