Uncertainty-Aware Semi-Supervised Method for Pectoral Muscle Segmentation
The consistency regularization method is a widely used semi-supervised method that uses regularization terms constructed from unlabeled data to improve model performance. Poor-quality target predictions in regularization terms produce noisy gradient flows during training, resulting in a degradation...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Bioengineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5354/12/1/36 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589030957514752 |
|---|---|
| author | Yutao Tang Yongze Guo Huayu Wang Ting Song Yao Lu |
| author_facet | Yutao Tang Yongze Guo Huayu Wang Ting Song Yao Lu |
| author_sort | Yutao Tang |
| collection | DOAJ |
| description | The consistency regularization method is a widely used semi-supervised method that uses regularization terms constructed from unlabeled data to improve model performance. Poor-quality target predictions in regularization terms produce noisy gradient flows during training, resulting in a degradation in model performance. Recent semi-supervised methods usually filter out low-confidence target predictions to alleviate this problem, but also prevent the model from learning features from unlabeled data in low-confidence regions. Specifically, in medical imaging and other cross-domain scenarios, models are prone to producing large numbers of low-confidence predictions. To improve the quality of target predictions while utilizing unlabeled data more efficiently, we propose an uncertainty-aware semi-supervised method that incorporates the breast anatomical prior, for pectoral muscle segmentation. Our method has a typical teacher-student dual model structure, where uncertainty is used to distinguish between high- and low-confidence predictions in the teacher model output. A low-confidence prediction refinement module was designed to refine the low-confidence predictions by incorporating high-confidence predictions and a learned anatomical prior. The anatomical prior, as regularization of the target predictions, was learned from annotations and an auxiliary task. The final target predictions are a combination of high-confidence teacher predictions and refined low-confidence predictions. The proposed method was evaluated on a dataset containing 635 data points from three data centers. Compared with the baseline method, the proposed method showed an average improvement in DICE index of 1.76, an average reduction in IoU index of 3.21, and an average reduction in HD index of 5.48. The experimental results show that our method generalizes well to the test set and outperforms other methods in all evaluation metrics. |
| format | Article |
| id | doaj-art-257b9a69c44d4ab4beed85ca6379104d |
| institution | Kabale University |
| issn | 2306-5354 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Bioengineering |
| spelling | doaj-art-257b9a69c44d4ab4beed85ca6379104d2025-01-24T13:23:02ZengMDPI AGBioengineering2306-53542025-01-011213610.3390/bioengineering12010036Uncertainty-Aware Semi-Supervised Method for Pectoral Muscle SegmentationYutao Tang0Yongze Guo1Huayu Wang2Ting Song3Yao Lu4School of Computer Science and Engineering, Sun-Yat sen University, Guanghzou 510006, ChinaSchool of Computer Science and Engineering, Sun-Yat sen University, Guanghzou 510006, ChinaSchool of Computer Science and Engineering, Sun-Yat sen University, Guanghzou 510006, ChinaThe Third Affiliated Hospital of Guangzhou Medical University, Guanghzou 510150, ChinaSchool of Computer Science and Engineering, Sun-Yat sen University, Guanghzou 510006, ChinaThe consistency regularization method is a widely used semi-supervised method that uses regularization terms constructed from unlabeled data to improve model performance. Poor-quality target predictions in regularization terms produce noisy gradient flows during training, resulting in a degradation in model performance. Recent semi-supervised methods usually filter out low-confidence target predictions to alleviate this problem, but also prevent the model from learning features from unlabeled data in low-confidence regions. Specifically, in medical imaging and other cross-domain scenarios, models are prone to producing large numbers of low-confidence predictions. To improve the quality of target predictions while utilizing unlabeled data more efficiently, we propose an uncertainty-aware semi-supervised method that incorporates the breast anatomical prior, for pectoral muscle segmentation. Our method has a typical teacher-student dual model structure, where uncertainty is used to distinguish between high- and low-confidence predictions in the teacher model output. A low-confidence prediction refinement module was designed to refine the low-confidence predictions by incorporating high-confidence predictions and a learned anatomical prior. The anatomical prior, as regularization of the target predictions, was learned from annotations and an auxiliary task. The final target predictions are a combination of high-confidence teacher predictions and refined low-confidence predictions. The proposed method was evaluated on a dataset containing 635 data points from three data centers. Compared with the baseline method, the proposed method showed an average improvement in DICE index of 1.76, an average reduction in IoU index of 3.21, and an average reduction in HD index of 5.48. The experimental results show that our method generalizes well to the test set and outperforms other methods in all evaluation metrics.https://www.mdpi.com/2306-5354/12/1/36deep learningsegmentationuncertainty |
| spellingShingle | Yutao Tang Yongze Guo Huayu Wang Ting Song Yao Lu Uncertainty-Aware Semi-Supervised Method for Pectoral Muscle Segmentation Bioengineering deep learning segmentation uncertainty |
| title | Uncertainty-Aware Semi-Supervised Method for Pectoral Muscle Segmentation |
| title_full | Uncertainty-Aware Semi-Supervised Method for Pectoral Muscle Segmentation |
| title_fullStr | Uncertainty-Aware Semi-Supervised Method for Pectoral Muscle Segmentation |
| title_full_unstemmed | Uncertainty-Aware Semi-Supervised Method for Pectoral Muscle Segmentation |
| title_short | Uncertainty-Aware Semi-Supervised Method for Pectoral Muscle Segmentation |
| title_sort | uncertainty aware semi supervised method for pectoral muscle segmentation |
| topic | deep learning segmentation uncertainty |
| url | https://www.mdpi.com/2306-5354/12/1/36 |
| work_keys_str_mv | AT yutaotang uncertaintyawaresemisupervisedmethodforpectoralmusclesegmentation AT yongzeguo uncertaintyawaresemisupervisedmethodforpectoralmusclesegmentation AT huayuwang uncertaintyawaresemisupervisedmethodforpectoralmusclesegmentation AT tingsong uncertaintyawaresemisupervisedmethodforpectoralmusclesegmentation AT yaolu uncertaintyawaresemisupervisedmethodforpectoralmusclesegmentation |