December 6, 2025IEEE Journal of Biomedical and Health Informatics

Deep Unfolding Segmentation Network for Under-sampled Magnetic Resonance Images

View Full Paper

Ask AI

Bookmark

Authors

GZGuixu ZhangEast China Normal University

Discussion

Member takes

Implication

This approach demonstrates improved segmentation accuracy in under-sampled MR images, suggesting effective handling of artifacts and enhancing region detail capture.

Key Points

Improved segmentation accuracy is achieved in under-sampled MR images, addressing common artifacts.
Utilizing DUSNet, which incorporates compressive sensing for enhanced image reconstruction, significantly boosts performance.
This method employs the Augmented Lagrangian Method to optimize the unfolding iterative algorithm within the neural network framework.
Highlights the potential of DUSNet in capturing crucial edge details, which may enable better disease diagnosis in clinical settings.

Cite This Study

Guixu Zhang (2025) studied this question.

synapsesocial.com/papers/69337cefb3f947a0a125a2ba https://doi.org/10.1109/jbhi.2025.3639469

View Full Paper

Ask AI

Bookmark

Implication

This approach demonstrates improved segmentation accuracy in under-sampled MR images, suggesting effective handling of artifacts and enhancing region detail capture.

Deep Unfolding Segmentation Network for Under-sampled Magnetic Resonance Images

Discussion

Member takes

Implication

Key Points

Cite This Study

Deep Unfolding Segmentation Network for Under-sampled Magnetic Resonance Images

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider