What type of study is this?

This is a Quantitative Study study.

October 20, 2025Open Access

Pancreas Part Segmentation under Federated Learning Paradigm

View Full Paper

Ask AI

Bookmark

Authors

This innovative method addresses pancreas part segmentation in MRI, suggesting effective solutions for data scarcity and technical complexity.

The approach achieves clinically viable performance in pancreas part segmentation despite data scarcity.
Attention U-Net with FedAvg was identified as the optimal combination for addressing pancreatic heterogeneity.
A privacy-preserving federated learning framework was used for collaborative model training across seven medical institutions.
A novel anatomically-informed loss function was introduced to enhance the texture contrasts in MRI for segmentation.

Hong et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

This innovative method addresses pancreas part segmentation in MRI, suggesting effective solutions for data scarcity and technical complexity.