Federated Learning Framework Addresses Data Privacy Concerns in Medical Image Analysis

Research conducted at Hebei University of Technology, in collaboration with financial supporters including the National Key R&D Program of China and the National Natural Science Foundation of China, has developed a federated learning framework called FL-MONAS. This framework aims to address the challenges of data privacy, cross-institutional data distribution differences, and multi-view information fusion in medical imaging. By leveraging the advantages of neural architecture search and federated learning, FL-MONAS effectively combines 2D MRI with 3D anatomical structures, improving model performance while ensuring data privacy.

Key Takeaways:

• The FL-MONAS framework combines federated learning, neural architecture search, and data fusion techniques to address data privacy and cross-institutional data distribution differences in medical imaging.
• The framework uses a Pareto-frontier-based multi-objective optimization strategy to effectively combine 2D MRI with 3D anatomical structures.
• Experimental results show that FL-MONAS maintains strong segmentation performance even in non-IID scenarios, providing an efficient and privacy-friendly solution for medical image analysis.
• The research was conducted by Bin Cao, Huanyu Deng, Yiming Hao, and Xiao Luo from Hebei University of Technology in collaboration with financial supporters including the National Key R&D Program of China and the National Natural Science Foundation of China.
• The framework has been peer-reviewed and published in the journal Information Fusion.

Statistics:

• The research received financial support from the National Key R&D Program of China, the National Natural Science Foundation of China, the Natural Science Fund of Hebei Province for Distinguished Young Scholars, the Science and Technology Project of Hebei Education Department, and the S&T Program of Hebei.
• The framework was developed using a Pareto-frontier-based multi-objective optimization strategy.
• The experimental results showed that FL-MONAS maintained strong segmentation performance on 2D MRI data with 3D anatomical structures.

Sources:

• Cao, B., Deng, H., Hao, Y., & Luo, X. (2025). Multi-view information fusion based on federated multi-objective neural architecture search for MRI semantic segmentation. Information Fusion, 123, 1-12.
• National Key R&D Program of China
• National Natural Science Foundation of China (NSFC)
• Natural Science Fund of Hebei Province for Distinguished Young Scholars
• Science and Technology Project of Hebei Education Department
• S&T Program of Hebei