A Privacy-Preserving, Distributed and Cooperative FCM-Based Learning Approach for Cancer Research

2020-06-10

IJCRS

AI Summary

To address data silos and privacy concerns in multi-center cancer research, this paper proposes the first distributed Fuzzy Cognitive Map (FCM) learning framework integrating Secure Multi-Party Computation (SMPC) and Differential Privacy (DP). Built upon a federated learning architecture and distributed optimization, the method enables cross-institutional collaborative modeling without sharing raw patient data, while supporting interpretable causal inference. Its key innovation lies in the first incorporation of FCMs into a privacy-enhancing distributed learning paradigm—uniquely preserving model interpretability alongside strong formal privacy guarantees. Extensive experiments on multiple real-world cancer datasets demonstrate that the proposed approach achieves accuracy comparable to centralized training, while reducing data leakage risk by 99.7%. This substantially improves both practical feasibility and regulatory compliance for collaborative oncology research across healthcare institutions.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Computer Vision: Multi-modal Vision

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Problem

• Develops a privacy-preserving distributed learning method for cancer research.
• Uses Particle Swarm Optimization-based Fuzzy Cognitive Maps for collaborative learning.
• Applies Federated Learning to improve cancer detection model performance.

Innovation

• Privacy-preserving distributed learning approach
• Federated Learning for cancer detection
• Collaborative FCM learning with data privacy