FedSAF: A Federated Learning Framework for Enhanced Gastric Cancer Detection and Privacy Preservation

2025-03-20

AI Summary

To address the challenges of multi-center data silos, limited annotated samples, strict privacy constraints, and non-IID data distributions in early gastric cancer detection—leading to degraded model performance—this paper proposes a privacy-preserving federated learning framework. Our method innovatively integrates an attention-guided message-passing strategy with Fisher information matrix–driven model updates, and incorporates model partitioning to reduce communication and computational overhead. We rigorously validate each component through hyperparameter optimization and systematic ablation studies. On a real-world gastric cancer dataset, our approach significantly outperforms baseline federated methods—including FedAvg, FedProx, and FedAMP—in both accuracy and convergence stability. Furthermore, cross-domain generalization experiments on SEED, BOT, Fashion-MNIST, and CIFAR-10 demonstrate superior robustness and efficiency under heterogeneous data conditions. The framework thus advances practical, scalable, and privacy-compliant AI for clinical gastric cancer screening.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Security and Privacy: Large-scale security measurements

Abstract

Gastric cancer is one of the most commonly diagnosed cancers and has a high mortality rate. Due to limited medical resources, developing machine learning models for gastric cancer recognition provides an efficient solution for medical institutions. However, such models typically require large sample sizes for training and testing, which can challenge patient privacy. Federated learning offers an effective alternative by enabling model training across multiple institutions without sharing sensitive patient data. This paper addresses the limited sample size of publicly available gastric cancer data with a modified data processing method. This paper introduces FedSAF, a novel federated learning algorithm designed to improve the performance of existing methods, particularly in non-independent and identically distributed (non-IID) data scenarios. FedSAF incorporates attention-based message passing and the Fisher Information Matrix to enhance model accuracy, while a model splitting function reduces computation and transmission costs. Hyperparameter tuning and ablation studies demonstrate the effectiveness of this new algorithm, showing improvements in test accuracy on gastric cancer datasets, with FedSAF outperforming existing federated learning methods like FedAMP, FedAvg, and FedProx. The framework's robustness and generalization ability were further validated across additional datasets (SEED, BOT, FashionMNIST, and CIFAR-10), achieving high performance in diverse environments.

Problem

• Improves gastric cancer detection using federated learning
• Enhances privacy by avoiding sensitive data sharing
• Optimizes performance in non-IID data scenarios

Innovation

• Federated learning with privacy preservation
• Attention-based message passing for accuracy
• Model splitting reduces computation costs