EasyGuide International EasyGuideYour Partner in Progress

Journal Article · 2025

A Blockchain-Secured Federated Learning System for Detection and Early Warning of Foot-and-Mouth Disease in Ugandan Cattle Farms

61 views 3 downloads

Authors

Abstract

Training machine learning models traditionally requires collecting all data in one central location. While this works in well-resourced environments with reliable internet, it creates problems in resource-limited settings. Privacy concerns prevent farmers and health facilities from sharing sensitive data, and poor connectivity makes uploading large datasets impractical. Centralizing data also creates security risks and trust issues when multiple organizations need to collaborate. Federated learning offers an alternative by training models on distributed data without moving it from its source. Each participant trains locally and shares only model updates, which are aggregated to create a global model. However, this introduces a trust problem: participants cannot verify whether aggregation was done correctly or whether some contributed manipulated updates. This research addresses the trust problem by integrating blockchain technology with federated learning. The blockchain provides a transparent, tamper-proof record of the training process, allowing participants to verify that all steps were executed correctly. This study designs and evaluates such a system specifically for resource-limited environments, using Foot-and-Mouth Disease prediction in Ugandan cattle farms as a practical test case. The system demonstrates that collaborative machine learning can work in settings with limited resources and unreliable connectivity while maintaining privacy and trust. The architecture balances security with computational efficiency, making blockchain-verified federated learning practical for developing regions. This has broader implications for privacy-preserving AI in agriculture, healthcare, and other domains where data sharing faces technical and trust barriers.

Keywords

Federated Learning Blockchain Privacy-Preserving Machine Learning Disease Prediction Distributed Systems Differential Privacy Byzantine Fault Tolerance

Research areas

All publications

Join Us

Ready to join MRU?

Applications are open for the next intake. Apply on the E-Portal in under twenty minutes, or talk to the admissions team first — whichever helps you more.