Federated Learning Achieves 99% Accuracy in CKD Prediction with XAI

A new study demonstrates that federated learning (FL) combined with explainable AI (XAI) can predict chronic kidney disease (CKD) with 99% average accuracy. Researchers used a VotingClassifier ensemble of Random Forest, AdaBoost, and XGBoost models trained on a clinical dataset, with GridSearchCV optimizing client-side performance. The global model's high accuracy highlights the potential of interpretable FL for early CKD diagnosis, addressing both data privacy and model transparency in healthcare.