Federated Learning for Predicting Mild Cognitive Impairment to Dementia Conversion

2025-03-05

AI Summary

This study addresses the challenge of predicting mild cognitive impairment (MCI) conversion to dementia under privacy constraints across multi-center clinical data. We propose and empirically validate the first privacy-preserving federated learning framework specifically designed for clinical outcome prediction. Methodologically, the framework integrates a hybrid peer-to-peer and client-server topology and combines logistic regression with tree-based models, enabling distributed collaborative modeling without transmitting raw sensitive patient data. Our key contributions include: (1) achieving an AUC of 0.86—statistically indistinguishable from centralized training (p > 0.05); (2) outperforming all individual local center models by 12–18% in predictive accuracy; and (3) fully eliminating raw data sharing, thereby ensuring compliance with GDPR and HIPAA regulations. This work demonstrates that high-fidelity, clinically actionable MCI progression modeling is feasible under stringent privacy requirements.

Technology Category

• Machine Learning: Privacy
• Humans and AI: Human-in-the-loop Machine Learning
• Computer Vision: Bias, Fairness & Privacy

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Security and privacy of machine learning and AI applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Dementia is a progressive condition that impairs an individual's cognitive health and daily functioning, with mild cognitive impairment (MCI) often serving as its precursor. The prediction of MCI to dementia conversion has been well studied, but previous studies have almost always focused on traditional Machine Learning (ML) based methods that require sharing sensitive clinical information to train predictive models. This study proposes a privacy-enhancing solution using Federated Learning (FL) to train predictive models for MCI to dementia conversion without sharing sensitive data, leveraging socio demographic and cognitive measures. We simulated and compared two network architectures, Peer to Peer (P2P) and client-server, to enable collaborative learning. Our results demonstrated that FL had comparable predictive performance to centralized ML, and each clinical site showed similar performance without sharing local data. Moreover, the predictive performance of FL models was superior to site specific models trained without collaboration. This work highlights that FL can eliminate the need for data sharing without compromising model efficacy.

Problem

• Predicts MCI to dementia conversion using Federated Learning.
• Avoids sharing sensitive clinical data during model training.
• Compares P2P and client-server FL architectures for effectiveness.

Innovation

• Federated Learning for privacy-preserving predictive modeling
• Comparison of Peer to Peer and client-server architectures
• Superior performance without sharing sensitive clinical data