Federated learning enables privacy-preserving AI training across pathology networks
The review examines federated learning as a framework for training artificial intelligence models across multiple pathology institutions without sharing raw image data. Investigators summarize current use cases, technical architectures, and the regulatory and ethical barriers to implementation. The authors argue that this decentralized approach can overcome data silos and privacy constraints while enabling scalable model development. Standardizing and prospectively validating federated learning pipelines will be essential before these systems can be integrated into routine diagnostic workflows.
The original study
Federated learning for multi-institutional AI in healthcare via digital pathology.
- Authors
- Soleimani R, Azimi M, Talebi N
- Journal
- Biomedizinische Technik. Biomedical engineering
- Type
- Journal Article, Review
- PMID
- 42683630
Original abstract
The integration of artificial intelligence (AI) in digital pathology has shown significant promise in advancing cancer diagnostics, grading, and treatment response prediction. However, widespread development and deployment of robust AI models face critical challenges due to data silos, privacy concerns, and the need for large-scale multi-institutional datasets. Federated Learning (FL) presents a transformative approach by enabling collaborative model training across hospitals without direct data sharing. In this review, we summarize recent developments in FL as applied to digital pathology, highlight pioneering use cases, and explore the technical, regulatory, and ethical hurdles. We discuss how FL can enable scalable, privacy-preserving AI models, and outline future directions for standardizing and validating FL-based approaches in clinical workflows.