Federated Learning for Privacy-Preserving Sentiment Analysis in Distributed Electronic Health Record Environments: A Systematic Literature Review
Abstract
Federated learning (FL) has emerged as a privacy-preserving approach for distributed healthcare analytics, yet its application to sentiment analysis of unstructured electronic health record (EHR) narratives remains limited. This systematic review examined the empirical maturity, methodological trends, and governance implications of federated sentiment-aware learning in distributed EHR settings. Following PRISMA 2020, searches were conducted in IEEE Xplore, Scopus, Web of Science, ScienceDirect, and PubMed on January 5, 2026, covering peer-reviewed studies published from January 2021 to January 2026. After screening and eligibility assessment, 29 empirical implementation studies were included in the qualitative synthesis, while conceptual and survey papers were reviewed contextually but excluded from the core analysis. The evidence shows that FL in healthcare is advancing mainly in structured prediction and privacy-preserving infrastructure. By contrast, sentiment-aware learning on unstructured clinical narratives remains at an early stage, with limited implementation and validation. This review distinguishes empirical from conceptual contributions and proposes a governance-aware, literature-derived framework to guide future implementation-focused research.
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.