Privacy-Preserving Healthcare Fraud Detection: A Federated and Data Mining–Centric Survey

Healthcare fraud poses a significant financial and operational burden on healthcare systems, while the sensitive nature of medical data imposes strict privacy requirements on analytical solutions. Traditional fraud detection approaches often rely on centralized data collection, which conflicts with regulatory and ethical constraints related to data sharing. To address this challenge, privacy-preserving data mining and machine learning techniques have gained increasing attention. This paper presents a concise survey of privacy-preserving approaches for healthcare fraud detection, with a particular focus on federated learning, differential privacy, cryptographic methods, and privacy-aware data mining techniques. We introduce a taxonomy that categorizes existing methods based on their underlying privacy mechanisms and analyze representative approaches through a structured comparative study. Furthermore, we discuss practical deployment scenarios and identify key open challenges that hinder real-world adoption, including privacy–utility trade-offs, scalability, and system heterogeneity. By synthesizing recent advances and highlighting unresolved research gaps, this survey aims to provide researchers and practitioners with a clear understanding of the current landscape and future directions of privacy-preserving healthcare fraud detection satisfaction.