Scholarlink

<p>Data-driven medical applications, powered by big data and artificial intelligence, generate scalable models from extensive datasets. This innovation attracts both academia and industry, significantly enhancing healthcare quality while posing integration challenges. Federated learning emerges as a transformative approach in healthcare, facilitating collaborative machine learning while preserving data privacy. This article reviews research on federated learning in healthcare, utilizing the "Web of Science" database to examine development trends and application progress. Initially, the federated learning system is dissected, and its application methods are analyzed, summarizing 18 federated learning frameworks and 9 privacy-preserving methods suitable for federated learning, and exploring its limitations in healthcare applications. Subsequently, recent literature on federated learning in four major areas—disease diagnosis and risk assessment, medical image analysis, drug discovery and development, and disease management—is reviewed, summarizing the general process of applying Federated Learning in healthcare. Finally, the challenges faced by federated learning in the medical field and the solutions currently being explored by scholars are summarized. This article aims to comprehensively summarize the research progress and application trends of federated learning technology in the medical field, analyze its limitations and challenges, and anticipate its future development to further promote sustainable advancement in the healthcare industry.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A review of federated learning technology and its research progress in healthcare applications

  • 期刊论文

  • 发表日期：

    2025年6月18日

查看全文
  • Zezhong Ma,
  • Nur Intan Raihana Ruhaiyem,
  • Meng Zhang,
  • Kamarul Imran Musa,
  • Tengku Muhammad Hanis,
  • Tianyun Xiao,
  • Dianbo Hua,
  • Hao Li

摘要

Data-driven medical applications, powered by big data and artificial intelligence, generate scalable models from extensive datasets. This innovation attracts both academia and industry, significantly enhancing healthcare quality while posing integration challenges. Federated learning emerges as a transformative approach in healthcare, facilitating collaborative machine learning while preserving data privacy. This article reviews research on federated learning in healthcare, utilizing the "Web of Science" database to examine development trends and application progress. Initially, the federated learning system is dissected, and its application methods are analyzed, summarizing 18 federated learning frameworks and 9 privacy-preserving methods suitable for federated learning, and exploring its limitations in healthcare applications. Subsequently, recent literature on federated learning in four major areas—disease diagnosis and risk assessment, medical image analysis, drug discovery and development, and disease management—is reviewed, summarizing the general process of applying Federated Learning in healthcare. Finally, the challenges faced by federated learning in the medical field and the solutions currently being explored by scholars are summarized. This article aims to comprehensively summarize the research progress and application trends of federated learning technology in the medical field, analyze its limitations and challenges, and anticipate its future development to further promote sustainable advancement in the healthcare industry.