Scholarlink

Traditional lung tumor detection methods often conflict with patient privacy concerns. Forest Federated Learning (FFL) bridges this gap by allowing hospitals to analyze local data and share only the derived insights, not the raw data. This approach forms a collective “forest” model, promoting collaboration while ensuring data privacy. The methodology involves local model training and secure knowledge aggregation, facilitating a decentralized and privacy-preserving solution. Extensive testing on real lung cancer datasets demonstrates that FFL's accuracy rivals traditional centralized methods, offering significant advantages in scalability and privacy. FFL's adaptability and robustness make it a promising universal solution for lung tumor detection, representing a significant advancement in responsible and scalable healthcare analytics in the era of decentralized data.

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

Harnessing Collective Intelligence Without Data Sharing: A Federated Random Forest Approach for Lung Tumor Prediction

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ch. Srividya,
  • K. RamaSubramanian,
  • M. Madhu Bala

摘要

Traditional lung tumor detection methods often conflict with patient privacy concerns. Forest Federated Learning (FFL) bridges this gap by allowing hospitals to analyze local data and share only the derived insights, not the raw data. This approach forms a collective “forest” model, promoting collaboration while ensuring data privacy. The methodology involves local model training and secure knowledge aggregation, facilitating a decentralized and privacy-preserving solution. Extensive testing on real lung cancer datasets demonstrates that FFL's accuracy rivals traditional centralized methods, offering significant advantages in scalability and privacy. FFL's adaptability and robustness make it a promising universal solution for lung tumor detection, representing a significant advancement in responsible and scalable healthcare analytics in the era of decentralized data.