Scholarlink

<p>Federated learning provides a comprehensive distributed AI paradigm in the healthcare sector by effectively managing and ensuring the privacy of sensitive records through the training of multiple local models without requiring data sharing or access. Several research efforts have been made by the researchers in the FL-based AI concerns by focusing on the general architecture, protocols, designs, privacy issues and prediction of records in local data sets. Moreover, the FL applications needed to be incorporated into predicting and classifying medical records more systematically and accurately, while facilitating accurate decision-making. This paper has proposed an accurate decision-making and classification method using ontology and prediction models for handling the distributed medical records of patients collected by heterogeneous intelligent devices. The proposed mechanism ensures an accurate decision-making and classification method using ontology and a rational decision prediction model for handling the distributed medical records. The proposed mechanism efficiently provides accurate decision-making while sharing and recording the information in a heterogeneous network. The proposed model is further validated and experimented against heterogeneous data sets in terms of accuracy, prediction, and classification metrics compared to existing models.</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 distributed classification and prediction model using federated learning in healthcare

  • 期刊论文

  • 发表日期：

    2025年6月30日

查看全文
  • Geetanjali Rathee,
  • Aparna Singh,
  • Gaurav Singal,
  • Abhinav Tomar

摘要

Federated learning provides a comprehensive distributed AI paradigm in the healthcare sector by effectively managing and ensuring the privacy of sensitive records through the training of multiple local models without requiring data sharing or access. Several research efforts have been made by the researchers in the FL-based AI concerns by focusing on the general architecture, protocols, designs, privacy issues and prediction of records in local data sets. Moreover, the FL applications needed to be incorporated into predicting and classifying medical records more systematically and accurately, while facilitating accurate decision-making. This paper has proposed an accurate decision-making and classification method using ontology and prediction models for handling the distributed medical records of patients collected by heterogeneous intelligent devices. The proposed mechanism ensures an accurate decision-making and classification method using ontology and a rational decision prediction model for handling the distributed medical records. The proposed mechanism efficiently provides accurate decision-making while sharing and recording the information in a heterogeneous network. The proposed model is further validated and experimented against heterogeneous data sets in terms of accuracy, prediction, and classification metrics compared to existing models.