Security and Privacy in Healthcare Applications Using Federated Learning - Review

This review explored data security and privacy in Internet of Health Things (IoHT) networks, focusing on local training for AI models and Federated Learning (FL) to ensure privacy and integrity. This highlights the importance of leveraging technological advancements for disease prediction and data exchange. Functional programming can be used in various industries and applications under real-world scenarios. In the medical industry, safeguarding the confidentiality of patient records and medical status is essential. This is where collaborative or federated learning becomes relevant. On the other hand, creating an intelligent system that helps medical personnel without exposing data can result in a Federated Learning concept. An example is an AI-based intelligent system for diagnosing brain tumors, which can effectively operate within a teamwork setting. Advancements in smart devices and applications designed for the Internet of Healthcare Things have created an ideal setting for implementing Machine Learning techniques. Nevertheless, conventional ML solutions work by collecting and processing data in a centralized manner. Federated Learning (FL) offers a potential solution for training machine-learning models on numerous separate devices without the need to share private data. As a result, FL provides a secure framework for managing extremely sensitive data in the context of the IoHT. This survey provides a detailed overview of new data security and privacy tools for Federated Learning in Internet of Health Things networks. Initially, we introduced the fundamental concepts of Federated Learning (FL) as it is applied to the IoHT.