Application of XGBoost Algorithm for the Analysis of Healthcare Data at the Fog Layer

Abstract

The present paper focuses on the analysis of healthcare data using XGBoost algorithm at the Fog Layer. The present paper is based on an implementation of the Edge-Fog Layer layered architecture by considering the ESP8266 as the Edge node and Raspberry Pi is the Fog Node. The data has been captured, encrypted based on a custom-built encryption algorithm named SPADE and forwarded to the Raspberry Pi using the MQTT publish subscribe model. The Raspberry Pi acts as the data aggregator from multiple other Edge Nodes. The data within Raspberry Pi then analyzed based on the XGBoost algorithm after decryption which is done using Reverse SPADE algorithm. The results of analysis are then communicated to relevant application programs using FastAPI while also being stored on the Firebase database. The present framework provides a cost-effective implementation mechanism for analysis of healthcare data received from Edge nodes in a secure manner. The results of the analysis are presented in the form of prediction whether the health condition of the patient is critical. This results in the healthcare providers being able to initiate necessary healthcare procedures required to improve the healthcare condition of the patient. General Terms Internet of Things, Healthcare, Machine Learning Algorithms.