Skip to main content
School of Electronic Engineering and Computer Science

Introducing HealthFog, a Smart Healthcare System for the Automatic Diagnosis of Heart Diseases

HealthFog: An Ensemble Deep Learning based Smart Healthcare System for Automatic Diagnosis of Heart Diseases in Integrated Internet of Things (IOT) and Fog Computing Environments

A infographic displaying how a Fog computing system works

HealthFog is a new project focusing on healthcare aspects for heart patients. A real-life healthcare application platform, it analyses and identifies heart diseases automatically, and delivers diverse healthcare configurations for different user requirements, to efficiently manage the data of heart patients.

By utilising a new Fog based smart healthcare system, the project aims to provide automatic diagnosis of heart diseases using deep learning and Internet of Things (IoT). HealthFog provides healthcare as a Fog service and efficiently manages the data of heart patients which is coming from different IoT devices. HealthFog integrates deep learning in Edge computing devices and has already been deployed in a real-life application for heart disease analysis.

Deep learning-based models with very high accuracy require very high compute resources, such as CPU and GPU, both for training and prediction. By using a Fog service, HealthFog allows complex deep learning networks to be embedded at the edge of the network, with low latencies that were previously deployed on cloud platforms. This allows high accuracy to be achieved with very low response times, which is critical for healthcare applications.

The system has been validated for real-life heart patient data analysis by training neural networks on popular datasets, and deploying a working system that provides prediction results in real-time.

HeathFog can also be implemented as a real-time mobile application to provide a healthcare service to heart patients remotely. Currently, HealthFog works with file-based input data which can be seamlessly integrated with data streaming sensors for more pragmatic deployments. More intelligent ensemble models can be deployed for further improving the accuracy. HealthFog can be made robust and generic to incorporate other IoT applications such as agriculture, weather forecasting, traffic management and smart city.

This work is a joint venture, including Shreshth Tuli from IIT Delhi (India) as a leading author, Nipam Basumatary from IIT Madras (India), Dr Sukhpal Singh Gill from the School of EECS, QMUL, Prof. Mohsen Kahani from Ferdowski University (Iran), Prof. Rajkumar Buyya from University of Melbourne (Australia) and medical doctors: Dr. Rajesh Chandra Arya and Dr. Gurpreet Singh Wander from DMC Hospital India.

To date, the work has been cited by more than 110 research papers, and is currently being developed upon by PhD students across the globe. Many researchers from top-ranked universities and research labs have considered this work as a benchmark, and have developed IoT applications for other domains of healthcare, such as COVID-19, diabetes, cancer and hepatitis, to improve the healthcare service.

HealthFog has been released as an open-source software. The implementation code with experiment scripts and results can be found at the GitHub repository:

For further information, watch this video on YouTube:

Publication Details

Shreshth Tuli, Nipam Basumatary, Sukhpal Singh Gill, Mohsen Kahani, Rajesh Chand Arya, Gurpreet Singh Wander, and Rajkumar Buyya, HealthFog: An Ensemble Deep Learning based Smart Healthcare System for Automatic Diagnosis of Heart Diseases in Integrated IoT and Fog Computing Environments, Future Generation Computer Systems (FGCS), Volume 104, Pages: 187-200, Elsevier, March 2020.

Read the full publication here:



Back to top