Health Care : Current Reviews

Privacy preserving health care using wi-fi-based deep learning techniques

International Congress on Global Healthcare

August 24-25, 2023 | WEBINAR

Seyed Ali Ghorashi

University of East London, UK

Scientific Tracks Abstracts: Health Care Curr Rev

Abstract:

Statement of the problem: The Internet of Things (IoT) consists of things or objects which are connected to the internet and exchange data with each other over the network. IoT can be utilised in health care to help monitoring the patients and old adults and new techniques such as vision-based, skeleton-based, and contactless Human Activity Recognition (HAR) have been introduced and accordingly, different devices such as wearable or environmental sensors, cameras, and smartphones have been developed in the research community as well as the market. Contactless HAR using Wi-Fi have recently attracted much interest compared to the other techniques because of its simplicity, availability, scalability, affordability, and most importantly, its privacy preserving characteristics, however, the accuracy of this technique needs to be improved to be used in health care applications properly. In this talk, a summary of my recent achievements in this area will be presented[1-5]. Methodology & theoretical orientation: Wi-Fi based HAR is conducted using one of the two channel metrics of Received Signal Strength (RSS) or Channel State Information (CSI). Deep Learning (DL) algorithms such as Recurrent Neural Network (RNN), Long short-term memory (LSTM), and Convolutional Neural Networks (CNN) as well as different pre-processing algorithms to convert sensory data into images are used for the classification of the received channel metrics. Findings: A new CSI dataset is collected, and four artificial neural networks’ performance are compared in terms of complexity and accuracy. We also showed that pre-processing can enhance the performance of the classifiers, considerably. Conclusion: Deep Learning techniques can boost the performance of indoor health monitoring systems to help meet the requirements for acceptable recognition accuracy and recognition speed for the activities that have been investigated in this research.

Biography :

Seyed Ali Ghorashi received his B.Sc. and M.Sc. degrees in Electrical Engineering from the University of Tehran, Iran and his Ph.D. degree from Kings’ College London, UK. He has worked for Samsung Electronics (UK) Ltd, Middlesex University, Shahid Beheshti University and University of East London. He is a senior member of IEEE & BCS, holds international patents and has published over 130 technical & peer-reviewed papers mainly related to the applications of optimization, artificial intelligence and machine learning in localization, human activity recognition, internet of things and wireless communications.