Towards Cognitive Authentication for Smart Healthcare Applications

Secure and reliable sensing plays the key role for cognitive tracking i.e., activity identification and cognitive monitoring of every individual. Over the last years there has been an increasing interest from both academia and industry in cognitive authentication also known as biometric recognition....

Full description

Saved in:
Bibliographic Details
Published in:Sensors (Basel, Switzerland) Switzerland), 2022-03, Vol.22 (6), p.2101
Main Authors: Sodhro, Ali Hassan, Sennersten, Charlotte, Ahmad, Awais
Format: Article
Language:English
Subjects:
Authentication
Biological effects
Biometric Identification - methods
Biometrics
Biometry - methods
Brain research
Cognition
Cognitive authentication
Computer and Information Sciences
Computer Sciences
Data- och informationsvetenskap
Datavetenskap
Deep learning
Delivery of Health Care
EEG
Electroencephalography
Healthcare
Humans
Identity
IoT
Natural sciences
Naturvetenskap
Passwords
Physiological effects
Physiology
Privacy
Sensing
Trends
User needs
Wireless networks
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Secure and reliable sensing plays the key role for cognitive tracking i.e., activity identification and cognitive monitoring of every individual. Over the last years there has been an increasing interest from both academia and industry in cognitive authentication also known as biometric recognition. These are an effect of individuals' biological and physiological traits. Among various traditional biometric and physiological features, we include cognitive/brainwaves via electroencephalogram (EEG) which function as a unique performance indicator due to its reliable, flexible, and unique trait resulting in why it is hard for an un-authorized entity(ies) to breach the boundaries by stealing or mimicking them. Conventional security and privacy techniques in the medical domain are not the potential candidates to simultaneously provide both security and energy efficiency. Therefore, state-of-the art biometrics methods (i.e., machine learning, deep learning, etc.) their applications with novel solutions are investigated and recommended. The experimental setup considers EEG data analysis and interpretation of BCI. The key purpose of this setup is to reduce the number of electrodes and hence the computational power of the Random Forest (RF) classifier while testing EEG data. The performance of the random forest classifier was based on EEG datasets for 20 subjects. We found that the total number of occurred events revealed 96.1% precision in terms of chosen events.
ISSN:1424-8220
1424-8220
DOI:10.3390/s22062101