Photoplethysmography Biometric Recognition Using Deep Learning

With the alarming rise of cyberattacks every year, securing electronic data has become an urgent matter. This study explores the possibility of using Photoplethysmographic signals (PPG signals) as a biometric method for data protection purposes. Our dataset was sourced online and consists of PPG rea...

Full description

Saved in:
Bibliographic Details
Main Authors: Cherry, Ali, Abbani, Mohammad, Sleiman, Ali, Hage-Diab, Ali, Hajj-Hassan, Mohamad
Format: Conference Proceeding
Language:English
Subjects:
(BiLSTM) Bidirectional Long Short-Term Memory
Biological system modeling
Biometrics (access control)
Deep learning
Distance measurement
Photoplethysmography
Security
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the alarming rise of cyberattacks every year, securing electronic data has become an urgent matter. This study explores the possibility of using Photoplethysmographic signals (PPG signals) as a biometric method for data protection purposes. Our dataset was sourced online and consists of PPG readings from 34 healthy individuals ranging in age from 10 to 75 years old, with 78 PPG readings each for training and testing purposes. To develop accurate models for testing, we utilized advanced pre-processing techniques along with Bidirectional Long Short-Term Memory (BiLSTM) deep learning algorithms. We were able to achieve a 100% accuracy during training and up to 95.26% accuracy during testing, clearly demonstrating the effectiveness of our approach. Harnessing PPG signals for biometric recognition through deep learning machines has excellent potential for enhancing security measures across various practical applications.
ISSN:2377-5696
DOI:10.1109/ICABME59496.2023.10293049