English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

This study aims to develop a respiratory-based biometric approach that predict the identities of individuals using electromyography (EMG) signals generated from diaphragm movements. Therefore, the examination of these signals is effective and offer a development of biometric systems. Researchers use bioelectric signals for biometric authentication. Electromyography (EMG), one of these signals, is a muscle-based signal used in many areas. This signal provides biometrics such as taking precautions against forgery and meeting security requirements. To the best of our knowledge, this is the first study to investigate a respiratory-based biometric approach using EMG signals from a single-lead generated by diaphragm movements. Statistical feature extraction techniques, including RMS, STD, MAV, and VAR, were performed usingshallow classifiers. Moreover, Recurrent Neural Network (RNN)-based deep learning models were also implemented after the time-series data augmentation process against the overfitting problem and multi-session issues of data recording. As known, deep learning models need high-size data to obtain better performances in pattern recognition. As far as we know, this study is the first attempt to investigate data augmentation with RNN-based deep models for the diaphragmatic respiratory movement-based biometric system performance. In this study, the highest accuracy of 90.9% was obtained for training models using the EMG data of seven people in statistical feature-trained shallow models. Furthermore, RNN-based deep models have reached up to &gt;96% accuracy in the testing phase. The framework of a Probabilistic Neural Network with a 2D-reshaped pattern has the greatest correct recognition rate over &gt;99 accuracies among seven people in a diverse cross-validation strategy.

Novel Biometric Approach Based on Diaphragmatic Respiratory Movements Using Single-Lead EMG Signals