Key result
ECG biometric validation model using LS-SVM achieves 95% accuracy for individual identification.
Why the study?
Passwords have major flaws in speed and security for device data integrity, driving the need for biometric authentication using unique, hard-to-falsify ECG signals.
A novel machine learning model using LS-SVM and frequency domain characteristics of ECG signals was developed for biometric authentication.
May support ECG biometrics for identification; leaves open prospective validation before clinical adoption.
The development of authentication and identity mechanisms has become a vital requirement to secure device data integrity, although passwords provide sufficient control and authentication, they have shown major flaws in speed and security, Biometrics is becoming the primary authentication method. Electrocardiogram (ECG) signals are created as a consequence. Due to their unique character, which makes them difficult to falsify and ubiquitous, ECG signals have attracted much attention in most authentication systems. We introduce a novel ECG validation model that combines frequency domain-based characteristics with the least-squares technique in this work (LS-SVM). Two types of frequency field characteristics were examined in order to find the best combination of high precision and speed. ECG Signal Characteristics and Frequency Domain Characteristics ECG signals are used to determine the optimal triple-band filter bank. To access the most important characteristics, we removed the extraneous features and extracted others.
No takes yet. Share an insight, caveat, or question.
Majeed et al. (2022) studied Healthy individuals (ECG biometrics) (n=47). Least Square Support Vector Machine (LS-SVM) with frequency domain features was evaluated on Classification accuracy for individual identification. The proposed ECG biometric validation model using frequency domain features and LS-SVM achieved an accuracy of 95%, sensitivity of 92%, and specificity of 92% for individual identification.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: