A proposed ECG-based biometric recognition method using a BP neural network achieved high accuracy (>97.9%) for identity verification, with potential applications for wearable devices.
Answers are grounded in this paper and cite their sources.
May advance wearable ECG authentication; leaves open prospective validation before clinical adoption.
As wearable devices gain widespread acceptance among the general population, there is a crying need to ensure that relevant privacy and security vulnerabilities are to minimize the hazards. Biometrics become very important in our daily lives due to its relative security and convenience compared with traditional personal identification data. In this paper, We propose a live biometric recognition method of ECG for wearable devices. Specifically, we locate the feature points and recognize Q-R-S wave, P wave, R wave respectively. Then we analyze and extract 16 individualized features for each complete ECG segment of all subjects. At last, we use BP neural network to get the classifier for randomly selected subjects and identify their identity. The experiment result shows that the proposed method achieves promising accuracy, FRR is only 0.015%, FAR is 0.01%. The classifiers' average recognition rate is more than 97.9% and robust enough in a considerable amount of identities.
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2016) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: