A recent study published in Scientific Reports suggests a new form of biometric verification that examines beat-to-beat heart signals may be possible.
Unlike several stationary physiological signals like fingerprint or facial recognition, hand-written signature, or voice analysis, the beat-to-beat information embedded within the time intervals between consecutive heart beats is non-stationary. On the basis that this reduces the ability to forge, the study looked at a new method of biometric verification termed “CompaRR”.
The research marked this as a preferrable method of personal identification to electrocardiogram (ECG), a method used over the past decade to register of register electrical activity of the heart.
It said: “The heart beat-to-beat time intervals may reflect the uniqueness of the heart’s physiological system and can be measured by compact wearable devices such as watches or wristbands and even remotely by video camera. Moreover, it substantially reduces the data dimensionality needed for verification in comparison to ECG.”
To examine the effect of age on biometric verification performance heartbeat was extracted from longitudinal recordings from 30 mice ranging from 6 to 24 months of age, or 20 to 75 in human years. Fifty heartbeats, which is close to resting human heartbeats in a minute, were deemed sufficient for the verification task, achieving a minimal equal error rate of 0.21.
When trained on 6-month-old mice and tested on unseen mice up to 18-months of age, or 50 human years, no significant change in the verification performance was noted.
Considering the fact that medications are commonly used in everyday situations the research also examined the effect of drug administration on biometric verification performance. In this case, drug-treated mice were used. Among the study’s results, it was noted that even when using this model verification was still possible.
In the study’s concluding notes, the author’s remarked, “This work demonstrated the feasibility of a heartbeat interval-based biometric verification method under real world limitations. The system termed “CompaRR” can perform verification on subjects whether they were part of the training process or entirely new and added to the database after training.”
They added: “In addition, the method performed well whether the calibration and test were performed with or without drugs. Therefore, the method can be used as complementary biometric agent for other widely-used biometrics for verification tasks.”
