A study conducted by the Carlos III University of Madrid (UC3M), in collaboration with the Shahid Rajaee Teacher Training University of Iran, concludes that the heartbeat can be used as a biometric tool to identify individuals.
Biometrics is the science that studies the identification of humans & animals based on biological measurements or physical characteristics, such as fingerprints or the iris of the eye. With this in mind, biometrics-based tools are increasingly being used in fields such as security to-supplement or replace the password systems; and in the field of civil administration in the registration & provision of identity documents.
This research proposes an innovative technique for identifying people based on the exclusive characteristics of their heartbeat. It does this by using electrocardiograms (ECG) & analyzing 5 musical qualities: dynamics, rhythm, timbre, pitch & tonality, which are commonly used to characterize audio files by applying them to the sound of heartbeats. Thus, from these 5 qualities, a combination of parameters is obtained that is unique to each person and with an accuracy rate of 96.6%.
“Biometric identification based on cardiac recording has been studied for years and is found to be effective. The main novelty of our work is that we look at the ECG recording, which is a temporary signal, as if it were of a sound wave. From there, we analyze this sound wave using the qualities that are commonly used to characterize music”, explains Carmen Cámara, researcher in the computer science department of UC3M.
The main advantage of this technique is the universality of its identification, since, even today, some people cannot be recognized by certain types of biometrics – in the case of injuries, amputations or disabling physical characteristics – but the heartbeat is a biosignal that is present in all human beings, without exception. Another advantage is its inexpensive & non-invasive operation: “There are already smart bracelets & smart watches today that perform ECG recordings, so it would be enough to install an application on them that uses our identification algorithm”, explains the researcher Pedro Peris López, also from the computer science department of UC3M.
This technique is currently under development. While the future of cardiac identification is promising, the researchers say they need to “continue this line of research before considering commercialization.” One of the critical aspects of this study is to analyze the behavior of the system according to the different activities that the person performs, such as walking, running, resting, exercising or being in a stressful situation. In addition, factors such as the use of a pacemaker or the effects of arrhythmia should be considered.
Age is also a factor to take into account: “Because of our aging, the signal changes slightly over-time. This means that our system must be updated approx. every 5 years”, explain the researchers.
The research published in the Biomedical Signal Processing And Control.
