Because "pulse rate" is nowhere near synonymous with "electrocardiogram," the headline in the article linked above is wrong and a bit misleading. The concept described in the article, however, is very interesting.
Pulse rate is a number, usually expressed as beats per minute (bpm). The average resting heart rate for an adult is 60-90 bpm. An individual's pulse rate varies not just with levels of activity or excitement; it also varies depending upon whether the person is inhaling or exhaling at the time. Trying to account for individuality among the entire adult population with a simple biostatistic that varies within an individual every few seconds, but among the whole population by only 30 bpm is absurd. Pulse rate is wholly unsuited to biometric identification.
But what about ECG?
The electrocardiogram (loose translation: electric heart writing, or ECG) is a far more detailed representation of what the heart is up to. It looks like this:
 |
| Photo: Wikipedia |
This graph shows the electrical activity associated with two full heart beats, and it may be possible that the formula describing an individual ECG will turn out to be unique enough and stable enough over time to use as a biometric identifier.
The other good news is that there are probably millions of recorded individual ECG readings for biometric algorithm designers to work with. Even people without fingers have heartbeats. Another awesome upshot of an ECG biometric would be the promise that that, one day, your phone or laptop might be able to tell if you're having a heart attack. Unfortunately, the good news pretty much ends there.
The bad news:
Healthy hearts all look very similar on an ECG, which is part of what makes the ECG a useful diagnostic tool.
Where on the body ECG sensors are placed causes changes in the observed wave, mostly on the Y-axis.
The wave gets compressed or stretched horizontally (X-axis) based upon heart rate.
So, to summarize these first three points, the ECG, like heart rate, may vary by too much within an individual but not enough over the healthy population to make it a useful identifier.
The biometric is actually a biostatistic, which is problematic (as linked above).
An ECG is medical information, which has accompanying regulatory and privacy issues.
It's going to be difficult for this type of modality to displace other hand-based biometrics which have a huge head start in terms of price, proof of reliability, education and acceptance.
In short, this is one of those subjects that is intensely interesting from a Ph.D.'s point of view (invention) but not so much from an engineering or business perspective (innovation). ECG as a biometric will face significant — I dare say insurmountable — challenges in finding its way into wide use as a commercial ID management application any time soon.
