Heart Monitors and the Limit of Self-Knowledge
February 19, 2008
The heart rate is among the earliest biometrics used by humans to take stock of themselves. Before mechanical clocks were invented, this was hard. The first doctor credited with making objective measurements of the pulse was an Alexandrian physician named Herophilos, from the 3rd century, B.C., who used a water clock as his chronometer. Using a specified outflow of water to set the time interval, he counted the heart beats of four healthy individuals of different ages, which gave him a base rate against which to compare the pulse of his patients. Genius!
It’s easier to find our heart rate now. In fact, it’s so easy it’s become complicated again. What you used to be able to do with two fingers and a second hand now requires a cardiac monitor, sometimes with a chest strap and a wireless connection to the wearable computer on your wrist. These complications are associated with bigger benefits; we can correlate our heart rate with our exercise regime, for instance. There are more than 700 heart rate monitors listed on Amazon.
Many of the best monitors require a chest strap. Why not just put this in our clothing? The one pictured below is from Numetrex.
Pacing exercise is just one of the things we might want to do with data about the rhythm of our heart. The pattern of the heart beat is a clue to health, and, ideally, it would be tracked all the time. For monitoring serious conditions, how about if we move the monitor from our wrist or our clothing, and put it inside our body? The image below is of a tiny cardiac monitor the size of a small memory stick. It is implanted in a patient’s chest, and recorded measurements can be picked up from the outside.
As we generate more data, the patterns become too complex for our brains to recognize. ECG measurements have to be read by trained physicians. Or by artificial intelligences. While reporting an upcoming story in Wired about the great inventor Ray Kurzweil, who is best known for his reading machine and his theory of the singularity, I found how that his company has also been involved in researching the use of artificial intelligence for the interpretation of ECG. His friend Martine Rothblatt, the founder of United Therapeutics hired Kurzweil’s company to contribute some improvements to the algorithm underlying CardioPal, a 24/7 cardiac monitoring system designed to provide early warning of arrhythmias. CardioPal is produced by Medicomp, a United Therapeutics subsidiary. The underlying algorithm is named Diogenes.
There is a lot of interesting science behind the interpretation of ECG, and it is easy to imagine a not-too-distant era when internal cardiac monitors are a normal health maintenance device, automatically warning of impending problems. The curious thing about this vision of a totally monitored future is that the algorithms that interpret data from these monitors inevitably becomes more and more complex, easily outstripping our capacity for unaided interpretation. We will get a warning of impending doom, but not fully understand why this warning is issued. We will gain more power over ourselves, but not more self-understanding. Maybe we have to adjust our idea of who we are. The artificial intelligence upon which we rely – can this be understood as part of our self?