If you were wondering whether the sophisticated heart-rate monitor in the Apple Watch does actually generate more accurate results, the answer appears to be a pretty resounding ‘yes.’ Software engineer and SonoPlot founder Brad Larson extracted the raw data from his Apple Watch and compared it to that generated by a dedicated heart-rate monitor. The results can be seen above.
The comparison device, a Mio Alpha, has itself been shown to have 99% accuracy when compared to the gold standard of chest strap monitors.
A teardown of the Apple Watch revealed that it contains a pulse oximeter, suggesting that future updates or models may be able to measure the oxygen content of our blood, useful data for both health and fitness measures. A separate chip analysis reveals plenty of scope for both performance and energy efficiency gains in future models.
Via iClarified
FTC: We use income earning auto affiliate links. More.
Of course it’s accurate, it’s only detecting the timing of a bump on your wrist. You can measure this accurately using your fingers and a watch, so no reason why a 300 pound machine should have trouble doing it either.
Joe, it isn’t sensing pressure in your artery (the way you detect a heartbeat by feeling on your wrist). It is using spectrophotometry–sending a light of a particular frequency (or several different frequencies) and measuring the absorption of that light against time. This is actually a far superior way of detecting heart rate, since not all types of heartbeats give the same “pulse” of blood, depending on whether they are firing on-time or prematurely. (The detailed explanation is beyond the scope of the discussion, but suffice it to say that your palpable pulse can underestimate your heart rate, sometimes very significantly).
Not to throw the conversation any further off track, but the heart rate palpated on your radial artery should match your heart beat exactly. Any noted difference between your apical and radial pulses is an oddity and may be a sign of a serious heart condition for which you’ll want to see your doctor as soon as humanly possible.
samuelsnay, I wasn’t going into a long discussion of the medical implications of a discordant pulse, just that there could be one, and using a photoplethysmogram is more accurate that relying on a tactile impulse. If the Apple watch is ever used as a more sophisticated medical device, then accuracy and precision are both very important.
I agree that a discordant pulse CAN be a serious finding and requires evaluation by a medical professional, though there are plenty of people with occasional premature heartbeats that cause discordance but are completely benign.
Back on topic, it speaks well of the future uses of the Apple Watch that it uses the technology it does.
Yes, and you can do that WHILE you are exercising.
Except the watch rests on TOP of your wrist, not on the bottom where the major arteries used for measuring heart rates by hand are located. Last I tried, it’s pretty difficult to measure heart rate from the top of my wrist by pulse.
When I am working out in the gym the heart rate monitor doesn’t always work. For example, after an aerobic exercise, the rate was 64 bpm. Just can’t be right. Or it says, it is “measuring” and the result may or may not be accurate. I have to use the “Other” workout because there is no activity for gym workouts or even tennis.
That is great news. Since I just finished a quick speed walk around my office building.
This is great to know! Thanks Ben!
Made the pulse oximetry predication months ago, looks like they’re letting everyone get used to the basics before either software activating it, or leaving it until Gen 2. Be a cool feature to switch on to add to the Gen 1 uses though, here’s hoping :-)
They may not have implemented that function because of the need for FDA approval as a medical device. Besides, how would you use the information? It might be useful to record in HealthKit, to be later interpreted by your physician, but is of minimal use to the layperson.
An app for that.
Hi Ben, this is a great article. My primary reason for purchasing my Apple Watch is for fitness monitoring. Its also nice to see my heart rate logged on my iphone in the Health App. Good to know the data is accurate.
Today was my first day using my Apple Watch for exercise. I use a scosche rhythm plus heart rate monitor and was disturbed when my Apple Watch only logged half the calories for both my gym workout and my daily 4 mile walk. Now I don’t know which is accurate. My mio heart rate monitor and my rhythm plus monitor run about the same????
What about the pedometer to distance traveled? I ran this morning without my iPhone and the distance it said I traveled was not accurate. Can I calibrate it?
Yes, you can, run with your iPhone and the Watch to calibrate it.
Yes, like rnc said, the iPhone will use GPS information and the background software will match that against your steps. After a few runs / walks, it’ll have an average distance per stride and then will be more accurate without the iPhone. I’ve gone on many walks with both and then, without even knowing it would do this, found the Watch alone gave me a pretty accurate reading of distance after I double checked it.
There is a procedure for calibrating the Apple Watch. You can Google for the specific instructions, but there are a couple settings that have to be enabled on your phone (they were enabled by default for me) and you must use the Workout app on the watch and have your phone with you. Your watch will automatically use your phone’s GPS data and approximate your stride based on accelerometer data and distance traveled. You’re supposed to accumulate 20 minutes of walking/running this way in order to calibrate the watch. To be safe, I accumulated over an hour before running without my phone and the results were impressive. It actually seemed more accurate than GPS.
My last run was a 5 mile track run and every single lap was .25 miles at the same spot, give or take a few steps. Even after 19 laps, my run ended almost exactly where I started. When I used RunKeeper, I would notice each lap would become shorter and shorter and by the time I was finished, RunKeeper might have me cut a half a lap off my run.
They’d sell a lot more watches if it could measure the alcohol level in your blood… Just saying. ;)
That would be awesome!
“this is Siri, master you are drunk and I detected you are trying to drive. I’m going to call your mom on speakerphone”
LMAO
NICEEEE ty for this article!, i have always read that wrist HRM are not very good, but apples looks bigger and more sofisticated than others, like samsung. I would love to see some comparison on pedometer and HRM. =) plz do it hehe and sorry my english
Don’t worry, next year all the others will look and work just like the Apple Watch. :)
You comment was posted 11 months ago – today Fitbit released the “Blaze” and it looks just like the Apple Watch. Your prediction was perfectly accurate! Well done. :)
I’m not sure if it is a good idea to be shot every 10min by Infrared LEDs deep into your skin all day long. I’m not that much interested in these stats. Not to mention other potential health risks for being exposed to 2.4GHz radiation so close constantly.
These types of monitors are used continuously for days or weeks on patients in intensive care units around the world. You are exposed to far more infrared radiation of much broader frequency ranges by simple sunlight. As IR goes through most glass, you even get the exposure indoors in many cases. While it makes sense to be cautious about your health, many before you have asked this question prior to such devices being available for use. Don’t worry about it.
I’m assuming we haven’t see this video yet … http://youtu.be/qTFWXKdWUEs
I was in Boston recently; I have a paroxysmal afib w/ paroxysmal tachycardia. My pulse rate on 10/10/15 was in the 150’s and the watch did not capture it at all. Sad.
ZoeBushey – this is why we would buy it, for keeping tabs on tachycardia episodes. It totally missed it? Yikes.
Atrial fibrillation with a rapid ventricular response is not appropriately measured/monitored with a photoplethysmogram. The nature of atrial fibrillation is that the beats are irregular, and with a very fast rate the individual beats will not reliably perfuse the extremities, at least not to a level that can be consistently detected by an external device. It is a case of having the wrong tool for the job, rather than the tool functioning incorrectly. The best tool for the job is an implantable loop recorder–this measures continuous cardiac electrical activity (ECG) and will readily detect A-fib with RVR.
My experience has been that it is very accurate at rest and during aerobic exercise (walking/jogging on treadmill and road) but when I exercise with free weights /strength train it grossly underestimates, and consistently does this – possibly by exactly half? Not sure. Have tried tightening strap, adjusting strap. No improvement. I have had the watch for about 3 months now and exercise several times a week. Anyone else getting this issue?