Viewing entries in
HIIT

Apple Watch Heart Rate Accuracy: Part 2

10 Comments

Apple Watch Heart Rate Accuracy: Part 2

Now that the Apple Watch is out, I figured I’d follow up my previous post on Apple Watch Heart Rate Accuracy.  Apple has recently posted a bit more information on their site on the heart rate functions on the Apple watch and in a nutshell “it kind of works, but if it doesn’t work during your specified activity, you can connect another heart rate sensor to the watch to get an accurate heart rate.”  So, they make it clear – the Apple watch is really good at measuring activity, it probably gives you a decent heart rate measurement every 10 minutes using a low power infrared sensor, and when there is more movement detected, the green LEDs fire up and give you a heart rate for exercise.  If you are not receiving an accurate heart rate, make sure your wrist strap is tight enough and if it still isn’t accurate, then it is probably best to just connect an external sensor.  In other words, it might not be very accurate.

Hmmm… I hate to say it (sort of) – but I told you so.  Let’s take a quick look at why it struggles. Full disclosure - As I stated in my original blog, I test for Scosche on their optical heart rate armband and Apple is having to jump similar hurdles to the ones we had to in the 1st generation of Scosche optical heart rate sensors 4 years ago!  

#1: All in one sensors are not great at doing any one thing.  

A specialized single function sensor will almost always outperform a one size fits all sensor.  A machete will cut through the forest better than a Swiss Army knife.  Not an easy task.  Similarly, the Apple watch is the Swiss Army knife of watches, attempting to cut through a thick forest one branch at a time.  It’s just not an adequate enough tool for the job.  In reality, to get accurate heart rate data, you need a specialized, specific sensor such as a chest strap or armband such as a Rhythm+ to do the job correctly – a machete.   We could have put more features in the Rhythm+ at the expense of a great heart rate experience, and I’m glad we didn’t.  

#2 Construction:

The Apple watch is a giant, heavy mass of a watch that moves with the inertia of its’ high center of gravity and fashionable, comfortable wrist straps.  The ideal watchband for an accurate heart rate monitor should be flexible and stretchy.  The flexibility and stretchiness of the watchband allows the sensor to follow the movement of the arm, expand and contract with your muscles under constant tension, and dampen vibration.  The Apple watch does not do this, and how could they?  It is impossible with a metal or rubber strap – they are simply not elastic.  Along with the high center of gravity there’s too much movement on the skin for an accurate measurement.  It looks pretty though.  We spent a lot of time on this at Scosche to keep a low center of gravity and test the right materials to get an accurate, consistent signal.

#3 Optics

Apple is correct about green light being absorbed by the hemoglobin in blood, however certain pigments in darker skin tones can impede green light.  The Valencell PhD’s (whose Performtek sensor technology is in Rhythm+) discovered yellow light is less sensitive to skin pigment.  The best optical sensor will have a combination Green + Yellow light for the most accurate signal on all skin tones.  

#4 Algorithms

It is now clear that the new Apple Watch uses dynamic algorithms to compensate for the design errors in points 2 and 3.  A dynamic algorithm changes the colors, intensity, and sampling rate based on the signal integrity measured by the photosensor(s).  The best algorithms in the world cannot make up for poor signal quality.  A fixed algorithm on the other hand uses the optimal intensity, sampling rate, and a color usage and these parameters are locked from the moment the sensor is powered on.  In my experience with Scosche’s testing, the engineers and other smart people in the room have all agreed using dynamic algorithms are a bad idea and prone to infinite loops like a dog chasing it’s tale.  The human body can provide a dynamic response with endless parameters, which can change at the same time, all the time.  There are simply too many possible scenarios for the algorithms to figure out what is going and to be accurate in all situations, especially taking in to account the first 3 hardware hurdles mentioned above.  

I have still yet to fully test the Apple Watch, but after reading their latest information about the heart rate functionality, I have even less confidence in the ability for the Apple Watch to provide accurate heart rate data during most types of exercise.  It may even be difficult to export the heart rate data to compare accuracy versus other products!  As I said before, if anyone can do it, Apple can, and I hope they DO!  This product is still in its infancy, and it most certainly will evolve with generations.  For now, as far as heart rate training goes, we are left with a neat looking toy to send our significant others the pitter patter of our heart beat.  Awww.... how cute.  

10 Comments

Apple Watch Heart Rate Accuracy

20 Comments

Apple Watch Heart Rate Accuracy

I have been working with Scosche Industries as a consultant, assisting Scosche in testing and the development in their optical heart rate monitoring starting with MyTrek, then Rhythm 1.5, and now our latest and greatest optical heart rate monitor, the Scosche Rhythm+.  Over the 3 generations of Scosche heart rate monitors I have gotten a pretty good idea of what is needed to make optical heart rate monitors accurate, and equally important what makes accuracy fail.  I intend to give Apple the benefit of the doubt when it comes to their development of an accurate wrist solution for continuous heart rate monitoring in the Apple Watch, although I see some potential “hurdles” to overcome.  If anyone can overcome these “hurdles” it will be Apple, and I can’t wait to see how they deal with these potential issues.

Hurdle #1

The Apple watch is a substantial device in weight with a high center of gravity - especially compared to the Rhythm+    

I’m using the Rhythm+ as a benchmark for optical heart rate accuracy since it is the most accurate product on the market that I’ve tested.  I do know this from my testing - More weight and a high center of gravity = more movement on the skin which will affect the accuracy of the signal generated from the photosensor(s) on the bottom of the Watch.  This will not be a problem when you are sitting still at the desk, intimately “texting” a live heart beat to your girlfriend (aww), but it will certainly be an issue when you are exercising.  For example, every bump you hit on the bicycle or movement of the wrist can move the heavy watch and throw off the photosensor’s accurate signal and thus impact accurate data collection.  

Hurdle #2

apple-watch-tap-1500x1000.jpg

Will a magnetic wristband be tight enough to keep light out?

After testing products like the Mio Link/Alpha which are wrist monitoring optical heart rate monitors, I know you need to get the wrist monitor very tight - and nearly cut circulation off at your wrist to get a good signal.  Tim Cook was actually wearing the Apple Watch above the wrist bone similar to where Mio recommends.  Even above the wrist bone there’s not much meat, and lots of little bones that distort the photosensor’s clear signal and create noise, thus throwing off the algorithms.  Bottom line is, that and optical heart rate wrist band needs to be tight, especially because of Hurdle #1!  I have gotten great results testing the lightweight, low center of gravity Rhythm+ on the wrist, but on the forearm or upper arm is the most accurate location due to a clear signal that is generated on a meatier location on the body.  Less noise in the signal = more accurate heart rate #.  The forearm also has a larger diameter and we have had great success with the Rhythms’ breathable, flexible velcro strap for both accuracy and comfort.

Hurdle #3

2 Green LED’s vs. 2 Green + 1 Yellow LED

In our testing at Scosche with Valencell, we found that to get a good signal on ALL skintones Valencell added a Yellow LED which penetrates deeper into the skin than green LEDs.  Green lights are most effective for measuring just below the surface of the skin and have been proven to NOT work very well when used alone with darker pigmented skin. When used in combination with the yellow LEDs, the green and yellow tones combine to accurately read through all skin tones.  From what I can see, the Apple watch only has 2 green LED’s so it is unclear to me if it will be accurate with all skin tones.  

Hurdle #4

Keep Ambient Light/Sunlight OUT!

Heartbeat.jpg

As described above, it is really important to keep sunlight and ambient light away from the photosensor’s sight.  The Rhythm+ uses a “gasket” that presses the photosensors and LEDs into the skin .5mm to bury the photosensor into the skin along with the LEDs.  That way, the photosensor is reading a signal that is illuminated by the LEDs, not the fluorescent light bulbs in the gym, or the sunlight outdoors.  The bottom of the monitor must also have a sufficient footprint to block ambient light, therefore thinner optical monitors have shown to be less accurate.  I have not had the chance to test the Apple watch, and it is unclear to me if they have figured out how to block ambient light efficiently enough to gather good data.  

Is the Apple Watch Designed to be a Continuous Heart Rate Monitor for intense exercise?

watchactivityapp.jpg

The way I saw it, In Tim Cook’s presentation, the Apple Watch was shown to be an excellent Activity/Life monitor, where movement, lifestyle, and general health were emphasized, and not heart rate training.  I could be wrong, but I think Apple knows heart rate is not going to be super accurate during exercise with the Apple Watch, therefore they built the Watch around making a great interface for activity monitoring - IE: how often you stand up throughout the day, pedometer information, etc.  This will surely play out to be a thorn in the side of FitBit and other activity monitors, but I don’t think it poses a threat to serious heart rate training athletes.  

Data Centralization

I think Apple knows their job is to centralize the data collected from accurate sensors, not to necessarily create the sensor itself - except for activity monitoring.  For example, Health Kit will allow all connected apps to push data that has been collected from different sensors directly into Apple’s Health App.  I don’t see Apple getting into the Blood Pressure monitor or blood glucose monitor business any time soon, but apps that monitor blood pressure and blood glucose can push their data to Apple Health.  I DO like the way the Apple Health App will centralize that information into one health app, along with other health related metrics.  It makes sense to have all that data in one centralized location.  The bottom line is that you should be able to use other heart rate sensors to do your workout and gather accurate data, you won’t be confined to only using the Apple Watch to collect accurate heart rate data.

Benefit of the Doubt.

As I stated in the beginning, I am willing to give Apple the Benefit of the Doubt.  If anyone can do it, Apple and their unlimited resources can solve these potential issues.  For one, the Apple Watch might have 2 photosensors.  There appears to be 4 circles on the bottom of the Apple watch - 2 Green LED’s and 2 photosensors?  Along with super fast processing power the Watch might be crunching more data in “stereo” via an extra photosensor.  They certainly could have hired enough talent to write some super sophisticated algorithms to get an accurate heart rate number.

I don’t know what they have up their sleeve (pun intended) but I do think they know it is not going to be a great heart rate monitoring solution for serious athletes.  Only time will tell.  For the time being, I'll stick to my Rhythm+.

20 Comments

Gymnastics Strength Circuit

Comment

Gymnastics Strength Circuit

I’m getting ready to hopefully do the Illinois Gymnastics strength circuit next week, so I’ve been training! Today I’m going to go through some 20/20 timed intervals and I’ll go through these 10 exercises which are similar to what the Illinois gymnasts do for their circuit.  This should help me get ready for the butt kicking I'm about to receive.  Looks something like this - 

Comment