TOOLBOX: Because of advances in wearable biosensors, a contemporary bike owner can seem to be a Formulation 1 automobile within the huge array of knowledge accessible throughout and after a experience. Right here’s a more in-depth look the present state of wearable sensor tech, and a few concerns in deciding what would possibly aid you in well being, coaching, and racing.
As we speak, I’ll undergo some main ideas relating to evaluating a wearable biosensor. This text is NOT a overview of any particular person sensor.
I additionally extremely suggest the 2 articles I’ve positioned within the References part if you need a deeper dive into the rising wearables know-how and their potential in sport science (Ye et al. 2020; Shei et al. 2022).
Correct and/or Dependable?
One of many largest ideas to grasp a few sensor, whether or not it’s a rest room scale or a wearable sensor, is accuracy/validity and reliability/precision. These phrases are sometimes incorrectly used interchangeably or wrongly outlined, so let’s clear this up with an instance of you wanting an altimeter to measure the true elevation on the prime of Mount Everest (8,848 m):
Accuracy/validity: Does the sensor provide the true worth that you just’re attempting to measure? Within the Everest instance, does it really provide the worth of 8,848 m on the summit or does it provide you with 8,200 m?
Reliability/precision: Should you take a number of readings with the identical sensor, does it provide the similar worth? For the Everest instance, a sensor that provides you 8,200 m every time is NOT correct/legitimate, nevertheless it IS dependable/exact.
What do you really need in a sensor? Properly, in fact the best is that it’s each correct AND dependable. However in some contexts, the aim or optimum might fluctuate. For instance, if it’s a coronary heart charge monitor, I’d in all probability place larger worth on accuracy as a result of coronary heart charge by itself can fluctuate vastly day after day (stress, sleep, caffeine, and so on.) and I really wish to know what my coronary heart is doing.
In distinction, if I used to be a rest room scale or an influence meter, I’d in all probability place larger worth on reliability. In these circumstances, I’m solely evaluating myself to myself and I’m extra keen on monitoring variations over time. So I’m extra caring about how a lot weight I’ve gained over a month (e.g., 3 kg) than I care that I weight 65 kg. Equally, if I’m coaching with energy, I’ll care extra that my threshold energy elevated by 20 W over a coaching cycle even whether it is studying 220 W slightly than the precise correct worth of 250 W.
I’ve beforehand highlighted the problem of accuracy and reliability in a video article concerning the CORE physique temperature monitor.
Direct vs Derived Worth?
One other necessary idea to grasp about biometrics is whether or not a sensor is instantly measuring a price, or whether it is taking a instantly measured worth after which placing it via a fancy algorithm to present you a price.
The most typical instance might be coronary heart charge. We all know that coronary heart charge measured by way of a chest strap could be very correct and dependable, and that’s an instance of a direct measure. That’s, if you see your coronary heart charge in your bike laptop, it’s precisely that.
Nevertheless, one problem with many new biosensors is that they take that unique measure after which use software program to attempt to mathematically predict many issues. Utilizing coronary heart charge as a beginning measure, this could embrace coronary heart charge variability, oxygen saturation, restoration, VO2max, and so on.
Why is that this an issue? The additional you get from the unique measure, the extra you’re counting on algorithms and assumptions.
Assumptions are precisely that, they assume that you just fall into the overall center of a inhabitants in response, whereas we all know that particular person responses can fluctuate vastly. For instance, you’ve all probably seen the 220 – age equation on your max coronary heart charge. Moreover the truth that there isn’t any scientific foundation for this equation, this results in the charts at gyms suggesting train zones. Nevertheless, my instantly measured maximal coronary heart charge has all the time been a lot decrease than predicted by this equation by 20-30 bpm, so any predictions for me primarily based on this assumption can be wildly incorrect.
One other downside with algorithms is that they’re by and enormous black containers. Every sensor and firm has their proprietary algorithms, such that we don’t even know what measure or measures they could be utilizing to derive a price, not to mention the equations themselves. That makes it difficult to belief any derived biometric worth for my part like a restoration rating or VO2max.
Who’s the Goal Inhabitants?
A ultimate associated situation to assumptions and algorithms is WHO you’re being in comparison with in that giant database. Understand that most of wearable biosensor growth is pushed from a well being care and scientific crucial, not from an athletic perspective. This may result in a case the place a sensor is first developed for a comparatively passive and sedentary inhabitants, after which laterally “shoehorned” into an athletic software.
The problem with this may be that the sensor could also be very correct and dependable in comparatively sedentary use or with delicate train, however that will not all the time be instantly extrapolated to it being as correct or dependable with heavy train, motion, and sweating.
One other potential problem is that the big inhabitants database that’s getting used for comparability is from a largely sedentary inhabitants with potential scientific points, versus a more healthy and extremely energetic inhabitants. So it might be a case of evaluating apples and oranges.
Abstract
Once more, as a sport scientist, there’s a enormous potential for wearable tech, and this text isn’t meant to disparage the sphere as an entire or any specific unit. What I’ve hopefully supplied as an alternative are three key concerns for you to consider when deciding whether or not to leap in and purchase/undertake one of many many new wearable biosensors popping out onto the market, every with daring claims for a way they may advance your coaching.
Have enjoyable and experience quick!
References
Shei R-J, Holder IG, Oumsang AS, et al (2022) Wearable exercise trackers–superior know-how or superior advertising? Eur J Appl Physiol 122:1975–1990. https://doi.org/10.1007/s00421-022-04951-1
Ye S, Feng S, Huang L, Bian S (2020) Current Progress in Wearable Biosensors: From Healthcare Monitoring to Sports activities Analytics. Biosensors 10:205. https://doi.org/10.3390/bios10120205
Like PEZ? Why not subscribe to our weekly publication to obtain updates and reminders on what’s cool in highway biking?
