Next generation wearables for health monitoring: an interview with Dr Steven LeBoeuf

insights from industryDr Steven LeBoeufPresident and cofounder,
​Valencell, Inc.

In what ways are current wearables limited and what do consumers want to see from next generation wearables?

Wearables have proliferated in the marketplace, but there's still a lot of opportunity for growth. One of the ways in which wearables are limited is that people expect their wearables to really measure what they feel is important to be measured. When they find out that wearables are not working as they expected, they start to lose faith in them.

We saw this in our recent survey; one of the big reasons that people dropped their wearables was accuracy concerns. Another big issue was battery life. When you make wearables that are annoying to wear, not comfortable or where the reason to believe in the data and insights from the device has gone, then those wearables just simply don't cut it for the consumer.

One of the challenges for those who want to make consumer-directed wearables, whether that be consumer fitness, consumer sports or consumer medical, is that any of those types of wearable devices need to be comfortable.

The user experience needs to be continually interesting, and wearables also need to accurately measure what they're supposed to measure, so that the user experience has a meaning. Otherwise, people grow weary of those devices.

If the wearable’s sensor data is inaccurate, then the user experience built upon the data will be meaningless and uninteresting. A wearable that doesn’t measure biometrics accurately can still have a great out-of-box experience, but it’ll end up in the sock drawer as the product has no basis for generating actionable personal insights.

What we were also able to uncover in our particular survey was that about half of the people did not own wearables. It was an online survey, so it was already slanted towards people who were technically savvy.

Even in our tech-savvy group, more than half of them did not own wearables and one of the big reasons they didn't was because they didn't believe the wearables had applicability towards health.

If the accuracy was strong enough to provide some health outcome or health monitoring, that would have been something that would have made them reconsider wearing a wearable device.


By how much do current wearables vary in terms of accuracy and what advances in technology are needed to improve accuracy further?

This has been the focus of Valencell’s R&D for quite some time, and we’re the industry experts when it comes to accuracy in wearables  At Valencell, our hypothesis from the beginning was, that the only way to really make wearables work ultimately is to help people get healthier and live longer, healthier lives, with the devices they already wear.

We knew early on that, in order for that vision to be realized, the devices had to be accurate because all the models that use heart rate, breathing rate or other vital signs in order to predict health outcomes, rely on accurate data.

It’s really important to not only focus on the accuracy of the device, but to validate it to make sure that it truly is accurate. One mistake a lot people make is that they might prototype something, test it on a few people and find that the data looks really good.

They may even think it is working much better than they thought it could, and that their job is done! Then they’ll begin manufacturing their wearable, without proper validation testing, and sadly find that it only really works on a certain percentage of the population.

For example, the prototype may have worked great on 4 or 5 people, but the final product may only work on  70% of people, which is a catastrophe in consumer electronics and health products.

Therefore, it’s important to focus not only on the sensor accuracy, but on the validation. What we're seeing now is that there's a broad range of accuracy in these devices.

Some devices, for example, don't fit the definition of a heart rate monitor. Namely, some of these devices may work OK during resting or during low activity, but they fail during exercise, where accurate heart rate is needed the most for fitness training.  

By that definition, a lot of these devices are not true heart rate monitors because they fail when you need them most. If you’re looking for a sports monitor, make sure the product you buy has been independently validated in a sports environment.

If you’re looking for health monitoring, make sure the product has been approved for a medical use case. In each case, there should be a link to a study proving why you have a reason to believe the product will really work.  


If someone already owns a wearable, or if they're thinking of purchasing one, where can they find out information on how accurate they are in terms of heart rate monitoring, for example?

This is a big problem and there's really no good, solid answer to it today. The problem is that there are no independent organizations now that are broadcasting a standard.

It's important to say broadcasting because there are independent organizations that have standards, but they don't really broadcast and they haven't been policing.

It is very difficult to do this, but here's a rule of thumb you can use until we have some type of standard agreed to that is broadcast and some kind of seal of approval that you can count on: that is to find out whether the company's product has independent validation results.

There are different ways of doing it. One is to simply go to their website, but the problem there is that some of them are so ashamed of the accuracy level, that they don't put it on the website.

Alternatively, you can contact them and ask them for the independent study. If they don't have it, you can't trust their product, because it is very straightforward and actually quite inexpensive to get an independent validation of your product.

What we do at Valencell is, firstly, evaluate our prototypes and make absolutely certain that they are gold standard. We strive to make certain our sensor technology is  within machine error of a benchmark device for the use case.

In HRM technology, for example, we strive to be within machine error of an ECG piece of equipment. However, for various reasons, there's no guarantee that our sensor technology is going to work as accurately in a partner product as in our prototypes.

Therefore, when we transition our technology to a partners’ product, we test it again in our labs and confirm that it falls within our specification. If it doesn’t, then we provide support to modify the partner product and/or our sensor module for their particular use case.

Once our partners pass our in-house validation testing protocol (called “PerformTek Certification”), we suggest to our partners that they do not just trust our results, but get an independent test. We publish those tests.

You can see the products that have been launched with our technology and which ones to trust, based on those independent tests. It takes a little bit of work and digging, but you can do it.

If the company doesn't have an HRM study to show you that fits your type of exercise, then you know not to trust it. That's the best rule of thumb until we have a more globalized standard that people can really trust.

That is something that's going to have to happen, not only as the wearable market expands, but as all the different vital signs expand. Look at these new vital signs we can pull up such as RR interval, which can be used to detect atrial fibrillation or blood pressure, optically in ear buds and in other devices. There's going to need to be some overall standard people can trust.

How useful is step data from a health perspective?

I think step data is very useful, in context. There are a few things people want to know and a few things that can be measured and kept track of. People want to know whether they are eating right, exercising enough and whether their health is getting better or worse.

Another big thing people want to know is whether they are in a situation where they can perform their best or whether there is something wrong that might make them not perform so well, in terms of maybe their exercise, work or anything that they do.

Those are the four things people want to know: whether they are eating right, exercising enough, their health status, and their capability to perform well i.e. their aptitude.

Then you can ask how steps relate to these things that people really want to know. The best way they relate to that is by helping them understand if they’re exercising enough.

Now, the downside is that, although knowing whether you are stepping more or less is very helpful and generally good to know, researchers who understand this space are saying that step data is not enough.

They say people really need to know a couple of other things. One is how many calories you've really burned in context of what you're eating. You also need to know you are exercising with the right intensity for your own body.

One big limit with measuring steps alone, which the marketplace is starting to realize and is why we see more heart rate monitors coming into these products, is that just knowing your steps provides only one dimension of information. There’s no context for how an intervention can be used to help you.

You need to know the energy you're expending, which is usually measured in terms of calories burned and what intensity you are exercising at with respect to your maximum heart rate capacity.

If you look at exercise as medicine and as a dosage, the dosage is unique to different individuals. It's unique based on an individual’s own aerobic capacity, which is often called VO2 Max. If you can understand what a person’s aerobic capacity is, you can tell them what intensity of exercise they need, uniquely tailored to their own physical habitus.

It turns out it's important that you exercise within certain intensity bands and that can only be measured by your heart rate. Activity trackers can provide insight into what you're doing, but they can't tell you how your body is responding to what you're doing. This is where it becomes really important to go beyond steps.

I think step data is useful for giving people a general direction, but it's dangerous beyond that because it's hard to say if you're stepping too much for yourself or if you're doing the right kind of steps.

To put it simply, the right kind of steps might mean that you're stepping enough, but not putting enough energy into those steps: that would not help your fitness level; it would only help you lose weight.

What's known is that you don't get the cardiovascular benefit unless you exercise within certain intensity bands. You can lose weight, but you're not really helping your cardiovascular fitness.

Is it possible for next generation wearables to monitor additional metrics such as stress?

The answer is yes, depending on the definition of stress. There are many definitions of stress in the National Institutes of Health, so you have to be very specific.

For example, there's cardiac stress and a very specific definition of that, which has to do with your cardiac output and how your heart is working for that cardiac output.

Then, there's psycho-social stress, which is a psychological phenomenon that can manifest in physical phenomena. For example, your autonomic nervous system will respond in certain ways to psycho-social stress.

An example of psycho-social stress might be that you're really stressed over whether somebody's going to take your job at the office, or you're stressed because your wife is angry with you about something.

Those are examples of psycho-social stress, but another form of stress, which also affects your autonomic nervous system very similarly is physical stress.

For example, if you are working out really hard one day, that's going to affect your autonomic nervous system in a way that looks like psycho-social stress.

The way that we can assess stress now with these products, and Valencell just launched a product that is capable of doing this, is by using something called RR intervals.

An RR interval is the time between R beats in your heart rate. On looking at an ECG waveform, the R peak is that sharp peak you see on the screen, and the time between those separate R peaks has physiological meaning when you look at all of them and process them statistically. That is indicative of how stressed you are.

The problem is, you can't be 100% certain that the stress is because of psycho-social stress or due to physical stress. The models need to get better. What we can do today is tell you that your autonomic nervous system is behaving in a way that indicates you're stressed.

What we can't exactly do is tell you the origin of that stress. More models need to be developed to be more accurate in describing that and more selective and sensitive to that particular type of assessment.

What other metrics do you think next generation wearables could realistically measure?

The workhorse of biometric sensing technology in wearables today is opto-mechanical sensing technology. Basically, you direct light onto the body with a sensor module, and they you analyse the light that scatters back from the body.

After removing mechanical-oriented noise and environmental noise, you can generate a clean optical signal that can be used to accurately measure various biometrics.

Now, there are other sensors that people have been investigating. For example, some companies have put a lot of money into investigating something called bioimpedance, which basically refers to characterizing the electrical currents that run on top of your skin, as opposed to shining light on your body and scattering it.

Valencell investigated this technology as early as 2006, and we saw a great deal of practical near-term and long-term weaknesses with that approach, which is one of the reasons why Valencell decided not to adopt it.

The realistic thing that you can measure with  optical sensor technologies, which are based on an older technology called photoplethysmography (PPG), is anything that can be derived from blood flow information.

That means things like heart rate, respiration rate and blood pressure. It also means changes in your blood analyte level, such as oxygenation or hemoglobin saturation. You can also get a good estimate of cardiac output, which is how much blood your heart is really pumping out.

Because the blood flow is going to have certain properties associated with your hydration level, you can also get an estimate of your blood hydration level. Anything associated with blood flow or anything that modulates with blood flow in any way can, in principle, be pulled out from optical sensor technology.

The technologies aren't accurate enough yet for all of those metrics. Heart rate is the only one that's good enough for a lot of the consumer use cases and now, with more advanced technology, even medical use cases.

The other metrics are getting there and you can expect a much broader array of metrics available from a single device that is using optical sensor technology.

How long will it take to develop next generation wearables and when can we expect to see them on the market?

Since Valencell is a technology provider in this space, we get to see a lot of what the future road map is going to be because we work with so many companies. Unless we’re authorized by partners, we can’t provide specific information, but we can provide sweeping generalities.

For example, we see more and more companies coming to us for medical-oriented wearables, with real projects that make sense.

That did not use to be the case. Until about a year and a half ago, we had companies coming to us with ideas, but they just weren’t viable. Now, we're seeing use cases that make sense and have real money behind them.

If you then apply the timeline map to get these products to marketplace, the soonest some of these medical-oriented wearables can get in the marketplace would be 3 years from now because there's a 1 to 2 year development and validation cycle.

Also, there's an approval process required. You need to pass approvals for the FDA, HIPA, and FCC, for example. A number of different approvals need to be awarded before the product gets launched.

Realistically, if we want to think about true, wearable medical devices (devices that make medical claims) and not traditional wearables in the sense of an arm-band blood pressure cuff, but, for example, wristbands, ear buds, consumer worn wearables, that type of a timeline is going to be a little way out; I’d speculate that it's going to be about 3 years.

How do you think wearables for sleep tracking will progress?

What was interesting and surprising in the study we did, was the response when we asked people when they would like to wear their wearable? I was expecting something like 70% of people would want to wear them 24/7 during the day and when they sleep.

We found it was very clear that half the people did not want a sleep-worn wearable and only wanted to wear it during the day. Of that group where half of them wanted to wear it just during the day, most of them were interested in a day-worn wearable and almost but not quite half of them, wanted one just for exercise only.

A huge chunk of people fell into those categories, but it's not a niche. If you have a market that's 50 billion in revenue, 20% of the marketplace or 30% of the marketplace is huge. It's far from anything that could be called a niche.

What it tells me, is that some people just don't want to sleep while wearing anything. This will be one limiting factor for sleep tracking, but still, half the people do say they want this.

What people want to know about their sleep is typically one of two things: 1)whether they are sleeping well enough and if not, what can they do about it and 2)  how their sleep is affecting their health.

Those are the two big questions that need to be addressed and what we see right now is that with sleep, the activity trackers have been grossly under-performing.

What's odd and really interesting is that there was a company a few years back called Zeo that had a sleep band technology that would be placed across the forehead for measuring EEG, the brainwaves.

It's well known that the gold standard for characterizing your sleep stages is EEG, because we can see your brain waves and we know certain brain waves only exist when you're in certain stages of sleep.

This product was reportedly incredibly accurate and they had many models for it, but it failed to sell well because people didn't want to wear it on their forehead.

In the meantime, there were these activity trackers that enabled you to put a sleeve on your wrist and attach the activity tracker. It could be worn whilst you slept at night to give some indication of your sleep quality, but the accuracy was terrible.

People couldn't tell if they were really sleeping well; they couldn't tell if they were really up and they were getting things that were almost random number generators. But, these were selling much better than the headband that was more accurate because the form factor was just untenable to many folks.

I think wearables for sleep tracking have to get more accurate, but people also need to figure out how to integrate these more seamlessly for sleep. People often don't want to wear anything while they sleep or they may move around and dislodge the wearable from the location where it needs to be in order to sense correctly.

What I'm more bullish about, actually, in terms of sleep monitoring technologies you can have in the bedroom are ones where you go to bed, you lie in the bed, you don't have to change anything, you don't have to wear anything, but the technologies are able to assess your sleep while you're sleeping in that bed.

I think that will be a lot more useful for a lot of people who are trying to understand whether they are sleeping well enough and whether there is anything they can do about it, as well as for people who want to know if their sleep is affecting their health.

If you're wearing a biometric wearable and you get sleep information from some other source, then you can connect the dots between how, for example, your heart rate variability or heart rate recovery or VO2 Max is responding to your sleep quality.

Another issue with sleep that is really important to address here is the intervention side. Even when you do know someone is not sleeping well, how do you help them?

This is a problem because the interventions used to help people could be very dependent on the origin of the sleep problem. The industry is struggling with the intervention side and not just for sleeping, but for all the things we measure. It just turns out that for sleeping, it's more complicated.

I'll give you an example. With our technology that can measure RR interval, if you're doing a two-minute test where we're collecting this data and doing statistics, we can show whether or not there is a strong chance that you have atrial fibrillation.

What can you do with that? You can go to the doctor, you can get a test, maybe wear a Holter monitor for a week, see if we were correct and then address that. The intervention is pretty straightforward.

It's not necessarily easy to execute, but it's straightforward. We can tell you to go the doctor, and the doctor has medically accepted tests they can do to provide a professional diagnosis, confirm your situation, and then provide you with the appropriate therapy. This is a simple “prescreen” functionality that personal health wearables will soon provide to consumers.

But let's say, for example, you find out from one of your wearables that your sleep quality is poor. It could be because you have a disease, it could be because you have a lifestyle change you need to make or it could be because you're depressed.

How would you tell that person the next steps to take? That's one of the biggest issues with sleep tracking right now; the interventions can be very diverse and very dependent on the nature of the sleep problem.

What other advances do you expect for health-monitoring wearables moving forwards?

A big issue that is mentioned a lot, and actually came up in our survey, is that wearables are too much of a hassle to recharge. Finding ways of either making the battery life much longer or making the recharge time much shorter, would be really important for wearables.

This interplay between user experience and battery life is interesting. I identified this problem a few years back, which I called “Death by Discharge”.

It refers to when people get a wearable device and they're really excited by it and use it and then the battery dies. They plug in the charger to charge it up and they live the next few hours of the day without it on.

Then, when they remember to put it on, it's not as interesting as it was before. They don't see anything new.

The wearable may just let you know you’ve walked so many steps one week and so many steps the following week and that you've got to the point where you walk the same number of steps. There's nothing very interesting.

Each time you go to recharge it, you have less of a reason to put it back on and it ends up in the sock drawer. It's not just the battery life and the recharge time; it's combined with the user experience side.

If you have user experiences that make someone eager to wait for it to charge and put it back on, that's one thing. The problem is the user experiences that people have usually take longer than that.

The marketplace doesn’t typically story board the user experience well. One of the things we had to do, and this was another one of the advances that I think has to happen in health monitoring wearables, is you have to understand how different biometrics change over time with an intervention and combine that with the recharge cycle.

For example, let's say that you go to a doctor, find out you have cardiovascular disease and you’re among the large portion of the population with cardiovascular disease where exercise will correct it.

That is not the case for everyone; some people can't correct it with exercise, but by far, most people can. So, the doctor puts you on an exercise regimen. How do you know you're improving? Well, your resting heart rate will start to go down in a few days and you'll start to see that benefit, but at some point it's going to go as low as it's going to go.

You have to look at the other biometrics and how they change and it turns out your cardiac efficiency continues to change over the course of several weeks. The simplest way to explain cardiac efficiency, is that the fitter you are, the less heartbeats it takes to move a footstep, so you can look at it as the ratio of footsteps to heartbeats.

Cardiac efficiency continues to improve over the course of time, but then that flattens out because eventually you've become as efficient as you realistically can be. Then, there's another thing in your health that starts to improve and it improves over a longer timescale.

This is the thing doctor's really want to see improve and it's called your VO2 Max, but that changes over the course of months. If you can measure all these things accurately, you can put this in an interesting story board where there's always something interesting to learn about yourself, without the battery getting in the way of your interest in the product.

Where can readers find more information?

About Dr Steven LeBoeufSteven LeBoeuf

President and cofounder of Valencell, Inc.

As cofounder and president of Valencell, Inc., Dr LeBoeuf has developed ongoing strategic partnerships between Valencell and leaders in industry and academia.

He has raised more than $10M in funding for Valencell and is the inventor/co-inventor of more than 50 granted patents, including dozens of foundational patents in the field of accurate wearable sensors.

Prior to Valencell, LeBoeuf led the optoelectronic biosensor program at GE Global Research, where he managed the development and productization of biosensor systems and developed cutting-edge nanosensor technology.

Before joining GE, LeBoeuf developed optoelectronic solid-state materials and devices while researching at North Carolina State University. Dr. LeBoeuf holds a PhD in electrical engineering from N.C. State and a bachelor’s degree in mathematics and electrical engineering from Louisiana Tech University.

April Cashin-Garbutt

Written by

April Cashin-Garbutt

April graduated with a first-class honours degree in Natural Sciences from Pembroke College, University of Cambridge. During her time as Editor-in-Chief, News-Medical (2012-2017), she kickstarted the content production process and helped to grow the website readership to over 60 million visitors per year. Through interviewing global thought leaders in medicine and life sciences, including Nobel laureates, April developed a passion for neuroscience and now works at the Sainsbury Wellcome Centre for Neural Circuits and Behaviour, located within UCL.


Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Cashin-Garbutt, April. (2018, August 23). Next generation wearables for health monitoring: an interview with Dr Steven LeBoeuf. News-Medical. Retrieved on June 23, 2024 from

  • MLA

    Cashin-Garbutt, April. "Next generation wearables for health monitoring: an interview with Dr Steven LeBoeuf". News-Medical. 23 June 2024. <>.

  • Chicago

    Cashin-Garbutt, April. "Next generation wearables for health monitoring: an interview with Dr Steven LeBoeuf". News-Medical. (accessed June 23, 2024).

  • Harvard

    Cashin-Garbutt, April. 2018. Next generation wearables for health monitoring: an interview with Dr Steven LeBoeuf. News-Medical, viewed 23 June 2024,


The opinions expressed here are the views of the writer and do not necessarily reflect the views and opinions of News Medical.
Post a new comment

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.

You might also like...
Waist-to-height ratio surpasses conventional metrics in predicting cardiovascular disease risk