Frequently Asked Questions
BSX Insight is equipped with both ANT+ and BLE communication stacks. This allows it to communicate with any ANT or low energy Bluetooth enabled device, including sports watches and eventually even phones (for example new Samsung Galaxy and Note phones have built in ANT antennas).
Most watches on the market today use ANT+ (it has become the defacto standard for wireless transmission of sports data) including Garmin, Timex, Suunto, etc. I'd recommend checking with your watch manufacturer to confirm they use ANT and if they do, it will pair just like other peripheral devices you’re currently using (e.g. HRM).
Watch integration is our first priority and right now we are working with Garmin to fully integrate BSX Insight into both their watches as well as their software platform. As we grow, we will branch out to include other watch platforms too. This tight integration will provide a more seamless and automatic experience, allowing athletes to both download data from their watch after a workout, as well as push new workouts to the watch which include personalized training zones.Last updated:
Insight will connect via ANT+ with your watch just like a HR meter, power meter, pedometer, or other peripheral device. At the moment BSX Insight is a wearable lactate threshold sensor and our algorithms are built to identify target zones and then facilitate an athlete’s adherence to that zone. I.e. telling them when they need to speed up or slow down to obtain the best training outcome. We did this for two reasons: 1) the ANT+ communication standard does not contain LT profiles allowing us to prompt with exact LT information in real time. This is something we are working on with thisisant.com currently and hope to have added soon. 2) speed up/slow down is much simpler and more actionable than mmol concentrations! As we face a sizeable need to educate the endurance community on lactate dynamics this is a good first step. Long story short... in either a future feature update or second generation (if hardware requires it) we hope to include real-time mmol values as well.Last updated:
You DO NOT have to be a premium user (coached or uncoached) on trainBSX to analyze the data received from BSX Insight. In fact, you don’t have to use trainBSX at all. Our goal is to develop a truly agnostic piece of wearable tech that in time will sync with any other training platform—Training Peaks, Addaero, etc. Of course this will depend on the cooperation of the developers of those platforms. We are working on developing the relationships with those parties to make that possible very soon. At the same time, we wanted a platform that would seamlessly integrate with Insight from the very beginning.
If you think of BSX Insight as a wearable peripheral (similar to power meter, heart rate meter, pedometer, etc) this makes sense. Each of these devices transmit their data wirelessly (e.g. ANT+, BLE, etc) to a sports watch that packages the information into a workout and transmits that (e.g. TCX file) to a piece of software for storage and post workout analysis. We are eagerly working with thisiant.com and the developers of third party platforms to ensure lactate data can be both transmitted and stored no matter what watch or software you choose to use.
We naturally believe that trainBSX will be a pedestal product in this space. If you don’t want to use it, you can still get full value out of BSX Insight. If you do choose to use trainBSX for your training you don’t have to be a premium user. You will still have full ability to measure, record, and trend both your LT and other endurance metrics over time. If you choose to become a premium user ($3.99/mo) it costs less than fifty dollars per year. Additionally, the premium version will enable you to create your own workouts and push planned workouts to your watch. This will essentially allow you to take your coach with you whenever/wherever you train. The watch will tell you when to speed up/slow down based on your LT training zones and the schedule of your coach.Last updated:
Each device (Runner and Multisport) is only capable of maintaining training data and profiles on a single athlete. Sharing between multiple athletes will obscure results and result in inappropriate training zone calculations. For those of you looking for this shared functionality though, you’re not alone and we would recommend the Team Edition. It is perfect for this scenario. The Team Edition is capable of maintaining up to 10 lactate threshold profiles at once. You can learn more about the different models of BSX Insight in our other FAQs or the campaign page.Last updated:
What are the differences between the running and multi-sport editions? Why is cycling included in the multi-sport edition?
BSX Insight comes in 3 versions: Runners Edition, Multisport Edition, and Team Edition.
The *Runner’s Edition* is capable of measuring and maintaining a running lactate profile (including both lactate threshold and variable training zones) for a single user, along with traditional endurance metrics like heart rate, cadence, pace, and calories-all in a single device.
Initially we had only a single device type, but found that lactate profiles were significantly different between different sports in the same athlete. This made it impossible to have a single LT curve accurately depict optimal training paces for both running and cycling. In response, we have had to change some of the components and algorithms to be able to identify and maintain two separate lactate profiles. This is the reason for the two separate SKUs.
The *Multisport Edition* was developed at the behest of our cyclist and multisport friends who train on a bike and, whether in event preparation or cross training, also run. This edition allows a user to measure and simultaneously maintain multiple lactate profiles (including both lactate threshold and variable training zones) for each of their sports in addition to the other traditional endurance metrics found in the Runner’s Edition. For now, it will also include bike cadence. Depending on the level of support from the Kickstarter Community we also hope to include left/right balance at product launch or soon thereafter.
We’re excited to be including this edition in our initial market release to Kickstarter supporters this Fall! If it weren’t for the support of our local endurance community it would not have been possible to tackle both for joint release so soon.
The *Team Edition* was designed at the request of our MUCH APPRECIATED partner coaches who want to be able to use Insight for multiple users at once. It is capable of holding up to ten lactate profiles (including both lactate threshold and variable training zones) simultaneously on a single device. It is perfect for coaches and teams who understand the value of data or users who want to share Insight for only periodic lactate threshold measurements.Last updated:
The compression sleeve is primarily the apparatus used to hold BSX Insight in place so that it can get the most accurate LT reading during your assessment run and subsequent training and racing activity.
We’re well aware that some people don’t love compression sleeves for any number of reasons and we’re working with our fabric partners on multiple solutions, including smaller, less invasive sleeves.
Many athletes have asked, “How should I wear the sleeve? Does it matter if I switch legs or always keep it on the same?”
We'd recommend either of the following: 1) wearing all the time on one leg really doesn't matter, you won't notice a difference, 2) rotate legs if you'd like to, or 3) in response to the many questions we've also developed a matching pair of sleeves so you always get the support of both legs and can rotate the device if you'd like.Last updated:
Your initial lactate threshold and zone targets are set by the LT assessment run. Real-time target adherence is then monitored by continuous wearing of Insight and your watch. If you wanted, you could then use those zones and/or pace and plug them into your current training systems.
Here's two different use cases:
1. Daily training (Recommended): BSX Insight untangles you from the multitude of one-off wearable devices out there (including constrictive heart rate monitors) and measures all the endurance metrics you care about. We recommend wearing it continuously as part of your training regimen.
2. Monthly LT training zone updates: Alternatively you can use the device once to determine your lactate threshold and LT training paces and then plug the zones/pace into your smart watch or power meter to monitor adherence. In this instance, you'd still want to periodically (every 4-6 weeks) use BSX Insight to determine how your fitness has improved based on your improved training -- and to alter your zones/systems accordingly.Last updated:
BSX Insight is the product of many years of research to develop a noninvasive technique for measuring lactate threshold. Our work actually started in the medical space as we worked with local university and hospital research centers to develop translational technologies. Our efforts build on over a decade of scientific effort that has demonstrated the viability of using optical techniques (specifically near-infrared spectroscopy- aka NIRS) to noninvasively monitor athletic performance. NIRS uses light to measure and record activity inside living tissue in a safe manner. A light source shines a signal into the target which then passes through as it returns back to a detector. Depending on the physical and chemical properties of that tissue, the signal is distorted along its path. These distortions (also known as absorption spectra) are what contain the information we are able to process afterwards. For example, oxygenated and deoxygenated hemoglobin have very predictable and very distinct absorption spectra from each other (you mentioned Moxy and Portamonn which do just this). This allows us to measure their relative and absolute ratios as well as the total amount of blood flow through the target region. BSX has perfected both the front end electronics as well as the post processing intelligence to do this with much higher quality and exponentially lower cost than other comparables. These oxygenation parameters are just some of our predictors, along with others simultaneously collected by Insight that feed into the algorithms for surrogately measuring lactate threshold curves.
The concept of measuring surrogate markers to correlate with biologic events/conditions is common practice. In medicine we use surrogate markers all the time to measure things we otherwise couldn’t measure and /or quantify. When we want to know how the kidney is functioning we measure creatinine instead; when we want to assess liver health and/or toxicity we look at the transaminases. By measuring surrogates of lactate we are simply applying the same rigor a physician would use to help him/her understand the body system of their primary interest in a manner that overcomes many of the traditional limitations to blood lactate testing.
Using NIRS to measure this is a very standard practice which has been well documented in the scientific community for many years. As exercise physiologists though, we were further interested in understanding how changes to these oxygenation and other predictor variables occurred in real time during exercise in peripheral muscle tissue, and how (if at all) these changes could be used to correlate with actionable training decisions that would benefit athletes of all levels.
We knew that lactate threshold was the gold standard for quantifying endurance performance and the best way for an athlete to personal their training intensities for best outcomes. The problem with LT was it was a blood test and very expensive to complete. We wondered if we could somehow correlate the blood measurements we were noninvasively recording with the blood measurements of a lactate threshold test. In short, we hypothesized that there must be a predictable correlation between local blood oxygenation concentrations and lactate levels. The rationale went as such: muscles extract increasing quantities of oxygen during increasing workloads. This leads to recordable desaturation event. At the same time, increased workloads begin to stress cellular aerobic machinery beyond their capacity to produce sufficient energy leading to a state of acidosis and increasing lactate production. We believed there might be a predictable correlation between the downward flexion of blood oxygenation and upward flexion of lactate levels. Our hypothesized relationship looked something like this:
[Insert Graph Here]
In this graph the blue line represent oxygenated hemoglobin and the red line represents blood lactate over the course of an incremental exercise protocol. In the spring of 2011 we then set out to test our hypothesis by running a pilot study of 35 athletes through standard LT protocols while collecting NIRS measurements from our wearable device. As predicted, successive samples at the beginning of the test did not show much variability, and revealed a relative plateau. But as intensities continued to increase, there was a clear downward trend in oxygenation data that accelerated as the athlete increased their workload. When overlayed with the blood lactate samples taken there appeared to be a visual symmetry.
Initial visual inspection indicated a strong relationship between features measured by our wearable device and the LT sampling protocol. We then used statistical modeling to correlate the two data sets which revealed a 90-93% correlation between the two across our pilot subjects.
It appeared from our initial research that we had discovered an unprecedented opportunity for measuring the metabolic condition of exercising muscle in real-time and thereby providing both athletes and coaches with powerful predictive training information. This initial proof-of-concept research was completed in 2011 and since then we have continued testing hundreds of athletes (including runners, cyclists, and multisport athletes). Raw data samples look like this:
Our ongoing research and development efforts have successfully, overcome many of the major hurdles faced by other research groups to make a NIRS based approach feasible for use in a consumer electronic device. Based on further statistical analysis we have developed mathematical models to describe this relationship between BLa and oxygenation for automatic interpretation and recommendation algorithms while the athlete is training (or racing). For the first time ever they can know what is essentially their athletic horsepower and know if they need to speed up or slow down for optimal results.
Additionally, we have developed novel algorithms that use proprietary machine learning techniques leveraging hundreds of individual subject tests (n=240) to deliver a robust predictor (True Positive Rate > 0.95) of lactate threshold. What we mean by this, is that across all athletes in our sample population we were able to identify the lactate threshold event using Insight with 5% or less deviation from what independent professional assessment determined it to be using traditional blood sampling methods. Those errors measurements essentially represent the percent of deviation from what the blood tests determined an LT to be, and what Insight determined them to be. As mentioned previously, current efforts are investigating how much of this error is due to true inaccuracies in Insight's data collection and/or algorithms, vs how much is due to the increased resolution that Insight has over traditional measurements and therefore represents an overall improvement in accuracy.
This approach is far superior to the basic regression techniques that have been published to date, because it eliminates the need for the identification of a 'break-point', which is especially susceptible to bias. In addition, our machine learning techniques are data driven, which allow the system to improve its predictive power over time, as more data is collected with each new athlete that trains with BSX Insight. This is why we mentioned increasing our sample size so significantly. The more data that goes into the system. The better it gets!Last updated:
All of our testing has been to correlate Insight's noninvasive measurements with that of traditional blood lactate testing. At this point, our algorithms are slightly more than 95% accurate at identifying LT. With the increased sampling size (initial sample size was 240 athletes, but we anticipate increasing the training set to n=600-750 by the time we ship the device) and introduction of more sophisticated techniques, our ultimate goal is to increase this to 97%.
To clarify the 95%, what we mean is that across all athletes in our sample population we were able to identify the lactate threshold event using Insight with 5% or less deviation from what independent professional assessment determined it to be using traditional blood sampling methods. Those error measurements essentially represent the percent deviation from what the blood tests determined an LT to be, and what Insight determined them to be.
Current efforts are investigating how much of this error is due to true inaccuracies in Insight's data collection and/or algorithms vs. how much is due to the increased resolution that Insight has over traditional measurements and therefore represents an overall improvement in accuracy. In other words, traditional blood sampling is taken every 2-3 minutes. Insight records many times per second. This gives us much greater resolution and theoretically will allow Insight to be more accurate than the old finger-stick method. We believe any discrepancy between the blood and needless method can eventually be explained by the limitations of the old method. These are things we're still investigating.Last updated:
We've been overwhelmed by the support of the endurance community and the many offers to help test Insight. At some point (relatively soon) in this journey we will need additional beta testers in the field; however, at this point we’re doing most of the testing in our own lab so that we can carefully monitor the correlation between our device and the traditional finger-prick method of measuring lactate.Last updated:
Don't see the answer to your question? Ask the project creator directly.Ask a question