Thursday, January 17, 2019

Moov HR Sweat - part 2

In the previous post the Moov HR Sweat preformed extremely well during intense interval exercise.  Since consistent data is important for session to session comparisons, I wanted to explore some more tracings done with the Moov as well as the Hexoskin.  In addition, we will see some raw signal information from the Moov during higher exercise intensity.  I want to emphasize that the intervals below represent a torture test for an optical sensor.  Body motion is severe and road vibration is higher as well. 

Raw data screenshots of the Moov sensor:
This was done on an indoor trainer, with the top pane showing the signal from easy pedaling at 130 watts then followed by 2 different captures at 340 watts.

Although the higher intensity signal is not as orderly as the low power one(top figure), the overall morphology is good enough for an accurate reading.


7 minute interval at 285 watts:
This interval began with 60 sec of a fast start approach at 400 watts.  The Moov (red) tracks very closely with the Hexoskin (green) both early, middle, late and in recovery.  There are some discrepancies at about 2 minutes into the bout of about 5 BPM.

From a practical use case, my purpose in doing a fast start is to reach near VO2 peak quickly.  Part of that assessment is getting my heart rate to about 90-95% max at one minute.  If the heart rate monitor can't track this closely, then the advantage of doing a fast start strategy could be diminished.  One of my criteria for heart rate monitoring, is accuracy at the 30-60 second range of the fast start session to best optimize this pacing approach.

Hexoskin status:
A look at the raw signal in the Hexoskin shows a less than perfect tracing, but good enough for accurate heart rate.



Wingate 60 with extended observation:
The Moov and Hexoskin are still very close, but there is some discontinuity early on:

So there is a 5-6 second gap where the Moov does not seem to discern the rapid change in heart rate.  This also corresponds to the highest external motion of the body.  
However, after the brief mismatch, the correlation is excellent including the maximum peak HR and recovery dynamics.
As mentioned previously, the Wingate 60 should provide an index of VO2 peak, if the maximum heart rate is almost or completely reached.  Failure to track this precisely will therefore cloud whether the VO2 peak was reached or not (as well as information on over reaching).  Therefore some loss of data early on is certainly not a deal breaker for the Moov as long as accuracy returns by 30 seconds or so.

The raw Hexoskin data here is picture perfect:
This was later in the riding session and presumably I was sweating more, creating a better electrical contact for the Hexoskin shirt sensors.


Final thoughts:
  • When it comes to determining accurate heart rate, the Hexoskin shirt is as good as one could hope for (when the electrical contact is proper).  Fortunately, there is a one lead EKG tracing available for inspection after the ride (or in real time using the smartphone app).  Users can be assured of a precise heart rate by reviewing the raw data which is somewhat unique.
  • The Moov HR Sweat appears to be the most accurate optical based unit for interval exercise that I have personally used.  Although not always "EKG accurate", it is pretty close.  Many optical devices are fine for resting and moderate intensity exercise.  However when it comes to road riding, high intensity exertion and severe body motion, they generally do not track properly.  Forehead based measurements take advantage of the preserved blood flow and reduced motion of the face compared to the arms.

No comments:

Post a Comment