Heart Rate Monitor Recommendations for 24/7 Continuous Monitoring and Logging

Hello,

Does anyone have any suggestions for heart rate monitors (HRM) that can be worn close to 24/7 for continuous monitoring and logging? Can anyone that is currently doing continuous HR monitoring share there experiences and offer any advice?

Thank you,

Jason

You should look at the Memory Belt by Suunto.

Hi Jason,
I tried an extended test with my Garmin FR60 Heart strap and watch with the following results:

  1. The heart monitor band is not comfortable for long durations. (probably why they sell a soft band version)
  2. The watch itself has limited memory (I doubt it will hold more than a day).
  3. Because the heart monitor band is stiff, it loses contact (no data) at night if you sleep on your side.
  4. I would be concerned about battery life of the heart monitor in continuous use.
  5. The web page graphs of heart rate don’t match the downloaded data (they may be smoothing it).
  6. The watch time is not synced with the computer, so the time stamp on the data can be wrong.
    Good Luck

I’m intrigued by this. Mixed reviews for the Suunto Memory Belt on Amazon. Edison, have you tried it? Does it work well for you?

[quote=“Jason_Miller, post:1, topic:327”]
Hello,

Does anyone have any suggestions for heart rate monitors (HRM) that can be worn close to 24/7 for continuous monitoring and logging? Can anyone that is currently doing continuous HR monitoring share there experiences and offer any advice?

Thank you,

Jason[/quote]

Hi Jason,

It is not cheap, but the BioHarness by Zephyr could fit your need. It can do both continuous monitoring by memory storage or a combination of live and recorded. It only needs a about a 3hr recharge and if you do that just before you go to bed then it should work just fine for you.

Mike

I am surprised no one has mentioned any of the watch-based solutions, which seem to me to be more feasible for all day monitoring.

Hey there,

I’ve been using the BASIS wristband for about a month.
The most important arguments FOR the Basis:

a) it’s unobstrusive: just use it instead of your normal watch and wear it every day. you won’t lose it and you don’t have to do much, you don’t even have to take it off to charge or sync it
b) long battery life: charge/sync for 10-15 minutes a day while sitting at your computer and you’re fine, if you don’t have a computer or USB power it’ll go 4 days easy before it’ll run out
c) well done web-interface. They figured out some really nice ways to graph the data and has some useful metrics
d) a LOT of data for a wristwatch: it has optical, GSR, temperature and acceleration sensors to measure heart rate, perspiration, steps and consumed calories, it abstracts from this data how long (and well) you slept, when you went to bed, when you rose, whether you take a morning / evening lap, how often you awoke during the night and how many times you stand up and walk few steps while sitting on your desk.
e) HR measurements are relatively accurate if you don’t move: http://psychologygeek.tumblr.com/post/52500859364/comparing-the-basis-wrist-band-to-a-research-grade

And some arguments AGAINST the BASIS:

a) ability to measure heart rate is heavily affected by movement artefacts. This is something that can’t be overcome easily using optical methods. (Buy a pulsesensor.com sensor and look at the raw data and you’ll know what I mean.) However, the sensor weeds out most, but not all false values. But be aware that this means that you can’t use it well for athletic activities, thus it really might not be well suited for your purpose.
b) you don’t have any access to your data except via their interface. This is kind of a hit in the face for most of us. https://github.com/dunn5/MetricsCSV check here for a solution
c) gamification vs function: some of the above mentioned features, like info about when you went to bed or when you got up, are only available after you’ve completed ‘habits’. You have to set yourself goals (walk 8000 steps a day, burn 2000 calories). If you’ve kept the goals you set for a while you get points. With the points, you can buy new metrics. After more then a month, I still can’t count morning laps. Arrrrrr! Nice idea, it might be actually doing something for some people. Others might set their goals lower because they want the metrics. The worst part: The metrics only assess data from the point where you bought them, and not retrospectively. So any morning laps I do now will be lost forever…
d) awful display. just bad. really bad contrast, really hard to read during the day. uugh.
e) it’s a bit pricey (199 USD)
f) another limitation of optical methods is that HRV is really hard to assess reliably. this option lacks and is probably not coming soon.

So if you want continuous HR measurement without hassle, the BASIS is my first choice AT THE MOMENT.
If on the other hand, you want a device the measure while doing athletic activities, go for an ECG sensor strap (look for the ability to extract HRV data if you want more from your device).
To be honest, I love my BASIS it does many things right. But you might want to wait for the next generation to come along.

There’s another gem: the http://www.op-innovations.com biosensor. It’s a bit in-development still, but it’s just an excellent product that records 11hrs of raw data, (1ch EEG/ECG, temp & 3D acceleration per sensor, you can use several at once) is open-source, has some software to go with it and comes at a price you can’t beat. (And there are still plenty arguments for it I haven’t even mentioned.) If you’re comfortable with handling raw data, give it a try!!!

1 Like

Thanks for that review of the Basis. Can anyone contrast with the Mio Alpha?

The Mio Alpha provides reliable HR, even during movement. The device is meant to be used for sports, so who developed the technology (i.e. Philips) invested a lot of effort on motion artifact reduction. This means that:

  • the device is not meant to be used for 24/7 monitoring, and won’t last that long. If monitoring 10-15 hours is enough for you, then there is no problem.
  • you can’t go beyond HR

The last point is probably the most important. A wrist worn device able to provide accurate RR intervals could be useful for several applications, not only sports (e.g. meditation, stress monitoring), simply by extracting standard time and frequency domain features used for decades in research. However, this is not possible at the moment with the mio aplha.

All of these new sensors implement the bluetooth low energy (or bluetooth smart or whatever you call it) heart rate profile, which is great since you can easily connect the device to any existing app or very quickly write your own. According to the profile, the device can send a few things, one being the HR, that they all provide, and another one being RR intervals. The problem is that even if they do provide RR intervals, they are not accurate, and are also affected by everything that is going on in the watch (I guess some sort of averaging to keep the reading stable). I did a quick test writing some code for iOS to get all the info out of the watch and compared it to a Polar H7 and a research ECG device(http://www.imec.be/ScientificReport/SR2010/2010/1159154.html).

Bottom line: the MIO alpha is a great device for runners/bikers, it can be used for other sorts of tracking, but limited to mean HR (at the moment). The way to go for HRV analysis is probably still limited to chest belts (Polar, zephyr or other prototypes if you have access to them), but as we all know they are unbearable after a few hours.

Thanks for that review, Marco!

A technical question: Would you say that calculating (valid) measures of heart rate variablity from an optical measurement of a wrist-worn devide is generally unlinkely to yield usable results? My suspicion is that the time a pressure wave caused by a heartbeat takes to traverse the body (pulse transit time) varies too much between measurements and individuals to arrive at a single comparable measure of HRV. (For example: movements or changes in posture could cause the time between peaks in the raw data to change, even though heart rate variability itself is constant.)
But this influence might be minor, a quick google scholar search didn’t turn out anything of much value.
Looking at the data in the frequency domain might be a valuable approach. Did you ever play around with some raw data from optical sensors? If you’re up for writing some code: it would be trivial to just throw together a measurement device made from an arduino and a pulsesensor.com sensor, and quite cheap to have a couple of them made. But as I said, I doubt you get much valuable HRV information from an optical sensor. What do you think?

Robert:
And I know I say this in every second post I write but it really applies: You can give op-innovations.com sensor a try. It’s very small, you can attach it to your chest without a belt using adhesive band-aids that come with it. You can record 11 hours continous raw data with it (ECG & accelerometer/step counter on the chest): if you have two or three sensors, you just need to switch them once or twice a day for a fully charged device. They’re quite cheap <50 USD and data quality ofECG recordings is excellent!
The drawback of this device is that it’s not a fully developed product yet, but more of an open source effort. If you want to get acquainted with raw data and get some baseline data before the product we all wish existed is on the market, this is the best choice.

Best,
Martin

1 Like

The “adhesive” solution is getting some increasing attention - we had Steve Zadig from Vital Connect at the recent Bay Area Show&Tell, they are doing adhesive sensors that include heart rate and hrv, it was really interesting to see them. Their web site doesn’t have much info, but they are a sponsor of the fall conference so I think we can get them to bring lots of prototype sensors for people to try out.

Adhesive solutions are definitely a thing to watch. Also looking forward to the MyoLink by Somaxis, an adhesive EMG/ECG/EEG platform! (Did you pledge the kickstarter campaign, btw?)
If you’re OK with some continuous cost & logistics associated with buying new electrodes, it has big advantages compared to a chest- or wrist worn device.

The downside of adhesive solutions are skin-irritations that will be in issue if one is set on the 24/7 thing. Changing the location every time you attach new electrodes might help there, though.

Hi Martin,
Looking at the raw data, it seems feasible. Obviously, you can forget about having a nice QRS complex, but peaks between beats are there. Also the variability due to posture/movement could create issues, especially for some HRV features, but as always, it depends a lot on the application you have in mind. Typically HRV features are used at rest or during sedentary activities (e.g. to detect stress, or during meditation), so you would not be affected by such issues (fixed posture, limited motion). I guess it’s just a matter of time, before we see something more reliable.

Regarding the patch/adhesives, Delta (based in Denmark) also makes nice stuff(http://www.madebydelta.com/imported/images/DELTA_Web/documents/ME/ePatch_ECG_EMG.pdf), we (imec) collaborated with them for the development of some research prototypes in the past http://www2.imec.be/be_en/press/imec-news/ecgpatch.html

I’ve also use off-the-shelf electrodes for about a year 24/7, using the ECG Necklace I linked before, and they are all but comfortable (of course that’s not the intended use, but it was the cheapest option).

I think the emwave uses optical sensors for their measurement (finger or earlobe). I suspect the ear is a better place to measure given it is closer to the heart and probably has better pulse fidelity. It may also have better blood flow given it’s proximity to the brain. Also I would expect movement of the arm and finger would modulate the flow. On the heart math web site they gave both frequency and time domain displays of the HRV measurements. They looked similar to the electrical measurements. I think this is still predicated on people sitting down and breathing slowly to reduce the other contributors.


Update:
Attached is a screen shot from an Inner Balance ear HRV sensor pulse display. One is stationary, the other is more distorted from movement.
My initial reaction is that this optical approach is not suitable for 24 hr HRV monitoring. You may be able to obtain HR information, but the motion artifacts can be greater than the blood pulse.

Enjoy.

TrueSense Kit offers attractive, low cost and easy to wear solution for continuous ECG/HR monitoring, as pointed out by Martin.
It can be used with very low cost reusable gelpads (included) for reliable skin contact.
You can also attach the electrodes using Breathable, Waterproof film for long-duration comfortable use (see http://op-innovations.com/en/thinfilm)
http://youtu.be/uLWV-ze2MTY
the accessories are now available at very low cost(http://www.taipeitrading.com/op003a.html)

Hi,
Update on my heart rate (HR) and Heart Rate Variability (HRV) monitoring during sleep.

Hardware:
Garmin Premium Heart Rate Monitor (Soft Strap) ANT+ electronics only
Polar WearLink+ Replacement Strap (mates well with Garmin electronics)
Spectra 360 Electrode Gel, Parker Labs
Garmin USB ANT receiver (for HRV and data download from FR60 watch)
Garmin FR60 watch (for HR capture)
Mac laptop computer.
Software:
MAC OS 10.7
TCX Converter App (to translate Garmin downloaded TCX file to CSV format)
MS Excel to view HR data
HRV tracker.app (for HRV data)
WinOnX (to run Kubios)
Kubios v2.0 to view HRV results (problems installing matlab libraries for version 2.1)

Results:
The Polar strap with the Garmin transmitter works well.
I have it on light tension so it is comfortable. I do notice strap movement depending on sleep position so I may increase the tension.
The electrode gel made a big difference in improving performance.
I think this approach will work for short duration studies (days), but is still too uncomfortable for long term wear.
Able to get most of a full night data on heart rate (e.g. missed 33 samples out of over 5000)
HR samples are at 5 sec intervals so there may be more dropped data points I don’t see. This will show up in HRV numbers.
HRV Tracker.app only collects up to 20 min of data before having problems (13 missed samples in 1000). App is Beta version. Source code from google code does not run correctly (for me), so I am still investigating this problem. I suspect the app is a later version than the published source code.

I am playing around with overlaying HR and ZEO Sleep results to see what events correlate.

Enjoy

Thanks for the update OP.

So as I understand it this set up with Garmin/ HRV Tracker allows you to record data live only and that is why there is some loss at the HRV Tracker end point?

Put another way there is no memory storage in Garmin, so that you have to be close to a receiver when recording to track the data.

The only product I’ve seen so far that gets round this problem is the Bioharness mentioned above, which has the higher price point. Did you find anything else while looking for a solution? I’m prepared to pay some money e.g. Bioharness rate, if it works 24/7.

Hi skyline,

Yes the Garmin approach has no data storage in the heart rate monitor strap. In my case I had both the watch and the ANT USB stick / laptop next to my bed. Both were running concurrently recording live data. Both the watch and the HRV Tracker app were running the next morning (although I don't recall if the app was displaying real time RR measurements when I stopped it. The watch was displaying heart rate when I stopped it). I believe I stayed within signal range the whole time. Given the above, I am convinced the software has a glitch beyond 20 min of data storage, I just haven't gotten the source code I downloaded working to determine where the problem lies.
I am taking an incremental approach to this investigation. I have noticed heart rate spikes during sleep and I would like to know more. If I see value in the 24 hour Heart Rate Variability data then my next step may be an iOS app to log the data from an ANT+ adapter. This would provide portability for the data collection.

Thank you

Thanks for the update.

I know there are iOS apps like SweetBeat and the ithlete app. I’m investigating and will report back when i find something.

the ECG/HRV recording overnight has lots interesting but little discussed features: spikes, oscillations, very slow periods, etc., over the various sleep stages, and very worth exploring.
http://www.bioshare.info/en/ecg
[attachment=101]