Accuracy of the o2ring compared to checkme

[Zackio]

I own the o2 ring, but my hospital gave me the checkme for one night. So I had the opportunity to do a side by side comparison. I understand that the accuracy of the checkme will be higher as it reads every 1 second where as the o2 ring is every 4 seconds.

The checkme gave me a ODI of 2.4 for that night but the o2ring only have me 0.3. Checkme gave 21 dips whilst the o2 ring gave me 2.

Again, I get that it's a 4 second window and if a drop happened to fall within a four second window then it could completely miss it, fair enough. But to only get 2 out of 21 seems a bit much.

When I dig into the data and do a comparison of the tWo readings, I can see places where it could have picked up a drop where there is no way that it could just have missed it if it was taking every 4 seconds.

I have also compared it to a standard finger oximeter and I noticed the same thing. My oximeter dips lower than the o2ring does. It's almost like it smooths it over like the fit bit does with HR readings. 

Has anyone else noticed the same thing? I am considering returning it and switching it out for the checkme, but it could be that mine is faulty.

[wwcanoer]

Did you discover any more about the difference between the data from CheckMe vs the O2Ring?  Did you find any problem with your O2Ring?

Both sample every second.  Given the larger battery on the CheckMe, I wonder if it has a faster processor and does any more complex analysis vs the O2Ring or are they both the same except for the data recording every 2 vs 4 seconds.

I assume that the alarm points are decided by the device and output as times.  i.e. It's not the mobile or PC software that's deciding alarm points, so 2 vs 4 seconds should not have an impact.

I certainly prefer the physical format of the O2Ring but if the CheckMe provides significantly better data then I would choose it.  So far, this is the only review that I have found with real information on the comparison but it's only one anecdote.

[Crimson Nape]

wwcanoer - The 2-second data collection is a duplicate of the previous value. Ultimately, they both are 4-second reporting. Both have a maximum recording time per file of 10 hours. The O2Ring will start a new file after the maximum time is reached. According to another member and Checkme user, the Checkme will stop and not start a new session.

- Red

[Jeff8356]

I noticed with the SleepU that after the 10 hour limit it will start another recording, but there is a gap of 7-8 minutes before the new recording starts.  Not particularly useful.

[Crimson Nape]

The O2Ring and the SleepU share that time gap in between files. The nice thing about OSCAR from 1.3.0 onward is if you rename the 2nd file to match the ending time of the first file, it will pick up where the first one left off.

[wwcanoer]

wwcanoer  -  The 2-second data collection is a duplicate of the previous value.  Ultimately, they both are 4-second reporting. 

- Red

I doubt that the CheckMe simply reports the same number twice.  It is an FDA cleared medical device.  They should catch such a cheat.

Here's screenshots from one CheckMe O2 Max review that zooms in enough on the app to see that the data points are not simply repeated.  There are cases of a single point, three points and seven points at the same reading. Wellue Checkme™ O2 MAX review by Smartwatch Ticks Feb 7, 2021  youtube v=LPFpzBOwgjA

Both sample once per second, which is presumably used for alarm detection and eliminating outliers (bad readings), but recording is presumably averaged for 2-second or 4-second reporting.

[Crimson Nape]

If you say so, but every raw data file I have analyzed did exactly as I described.  In fact, the earlier versions of the Android and iPhone apps did the same thing for the SleepU and O2Ring data.  The later versions now only record an entry every 4-seconds.

[sawinglogz]

There’s a newer model (“Pro” I think?) that has a faster sample rate.

The older ones only advertised 4s, just some of the files recorded at 2s doubled.

[coolwater]

FYI, I also had the chance to check the two against each other and vindicate what OP saw (with the pro model though). I'll copy the post from another board:

I managed to get my hands on a checkme pro (not the max), a new o2ring and an old o2ring (looks like the same model, just has been used for a while) so I thought I'd test them against each other and share the results with the community. I've only had time to test for only a night or two so far.
I've used the term 'picked up' extra drops here, but obv I don't know that for sure, they could be 'made up'.

First night:

First night, I compared o2ring old (O2O) vs o2ring New (O2N)

The trends were generally the same. Similar/same avg spo2, heart-rate, min spo2, max hr.
O2N picked upmore drops. >3% 12 vs 5 and >4% 4 vs 3.

If you look at the graphs, they're quite similar the majority of the time. The big drops were shown at the same time by both.
There was just a short period of 20min where O2N showed disturbance, drops, and generally sub 96% oxygen which did not show up with O2O, which showed just flat 96%-97%ish spo2. This period probably accounts for most of the extra 7 drops picked up.

Likely the o2ring sensitivity has degraded over time. Or it could be the positioning of the ring got disturbed on O2O for a bit.

Second night:

Similar to night before between the two rings, despite wearing one on thumb, one on forefinger. O2N picked up an extra drop >3% 2 vs 1. Otherwise, similar averages, minimums and amount of time spo2 > 95%. Bit of a discrepancy with max heart-rate of around 5% (O2N showing higher).

Would've thought I was on my way to defeating my apnea through various exercises were it not for adding the checkme pro this night. Couldn't believe my eyes.
Basically, I vindicate what the user on the other board saw.
The o2rings picked up 1-2 drops > 3%. The checkme pro picked up 37! It also picked up 12 drops > 4% vs 1 on both rings. The spo2 average was a full 1-2% lower with checkme and the lowest was 86% vs 91/92% with the rings.
There was a good hour where checkme displayed a lot of disturbance in the graph- the rings showed nothing.

Stark difference between the two. It's anyone's guess which is more accurate, but given the checkme has that extra clearance and is used in hospitals, you'd have to assume it's more accurate and the rings are just not picking up all the drops. They seem good for trends because they were relatively consistent with each other, ie if your drops go from 30 to 20 with the ring over a few months, you're getting better.
However, there is a significant difference in my results between the ring and checkme, so bear that in mind...
If the results change dramatically after a bit more testing, I will update.

Regarding other factors, I'd say there's not a big difference in comfort. The checkme feels better on the thumb, but it comes with the big wrist strap. Connecting and retreiving data, they're the same, both use bluetooth and v easy.

[Crimson Nape]

Since there is a sensing difference between the various digits, did you swap the oximeters from the previous night's recording location?  This will tell you if the variance is caused by the digit or the device.

To support my statement above, here is a link to a NIH study on the subject: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4627972/