I just got my new charge 5.
I can't find a continuous graph of my heart rate over time in either the app or the web site.
Does anybody know how to get to this or perhaps someone knows about a 3rd party app?
Thanks
I am interested in developing an app to chart the heart rate. It would use the files downloaded from Data Export. Right now I need information from Fitbit about the meaning of the "waveform_samples" data in the ".\Heart\Afib ECG Readings.csv" file that is included in the download archive.
Here is a sample Heartbeat chart that my program generates:
Send me an email to pierre-at-kyber.ca for a copy of the program.
I use SpamArrest, just reply to the first challenge email.
I'm talking about a desktop app that can show BPM (pulse) over any period of time, in one chart, that can be saved as an image or PDF.
I can combine a heartbeat chart with exercise and sleep graph. More than you could ever accomplish with a mere app on a phone.
A desktop app that can extract, process, re-format and save to a spreadsheet for further analysis.
A desktop app that can combine Fitbit data with your dieting app (vg. MyNetDiary), and CPAP data (Philips, Resmed, Fisher & Paykel, etc.)
I'm curious, are there overnight spO2 data in this download? If so, how is it presented in the spreadsheet?
Here is what I can pull out of the downloaded archive:
| timestamp | average_value | lower_bound | upper_bound |
| 2021-11-17T00:00:00Z | 94.4 | 91.8 | 95.9 |
| 2021-11-18T00:00:00Z | 94.9 | 93.4 | 96.1 |
| 2021-11-19T00:00:00Z | 94.9 | 92.7 | 96.1 |
| 2021-11-20T00:00:00Z | 95.4 | 94.1 | 96.9 |
| 2021-11-21T00:00:00Z | 95.7 | 94.6 | 97.7 |
| 2021-11-22T00:00:00Z | 94.6 | 93.1 | 95.7 |
| 2021-11-23T00:00:00Z | 95.4 | 94.2 | 96.7 |
| 2021-11-24T00:00:00Z | 95.3 | 92.7 | 97.3 |
Values per row:
- timestamp: The time when the sleep period has ended
- average_value: The average SpO2 value computed during sleep
- lower_bound: Lower bound of SpO2 values
- upper_bound: Upper bound of SpO2 values
Is this of interest?
I also see a file called "estimated_oxygen_variation.csv", one for each day:
| timestamp | Infrared to Red Signal Ratio |
| 10/04/21 00:00:38 | 0 |
| 10/04/21 00:01:38 | -4 |
| 10/04/21 00:02:38 | 2 |
| 10/04/21 00:03:38 | 19 |
| 10/04/21 00:04:38 | 21 |
| 10/04/21 00:05:38 | -4 |
| 10/04/21 00:06:38 | -24 |
| 10/04/21 00:07:38 | -8 |
| 10/04/21 00:08:40 | -13 |
| 10/04/21 00:09:40 | -4 |
I see now how they are computing these, kinda, sorta. The second table that you give, the hourly one is interesting too. This is obviously taken from a much larger data set that go into making the weird sp02 sleep graph. It's been a long day and I'm too tired to do any more figuring!
OK, tell me how you would like to see this data formatted. I can combine this with other data, and display on a graph.
What is the significance of spO2?
No need to do that for me. I don't know what I want anyway! (as I said, long day which began with almost running into a skunk when I was out for a walk at dawn. Believe it or not a crow cawed repeatedly in a tree above me and when it saw I'd seen the skunk and moved away from it the crow flew away! Police crow!).
spO2 is the percentage of oxygen saturation in the blood. Large variations over time can be an indication of sleep apnea which is why some people had a cow when they saw this new chart presentation which shows large spikes while sleeping. When I saw mine I simply decided to discount the chart completely. Oh well.
When you go to the doc and they put that thing on your finger, which is called a pulse oximeter, it measures your heart rate and spO2 at that moment. People who use oxygen because of COPD, etc. use pulse oximeters to check themselves, especially if they start to feel out of breath.
It might be interesting, but I would be curious about how you could detect anomalies in the data. Earlier today my Charge 5 was showing my HR at exactly double my actual heart rate. As it happens, I was with my brother-in-law who's a surgeon, and we did a few basic verifications, (manual pulse reading, Beurer pulse oximeter), and the issue was the Fitbit reading. The Charge 5 also seems to pick up some extranous signals from unrelated activities such as driving. You could smooth out the data, but you could lose some important information such as a true SVT condition. It would be interesting if you could try and correlate the bpm with other data stored in the device, and to indicate zones where the reading may have been "corrupted".
Best AnswerYes, I have also observed improbable BPM values on my Charge 5. Ex: 150+ while walking on a hike. I'd be dead at that rate!
OK, I will add spO2 to the chart. Right now I'm adding Activity (calories burned, calories Active), and Sleep (Minutes REM, Light & Deep sleep).
What is "estimated_oxygen_variation.csv"? Does it have anything to do with spO2?
Thank you for your help and suggestions.
Best AnswerEOV is the first derivative of spO2.
Best AnswerI can generate spO2 graphs. See attached.
Send an email to pierre-at-kyber.ca and I can send you a copy of the program.