HRV4Training
  • Home
  • QuickStart Guide
  • PRO & TEAMS
  • FAQ
  • Privacy & Terms
  • Contact
  • Publications
  • Guides
  • Blog
  • Shop

iOS10: camera accuracy check & other minor changes

9/3/2016

0 Comments

 
Blog post by Marco Altini
Next week Apple will host their usual annual event in which they will unveil their new iPhone and also a release date for iOS10. With each new iOS version come a few challenges for developers, as some of the underlying mechanisms in the operating system might change. 

On our side, the main concern is always the same: will we be able to use the camera-based algorithms and maintain a high quality app which can effectively replace an ECG or chest strap? 

Short answer: Yes, while a few things changed, it all works the same and the camera based algorithms are as reliable under iOS10.

Long answer: check out the data below for a few examples similar to the ones we have shown in the past during our clinical validations. 

The last check we will have to run will be on the new iPhone 7, to make sure that also on new hardware everything works the same. However, for anyone with an iphone 5 or 6, we can already confirm that HRV4Training & the camera based algorithms won't be affected.

Camera vs Polar H7

The setup for these tests is always the same we use when we are not at the lab. We have a custom app that records PPG and Bluetooth data at the same time, and also computes already RR intervals for the camera based algorithms, storing them to file together with the RR intervals sent by the chest strap. You can see a screenshot of the app here.

​The only disclaimer here is that as the two system have to run together, for synchronization reasons, we might mess up the camera data a little. Hence, this is a worst case scenario analysis. The reason is that the bluetooth radio interrupts the camera every time we receive a packet, so every second, and this operation while necessary for synchronization, is not ideal as all the camera does is extremely time-sensitive. We always validated things this way and results turned out very good, so for now this will do.

We will look at three segments, one simply following a 10 breaths/minute breathing rate for 3 minutes, and other two alternating self paced breathing and deep breathing. This way we will be able to cover a decent range of heart rates and rMSSD, as well as clearly see breathing effects on RR modulation (RSA), so that RR intervals shorten when breathing in, and get longer when breathing out. 

Paced breathing, 10 breaths/min

The two time series of RR intervals shown below are very close, with a small artifact across the 1 minute vertical line, which is then handled by the artifact correction (as you can see from the overall rMSSD, which is only 1ms apart, considering that repeated measures can differ by up to 5-20ms, and synchronization here is not perfect, there is no difference between the two methods).
Picture

Alternating self-paced breathing and deep breathing

The two plots below show both a combination of self-paced breathing and deep breathing. We can clearly see the swings in RR intervals due to paced breathing, as the effect of RSA is much stronger. rMSSD is once again only 3 ms apart.
Picture
Second segment, with similar considerations (effect of RSA/deep breathing):
Picture

Minute by minute rMSSD & heart rate

Visually we can already see there is a very strong correlation between RR intervals captured between the two methods in all plots above, however correlation is not necessarily a good metric in this case, as rMSSD is very time sensitive and we need to make sure that RR intervals are not only correlated, but actually matching very closely what we get from our chest strap. 

I computed already rMSSD over the entire segment and printed it in the plot above (see the titles), where we can see they are indeed the same across conditions. However, I've also split each segment above in 1 minute windows, as most users use 1 minute measurements, to compare also minute by minute data. This way we can also see how the algorithms perform on different subsets of the signals, where HRV differs a lot due to breathing effects.

Below you can see both rMSSD and heart rate for minute by minute data, as collected by the camera and Polar H7 on iOS 10:
Picture
Results are very good even on a minute by minute basis, on a decent rMSSD range (45 to 120ms). This is a preliminary analysis, but everything seems very much in line with what we reported in the past.

Other minor changes

Not much else has changed in iOS10 for what concerns the current features in HRV4Training, with the main issues being some extra information we need to provide in order to write to the Health app. Make sure to keep your app up to date!
0 Comments

Your comment will be posted after it is approved.


Leave a Reply.

    Picture
    Picture
    Register to the mailing list
    and try the HRV4Training app!
    Picture
    Picture
    This blog is curated by
    Marco Altini, founder of HRV4Training


    ​Blog Index
    ​
    The Ultimate Guide to HRV
    1: Measurement setup
    2: Interpreting your data
    3: Case studies and practical examples

    How To
    1. Intro to HRV
    ​2. How to use HRV, the basics
    3. HRV guided training
    ​4. HRV and training load
    ​
    5. HRV, strength & power
    6. Overview in HRV4Training Pro​
    7. HRV in team sports
    ​

    HRV Measurements
    Best Practices

    1. Context & Time of the Day
    2. Duration
    ​
    3. Paced breathing
    4. Orthostatic Test
    5. Slides HRV overview
    6. Normal values and historical data
    ​7. HRV features
    ​
    Data Analysis
    1a. Acute Changes in HRV
    (individual level)

    1b. Acute Changes in HRV (population level)
    ​
    1c. Acute Changes in HRV & measurement consistency
    1d. Acute Changes in HRV in endurance and power sports​
    2a. Interpreting HRV Trends
    2​b. HRV Baseline Trends & CV
    3. ​Tags & Correlations​
    4. Ectopic beats & motion artifacts
    5. HRV4Training Insights
    6. HRV4Training & Sports Science
    7. HRV & fitness / training load
    ​8. HRV & performance
    9. VO2max models
    10. Repeated HRV measurements
    11. VO2max and performance
    12. HR, HRV and performance
    13. Training intensity & performance​
    14. Publication: VO2max & running performance
    ​
    15. Estimating running performance
    16. Coefficient of Variation
    17. More on CV and the big picture
    ​​​​​18. Case study marathon training
    19. Case study injury and lifestyle stress
    20. HRV and menstrual cycle
    21. Cardiac decoupling
    22. FTP, lactate threshold, half and full marathon time estimates
    ​23. Training Monotony
    ​
    Camera & Sensors
    1. ECG vs Polar & Mio Alpha
    2a. Camera vs Polar
    2b. Camera vs Polar iOS10
    2c. iPhone 7+ vs Polar
    2d. Comparison of PPG sensors
    3. Camera measurement guidelines
    4. Validation paper
    ​5. Android camera vs Chest strap
    ​6. Scosche Rhythm24
    ​7. Apple Watch
    8. CorSense
    ​
    9. Samsung Galaxy
    ​
    App Features
    ​1. Features and Recovery Points
    2. Daily advice
    3. HRV4Training insights
    4. Sleep tracking
    5. Training load analysis
    ​6a. Integration with Strava
    6b. Integration with TrainingPeaks
    6c. Integration with SportTracks
    6d. Integration with Genetrainer
    ​
    6e. Integration with Apple Health
    ​
    ​6f. Integration with Todays Plan
    7. Acute HRV changes by sport
    8. Remote tags in HRV4T Coach
    9. VO2max Estimation
    ​
    10. Acute stressors analysis
    11. Training Polarization
    ​
    12. Lactate Threshold Estimation
    13. Functional Threshold Power(FTP) Estimation for cyclists
    14. Aerobic Endurance analysis
    15. Intervals Analysis
    ​​​16. Training Planning
    17. Integration with Oura
    18. Aerobic efficiency and cardiac decoupling
    ​
    Other
    1. HRV normal values​
    ​2. HRV normalization by HR
    ​
    3. HRV 101

    RSS Feed

Picture
  • Home
  • QuickStart Guide
  • PRO & TEAMS
  • FAQ
  • Privacy & Terms
  • Contact
  • Publications
  • Guides
  • Blog
  • Shop