Researchers Struggle to Use Apple Watch in Studies Due to Sudden Algorithm Changes

[MacRumors]

Researchers are rethinking plans to use the Apple Watch in studies after finding inconsistencies in the data gathered by the device (via The Verge).

Inconsistencies in the Apple Watch's heart rate variability data are reportedly caused by Apple tweaking the device's algorithms, meaning that data from the same time period can change without warning. Associate professor of biostatistics at the Harvard T.H. Chan School of Public Health and developer of the Beiwe data platform, J.P. Onnela, told The Verge:

These algorithms are what we would call black boxes — they're not transparent. So it's impossible to know what's in them.

Research fellows at Brigham and Women's Hospital exported the heart rate data from Apple Watches for the same period of time, but several months apart. The data should have been identical, but since it is filtered through an algorithm prior to export it was drastically different.

Apple changes its algorithms regularly and without warning, so exporting the same data at different times may use different algorithms. Beyond heart rate variability, researchers looking into sleep tracking have experienced similar problems with changes to algorithms.

This mostly precludes commercial devices from being used by researchers, resulting in the need for devices specifically designed to collect data for scientific studies. While this is adequate for some studies, it heavily constrains research into the medical value of commercially available products for users. Some researchers say that Apple should publicize the changes it makes to algorithms or make the Apple Watch's raw data available.

Article Link: Researchers Struggle to Use Apple Watch in Studies Due to Sudden Algorithm Changes

 

[csfossett]

I have to admit I find this incredibly disconcerting. I get that companies like Apple are commercially driven, and algorithms tend to be in service to that, but they're also touting the Apple Watch as a major healthcare related device. Shouldn't medical/biological diagnoses and research be driven purely by data and information, and not by what amounts to programming developed to interpret data in advance of receiving it?

 

[NMBob]

There's no one there to say "no" anymore.

 

[hailey bee]

It's the same principle as with raw format in photos – pro users need raw data.

 

[Alfredo_Delgado]

I'm going to guess Apple didn't make the raw data available to avoid this very case where different applications could interpret the data differently. That at least reduces the number of culprits responsible for sketchy data interpretation to just Apple.

I wonder if they could put something in the EULA to make raw data available for scientific studies along with a vanilla sample algorithm for consuming it. Any studies could then publish the algorithm they settled on -- whether baseline or tweaked -- along with their raw data.

 

[jerryk]

Welcome to working with data. Getting and pre-processing data to improve quality uses upward of 70% of the project time. Data is constantly changing due to legal or technical issues or to make it better for users other than you. Any study based on data needs to be able to handle these changes.

 

[Velemaker]

Provide raw data

 

[sw1tcher]

Research fellows at Brigham and Women's Hospital exported the heart rate data from Apple Watches for the same period of time, but several months apart. The data should have been identical, but since it is filtered through an algorithm prior to export it was drastically different.

I'll call this a built-in feature. Since the data changes each time it's exported, it makes it harder for them to track you. This is Apple putting privacy above accuracy. 🤣

 

[Malus120]

This is so incredibly disappointing, and yet also so undeniably Apple.

It's both funny and sad that one of the few aspects of Steve Jobs Apple that Tim Cook has chosen not to change, the company's intense secrecy, continues to be its downfall in so many current and potential markets.

Apple is fortunate to be the only real player in the wearables market at the moment. Despite the complaints, I would imagine most researchers will continue to work with Apple for the time being as they're the only game in town. That said, just as in the pro and corporate markets, Apple will eventually struggle to retain research clients in the long run if it continues to cling to its asinine refusal to provide even a basic level of transparency.

If Apple wants to expand outside of the consumer market in a serious way it needs to understand that secrecy is not the asset it is there but a serious liability that many folks just aren't going to be willing or able to deal with.

 

[nt5672]

More arrogance from Apple. They seem to think they know everything better than everyone else. Of course, since this hit the media, maybe Apple will will consider making the raw data available as it should have been in the first place.

Of course, this also fits with Apple's attitude that it is not true unless Apple says it's true.

 

[rjohnstone]

Welcome to working with data. Getting and pre-processing data to improve quality uses upward of 70% of the project time. Data is constantly changing due to legal or technical issues or to make it better for users other than you. Any study based on data needs to be able to handle these changes.

In order for a study to account for the changes, it needs to know what was changed. It would need the raw data or the algorithms to compare, so they can account for those changes.
Since Apple doesn't provide that information, the data collected is largely useless.

 

[Mrjetsondc]

Apple: this watch is your health monitor!

also Apple: lol psych thanks for the 400 bucks

 

[Khedron]

Either Apple comes out with clarification or all medical studies using Apple devices should be retracted and rejected.

No data is better than bad data. Apple is playing marketing games with public health.

 

[augustrushrox]

Example #1957502759 of why “walled garden” approaches tend to be hamfisted.

 

[jerryk]

Provide raw data

Because of memory constraints, I bet they don't store the actual data stream coming off the sensors. Rather the software on the watch pre-processes the data. Perhaps the Apple Watch has a streaming mode that is accessible to apps on a connected phone. I know some Samsung watches have this.

 

[Wackery]

Everyone here is a medical expert apparently

 

[Pezimak]

If this is true then I would question the authenticity of the approval the medical associations around the globe have awarded the device. Many do use their Apple watches to keep an eye on heart related issues but I bet they do not know Apple is potentially changing their recorded data at will, without warning.

Apple needs to be careful with the way it advertises its health sensors, and it’s health programmes that it tends to present as being very proud off. It does give an impression of reliable health data is being presented, which this article seems to prove otherwise.

 

[rjohnstone]

Everyone here is a medical expert apparently

You don't have to be a medical expert to understand data analysis.
If the baseline data is changed via an algorithm update, you need to know how that algorithm was changed to account the data discrepancy.
Without the info, the data becomes useless.

 

[dsdevries]

I found the resting heart rate and heart rate variability data from the Apple Watch quite useless too. As an athlete, it's quite important that these measurements are taken every morning right after you wake up, or just before. However, I also found that Apple changed this behaviour multiple times. At one point, the measurements were taken while I was already under the shower, or even when I was already on my bicycle. Utterly useless, so I switched back to Elite HRV with a heart rate strap.

 

[deadlock]

Are these medical studies that Apple has advertised themselves that they are supporting, or ones that are trying to use data exports to do a studied but not officially partnered?

If the former then that is worrying as I’d expect Apple to be giving them the raw data and access to the algorithms (possibly under NDA).

if it’s the latter then why haven’t they partnered officially?

 

[matrix07]

There's no one there to say "no" anymore.

How’s that relevant?

 

[Khedron]

Are these medical studies that Apple has advertised themselves that they are supporting, or ones that are trying to use data exports to do a studied but not officially partnered?

If the former then that is worrying as I’d expect Apple to be giving them the raw data and access to the algorithms (possibly under NDA).

if it’s the latter then why haven’t they partnered officially?

Apple make it possible to share your data with your doctor. Does every watch owner now need to partner with Apple to get access to the ‘real’ data?

 

[Brandon42]

https://xkcd.com/2494/
Perfect timing from XKCD as usual.

 

[ericinboston]

<sarcasmoverload> Apple being compared to a black box? Never! </sarcasmoverload>

 

[matrix07]

Everyone here is a medical expert apparently

I don’t even think they read the same news as I. 😳

It sucks that researchers having a hard time with Apple Watch but it’s not wrong for Apple to change their algorithm periodically if they think it’s a better one.