How long until Apple Watch measures blood sugar and blood pressure?

[Jackbequickly]

Not going to happen anytime soon.

 

[The-Real-Deal82]

It certainly is basic, it's no tricorder.

I wonder what incentives the consultant had, if any. I know healthcare is not monetized in the UK to the same extent as the US, for example, but it would not be unheard of for some upselling to go on.

It wasn’t recommended for any backhand sort of reasons and no specific product was mentioned, just a recommendation to get a dedicated product.

 

[msackey]

Wrist BP measurements tend to be inaccurate.

yes, they tend to be finicky.

I don't think they are inaccurate if the conditions around how the measurement is taken is done appropriately according to the manufacturer. For example, some wrist bp measuring devices require the wrist to be elevated to the heart level with the elbows supported, back and feet all supported with feet flat on the ground. It's not always easy to get this right.

 

[msackey]

Could they partner with already existing medical under-the-skin implants that monitor blood sugar levels to transmit the data to the Watch? That would be helpful and Apple would not have to rely on any unreliable sensor on the Watch itself.

I think something like this exists!

There are CGM devices that can transmit to apps on the iPhone. I don't know if those apps also appear on the Watch, but that should not be a hard thing to do at all. But I know CGM devices that transmit to app on iPhone definitely already is there.

Update: Here is it! https://www.dexcom.com/en-gb/faqs/can-i-view-my-dexcom-cgm-data-apple-watch

 

[TechnoMonk]

The watch is very basic in regards to being a medical device though. My wife was having irregular heart rhythms and used the ECG feature to record certain episodes. She had to buy a dedicated ECG monitor as her consultant said the Apple watch simply wasn't accurate enough. The marketing Apple use does suggest otherwise, but in practice it is very different. Nice to have features though on a watch and I have been an AW wearer for more than 6 years now.

It’s not gonna replace a high grade medical equipment but very good starting point. It’s not like Apple is claiming it’s replacing devices in hospitals.

 

[TechnoMonk]

Could they partner with already existing medical under-the-skin implants that monitor blood sugar levels to transmit the data to the Watch? That would be helpful and Apple would not have to rely on any unreliable sensor on the Watch itself.

At least in the US, you need a prescription, they last 2-3 weeks and need to be replaced. If Apple Watch can measure BG levels with out replacing sensors, and not needing a prescription, It would be amazing.

 

[teh_hunterer]

I think the recent advancements in A.I. will actually speed up the development of these AW features.

One part of this will be from the dev and r&d teams having a.i. assistance when creating the hardware and software algorithms.

Yeah... nah. AI like Chat GPT are just language prediction engines based on existing human output.

 

[JustAnExpat]

Wasn't this question asked recently? Both of these are about 10 years away.

There are non invasive blood sugar monitors that can be worn around the arm. They are called "Continuous Glucose Monitors". The monitor currently is applied to the skin as a patch, and the needle goes inside the skin. This allows glucose to be monitored for up to 14 days, and then the patch is removed. See https://www.verywellhealth.com/freestyle-libre-a-glucose-meter-without-a-finger-prick-4154266 for an example of one. I suppose blood sugar can be measured by proxy potentially, and using AI to estimate a potential range. However, this requires identifying what that proxy is (I guess sweat?), developing the AI system to measure blood sugar, and then confirming the AI system and the sensor is both accurate. Blood sugar levels is not something that has a wide range of error.

Blood pressure sensors may be possible, however multiple problems exist in developing that technology. This article https://www.frontiersin.org/articles/10.3389/fcvm.2019.00040/full provides a good summary of the existing problems.

 

[JustAnExpat]

Yeah... nah. AI like Chat GPT are just language prediction engines based on existing human output.

Chat GPT scares me. Not because of the technology behind it, but because most people don't understand what it is and think it's a computer actually thinking.

 

[teh_hunterer]

Chat GPT scares me. Not because of the technology behind it, but because most people don't understand what it is and think it's a computer actually thinking.

Absolutely. At the risk of going off topic, if anyone wants some actual clarity on what ChatGPT is, this article explains it very well: https://www.newyorker.com/science/a...elligence/what-kind-of-mind-does-chatgpt-have

Assisting Apple R&D, lol, no.

 

[TechnoMonk]

Wasn't this question asked recently? Both of these are about 10 years away.

There are non invasive blood sugar monitors that can be worn around the arm. They are called "Continuous Glucose Monitors". The monitor currently is applied to the skin as a patch, and the needle goes inside the skin. This allows glucose to be monitored for up to 14 days, and then the patch is removed. See https://www.verywellhealth.com/freestyle-libre-a-glucose-meter-without-a-finger-prick-4154266 for an example of one. I suppose blood sugar can be measured by proxy potentially, and using AI to estimate a potential range. However, this requires identifying what that proxy is (I guess sweat?), developing the AI system to measure blood sugar, and then confirming the AI system and the sensor is both accurate. Blood sugar levels is not something that has a wide range of error.

Blood pressure sensors may be possible, however multiple problems exist in developing that technology. This article https://www.frontiersin.org/articles/10.3389/fcvm.2019.00040/full provides a good summary of the existing problems.

I don’t want AI anywhere close to estimating BG levels. May be in the future, if these AI systems stop Hallucinating.

 

[Tagbert]

For blood pressure, even the wrist mounted pressure cuffs have erratic readings because it’s not a good location to get reliable reading. Also, unless you sit down and raise your arm to heart level and sit quietly while doing it, you are going to get highly variable results that don’t compare well to other readings.

 

[Tagbert]

I don’t want AI anywhere close to estimating BG levels. May be in the future, if these AI systems stop Hallucinating.

It would be more of the Machine Learning style of AI that Apple has been using for several years. That is not the kind of AI that hallucinates. It’s more of an algorithm that adapts its results as the data changes.

 

[jz0309]

Samsung released a smartwatch last year that offered blood pressure capabilities, never seen reports on accuracy though.
As for CGM (continuous glucose monitoring), the technology exists but needs further refinement, and more important will require (in the US) FDA approval as a medical device (Type 1 depends on accurate readings for Insulin intake). the "standard" AW will not ever get those capabilities as the model will have to go thru years-long approval process. There could be a "special" AW that offers it, it will be VERY expensive (I believe a current Dexcom invasive CGM costs you several thousand over 3-4 years and is only available via prescription.
What the "standard" AW might get is a "trend" reader, not exact values, similar to ECG ...

there are lots of health sensor developments in process, in time those will become available ...

 

[HDFan]

There are non invasive blood sugar monitors that can be worn around the arm. They are called "Continuous Glucose Monitors". The monitor currently is applied to the skin as a patch, and the needle goes inside the skin.

"the needle goes inside the skin." meets my definition of invasive.