Study uses Apple Watch heart rate sensor to detect serious heart condition with 97% accura...

Posted:
in Apple Watch edited May 2017
New research conducted by the University of California, San Francisco, in collaboration with Apple Watch app Cardiogram, shows Apple Watch's heart rate sensor can aid in the detection of atrial fibrillation, a common heart arrhythmia that can lead to stroke.




As part of ongoing research, a deep neural network was trained and paired with Apple Watch's heart rate sensor to automatically distinguish atrial fibrillation from normal heart rhythm in a pool of test patients. Findings were presented at the Heart Rhythm Society's Heart Rhythm 2017 conference on Thursday.

To train the DNN, researchers collected data -- 139 million heart rate measurements and 6,338 mobile ECGs -- from 6,158 Cardiogram app users enrolled with the UCSF Health eHeart Study.

Explaining the process to AppleInsider, Cardiogram cofounder Brandon Ballinger said about 200 participants with diagnosed paroxysmal atrial fibrillation took part in the study. These patients were provided a mobile electrocardiogram and tasked with taking one reading per day, or when they felt an onset of symptoms ranging from lightheadedness to heart pains. Cardiogram staff used the gathered information to train the DNN, which was subsequently paired with Apple Watch heart rate data to identify AF.

The DNN was recently validated against a test sample of 51 patients scheduled to undergo treatment. Each person wore an Apple Watch for 20 minutes before and after cardioversion, a procedure that restores normal heart rhythms to patients with arrhythmias.

When compared against a 12-lead electrocardiogram reference, the resulting Apple Watch and DNN solution was found capable of identifying AF with an accuracy of 97 percent, sensitivity of 98 percent and specificity of 90.2 percent, all high marks compared to past detection algorithms.

"Our results show that common wearable trackers like smartwatches present a novel opportunity to monitor, capture and prompt medical therapy for atrial fibrillation without any active effort from patients," said the report's senior author Gregory M. Marcus, MD, MAS Endowed Professor of Atrial Fibrillation Research and Director of Clinical Research for the Division of Cardiology at UCSF. "While mobile technology screening won't replace more conventional monitoring methods, it has the potential to successfully screen those at an increased risk and lower the number of undiagnosed cases of AF."

At its heart, the project seeks to address AF through the use of common consumer devices. Currently, medical devices like Holter Monitors and wireless patches can monitor patient heart rhythms for 24 hours to 4 weeks, but these methods could take some 84 days to detect the first signs of AF. With the right software, new wearable technology like Apple Watch and other devices with accurate heart rate sensors can provide healthcare professionals with a more effective means of monitoring patients on a long-term basis.

The study builds on research first published last March. At the time, Cardiogram cofounders Ballinger and Johnson Hsieh were in the process of creating a preliminary machine learning algorithm capable of detecting AF using consumer grade heart rate sensors.

Ballinger and his team at Cardiogram are currently validating the DNN against a series of industry standards with plans to incorporate results into the app. Looking ahead, Cardiogram is investigating whether the DNN can be applied to other heart conditions.
«1

Comments

  • Reply 1 of 21
    netroxnetrox Posts: 1,464member
    This is what makes me think about getting Apple Watch but I am waiting for next generation.... I want it to monitor my sleep as well.
    jahbladealbegarc
  • Reply 2 of 21
    glynhglynh Posts: 133member
    Monitoring sleep would be a really nice feature to have but the ability to monitor blood glucose would IMHO be a game-changer.
    Soliradster360ravnorodom
  • Reply 3 of 21
    lkrupplkrupp Posts: 10,557member
    Apple already dominates the wearable market. If the Apple Watch becomes the de facto wearable medical device then the sky is the limit. Heart patient, diabetes patient, whatever, walks into their doctor’s office. The doctor writes a prescription for an Apple device specifically designed for the purpose, and Medicare pays for it like they pay for other necessary medical devices like oxygen tanks and the like. Or Apple licenses the technology to approved medical device manufacturers. The device uses AI and will notify the doctor’s office when an event occurs and send a quick analysis along too. It might also call 911 and request medical assistance. The imagination goes wild with the possibilities.
    edited May 2017 jahbladealbegarcwatto_cobrabrucemc
  • Reply 4 of 21
    cyberzombiecyberzombie Posts: 258member
    You don't need to wait for that. (note: just a satisfied customer) The apps HeartWatch and AutoSleep include Apple Watch apps. I use these to check my heartrate at any point (with much better history display than the Apple app) and AutoSleep tells me how long I've slept and the quality of that sleep (using the watch accelerometer to detect when I've gone to sleep so I don't have to do anything). With my old series 1 watch, I charged in either mornings or evenings so I could sleep with it. With this new 42mm series 2, I only have to charge every 2.5 days... Note: Just finished my Dr. checkup. At the point where my only options are to go on to insulin or completely change my lifestyle with exercise and dropping most the carbs. I still think I'll get 1.5 days out of my S2 (S1 was 1.5 days with no exercise). i do NOT want to go on to insulin.
    jahbladenetroxwatto_cobrabrucemc
  • Reply 5 of 21
    netroxnetrox Posts: 1,464member
    You don't need to wait for that. (note: just a satisfied customer) The apps HeartWatch and AutoSleep include Apple Watch apps. I use these to check my heartrate at any point (with much better history display than the Apple app) and AutoSleep tells me how long I've slept and the quality of that sleep (using the watch accelerometer to detect when I've gone to sleep so I don't have to do anything). With my old series 1 watch, I charged in either mornings or evenings so I could sleep with it. With this new 42mm series 2, I only have to charge every 2.5 days... Note: Just finished my Dr. checkup. At the point where my only options are to go on to insulin or completely change my lifestyle with exercise and dropping most the carbs. I still think I'll get 1.5 days out of my S2 (S1 was 1.5 days with no exercise). i do NOT want to go on to insulin.
    Thank you, much appreciated!
    watto_cobra
  • Reply 6 of 21
    kenckenc Posts: 195member
    Hope the validation is successful. Carrying around and wearing a Holter monitor is quite a nuisance. My mom has an Apple Watch and the AliveCor device that sticks to the back so she can take her single-trace EKG whenever she feels odd. Works extremely well. Hopefully the AliveCor Apple Watch band will be out soon.
    watto_cobra
  • Reply 7 of 21
    cjcampbellcjcampbell Posts: 115member
    I use the Sleep++ app. Works just fine. I recharge my watch while I'm getting dressed in the morning and it stays fully charged.
    watto_cobra
  • Reply 8 of 21
    I like the concept but Apple makes it really hard to find research kit apps
  • Reply 9 of 21
    cjcampbellcjcampbell Posts: 115member
    Cardiogram is in the Apple Store just like any other app. It is free.
    watto_cobra
  • Reply 10 of 21
    dws-2dws-2 Posts: 277member
    That's surprising. I've tried multiple Apple watches, and none of them can consistently track my heart rate. I can run at full sprint, and my heart rate is 55, or I can lay in bed with a heart rate of 140, then suddenly a heart rate back down to 68. My resting heart rate at the doctor's office was 58 a few weeks ago.

    It's all over the map. I'm sure it's right at some point, but so is a broken clock.
    edited May 2017
  • Reply 11 of 21
    StrangeDaysStrangeDays Posts: 12,955member
    dws-2 said:
    That's surprising. I've tried multiple Apple watches, and none of them can consistently track my heart rate. I can run at full sprint, and my heart rate is 55, or I can lay in bed with a heart rate of 140, then suddenly a heart rate back down to 68. My resting heart rate at the doctor's office was 58 a few weeks ago.

    It's all over the map. I'm sure it's right at some point, but so is a broken clock.
    Quite strange. Besides no real problems myself, several articles have declared it the most accurate of wrist-worn monitors. 
    edited May 2017 watto_cobra
  • Reply 12 of 21
    iqatedoiqatedo Posts: 1,825member
    dws-2 said:
    That's surprising. I've tried multiple Apple watches, and none of them can consistently track my heart rate. I can run at full sprint, and my heart rate is 55, or I can lay in bed with a heart rate of 140, then suddenly a heart rate back down to 68. My resting heart rate at the doctor's office was 58 a few weeks ago.

    It's all over the map. I'm sure it's right at some point, but so is a broken clock.
    I believe that one (big) problem Apple faced prior to introduction of the first watch, was building a device that would work with all skin types. My results are quite different to yours. The watch tracks quite well. Some blood vessels are either visible or hinted at quite strongly on my wrist. Do you pull the watch in tight?
    watto_cobra
  • Reply 13 of 21
    sflocalsflocal Posts: 6,110member
    It's only a matter of time before the Samsung's of the world attempt to (badly) copy their medical tech and try to pass it off as reliable as Apple's hardware.

    You never hear anything about anyone attempting all this on an Android watch.  That alone speaks volumes of the lack of interest using Android for something as important and lifesaving as a medical device.
    pscooter63watto_cobra
  • Reply 14 of 21
    anantksundaramanantksundaram Posts: 20,407member
    dws-2 said:
    That's surprising. I've tried multiple Apple watches, and none of them can consistently track my heart rate. I can run at full sprint, and my heart rate is 55, or I can lay in bed with a heart rate of 140, then suddenly a heart rate back down to 68. My resting heart rate at the doctor's office was 58 a few weeks ago.

    It's all over the map. I'm sure it's right at some point, but so is a broken clock.
    Did you try calibrating it initially with a 20-minute walk/run as Apple recommends? See here for tips to help you get the right reading: https://www.macrumors.com/how-to/apple-watch-accurate-heart-rate-reading/
    edited May 2017 watto_cobracornchip
  • Reply 15 of 21
    croprcropr Posts: 1,133member
    97% is indeed very high and a big achievement. But in the medical world it is not good enough.  What if a doctor would say to a patient:" I am 97% sure that you don't have cancer"  Would you feel reassured?  I doubt it.
    edited May 2017
  • Reply 16 of 21
    nhtnht Posts: 4,522member
    cropr said:
    97% is indeed very high and a big achievement. But in the medical world it is not good enough.  What if a doctor would say to a patient:" I am 97% sure that you don't have cancer"  Would you feel reassured?  I doubt it.
    Yes I would and have been.  The 5 year survival rates of many cancers are lower than 97% and recurrence rates higher than 3%.

    Medicine doesn't deal with absolutes.
    cornchipStrangeDays
  • Reply 17 of 21
    farmboyfarmboy Posts: 152member
    nht said:
    cropr said:
    97% is indeed very high and a big achievement. But in the medical world it is not good enough.  What if a doctor would say to a patient:" I am 97% sure that you don't have cancer"  Would you feel reassured?  I doubt it.
    Yes I would and have been.  The 5 year survival rates of many cancers are lower than 97% and recurrence rates higher than 3%.

    Medicine doesn't deal with absolutes.
    Me too. No doctor worth his/her malpractice insurance will give you an absolute. And patients shouldn't expect that, because any medical measurement of anything, or any statement by any medical professional is just a snapshot of your condition today, at that moment. Not next month or next year. Relax, take your meds, don't be an idiot and live your life.

    Ever notice that popularly-reported medical study stats are universally reported as "increases your risk / lowers your risk of..." without ever telling you what the original risk is and what the new risk is if the study results prove to be valid? If there is generally a 4% risk of colon cancer, and a new study says red meat every day will increase your risk a massive 25% (headline news), that means the risk is now 5% if the study is true and you are that 1-in-20 guy eating steak for breakfast (NOT headline news).
  • Reply 18 of 21
    brucemcbrucemc Posts: 1,541member
    glynh said:
    Monitoring sleep would be a really nice feature to have but the ability to monitor blood glucose would IMHO be a game-changer.
    It certainly would be, but we are talking bleeding edge "research" in this area, which often can be many years from an actual product, if not much longer.  I am glad that Apple does invest in these areas which can have a big payout, but are long-term bets.  

    That said, there should still be health benefits (like this study shows) that come from more accurate heart rate monitors, measuring internal temperatures, possibly detected blood pressure, etc.

    IMO Apple will go down the route with a smart band concept.

    /BMc
  • Reply 19 of 21
    foggyhillfoggyhill Posts: 4,767member
    farmboy said:
    nht said:
    cropr said:
    97% is indeed very high and a big achievement. But in the medical world it is not good enough.  What if a doctor would say to a patient:" I am 97% sure that you don't have cancer"  Would you feel reassured?  I doubt it.
    Yes I would and have been.  The 5 year survival rates of many cancers are lower than 97% and recurrence rates higher than 3%.

    Medicine doesn't deal with absolutes.
    Me too. No doctor worth his/her malpractice insurance will give you an absolute. And patients shouldn't expect that, because any medical measurement of anything, or any statement by any medical professional is just a snapshot of your condition today, at that moment. Not next month or next year. Relax, take your meds, don't be an idiot and live your life.

    Ever notice that popularly-reported medical study stats are universally reported as "increases your risk / lowers your risk of..." without ever telling you what the original risk is and what the new risk is if the study results prove to be valid? If there is generally a 4% risk of colon cancer, and a new study says red meat every day will increase your risk a massive 25% (headline news), that means the risk is now 5% if the study is true and you are that 1-in-20 guy eating steak for breakfast (NOT headline news).
    Well, even that could be misleading as someone with a high meat intake may have also many other things that impact his health.
    That what makes any one single study near useless and even meta studies barely useful except as broad guidelines.

    Even with things with distinct large metabolic impact like high glycemic index food consumption have been hard to prove harmful concluviely and broadly (though this seemingly has been the case progressively in the last decade).

    The reliance on single studies (or a few badly done ones, or even worse agenda driven ones) have lead to dietary recommendations being a mess generally.


  • Reply 20 of 21
    cornchipcornchip Posts: 1,954member
    sflocal said:
    It's only a matter of time before the Samsung's of the world attempt to (badly) copy their medical tech and try to pass it off as reliable as Apple's hardware.

    You never hear anything about anyone attempting all this on an Android watch.  That alone speaks volumes of the lack of interest using Android for something as important and lifesaving as a medical device.


    But you're right, what ever happened with that?
    edited May 2017
Sign In or Register to comment.