Doctors used an ER patient's FitBit data to justify electrical cardioversion | Hacker News Reader

Doctors used an ER patient's FitBit data to justify electrical cardioversion

slate.com

87 pointscurtis35 comments

We (Cardiogram) have been working with UCSF cardiology on exactly this problem: detecting abnormal heart rhythms using Apple Watch data.

The more data our algorithm sees, the more accurate it gets. If you own an Apple Watch, you can join our study / download Cardiogram for iOS: https://mRhythmStudy.org https://itunes.apple.com/us/app/cardiogram/id1000017994?ls=1...

We also gave a talk at Strata last week going into some of the technical and organizational challenges on applying deep learning to medicine: conferences.oreilly.com/strata/hadoop-big-data-ca/public/schedule/detail/47144

If anybody has questions on heart rate sensors, machine learning in medicine, or anything else, shout out!

And if you're curious as to what atrial fibrillation looks like when measured on an Apple Watch in workout mode, here's an example (bottom-right corner): https://twitter.com/bballinger/status/714500654308306946

I've been participating. I noticed that when I'm doing a workout my Watch"s heart rate monitor might drop to 65 bpm from 120, for example, then recover on another reading. Does this mean that the Watch isn't always accurate?

Thank you for participating! And that's exactly what we're measuring—how accurate is the Watch, and in what circumstances does it make errors? The anecdotal evidence is very conflicting, which is why we're comparing the AW to a few different clinical gold standards to get hard metrics.

Looks awesome. Did you find HIPAA or the need for an extremely low error rate to be burdensome in building the app?

We're lucky in that information covered by HIPAA (e.g., the medical record) is handled by our academic partner at UCSF, the Health eHeart study (health-eheartstudy.org).

I think it's a bit of a misconception that you always need an extremely low error rate in medicine. Many of the existing tools are quite poor. CHADS-VASC, for example, is the primary score that cardiologists use to prescribe blood thinners, and its c-statistic is an... unimpressive 0.673. (0.5 is random, and 1.0 is perfect.) Given that the side effects of unnecessary blood thinners include brain hemorrhage, even a modest improvement to CHADS-VASC could save hundreds of thousands of lives.

So my message to data scientists considering going into healthcare is... please come! You don't have to achieve perfection to save lives. Even a mediocre statistical model would be an improvement in so many areas.

This is really cool! It sounds like you're trying to determine whether you can identify AF out of data that's complicated by poor recording quality and regular activity? Is your longer term goal/business model to use this as a diagnostic tool (i.e. replace a Holter monitor)?

And since you offered to answer questions, I have a somewhat niche one. I'm an MD/PhD student doing my research in computational electrophysiological modeling, and I'm getting interested in the idea of using machine learning in medicine, specifically in signals analysis (i.e. EKG), and in time-series analysis (like heart rate variability). I'm teaching myself machine learning via a fantastic UW Coursera series that covers the basics (regression, classification, clustering, etc) with good mathematical rigor, but I'm not really sure where to go from there. What techniques would you suggest I look at? Would learning something like TensorFlow (I know Google has put out a course on Udacity) be a good idea? Obviously at some point, I just need to find some small projects to jump in on, but I want to build a decent skillset first if I can. Any thoughts would be appreciated!

If we succeed in training an accurate enough machine learning algorithm, I think the primary purpose will be to screen for asymptomatic atrial fibrillation, and then use a Holter or Zio Patch for a final diagnosis.

The main benefit is that you can prevent cryptogenic strokes, which are often caused by undiagnosed atrial fibrillation: http://www.nejm.org/doi/full/10.1056/NEJMoa1311376

For machine learning, I think you're on exactly the right path. The Udacity TensorFlow course taught by the Google Brain team is good; if you can complete that, you'll be as good as anyone at applying neural networks. The Stanford courses (http://cs231n.stanford.edu/, http://cs224d.stanford.edu/) are likewise great, as is Chris Olah's blog for intuition (http://colah.github.io/).

And down the road, if you're ever looking for an internship, definitely let us know! brandon@cardiogr.am.

Having worked with some physicians on AF diagnostic instrumentation, I took away that an outsized difficulty in this diagnosis is patient compliance. One doc called it a "one shower" problem; very many patients would wear holters as prescribed until they needed to take it off, and then fail to don it again. Clearly this is an issue when hunting for an infrequent arrhythmia, and it seems that the comfort and ease of use of the device are inherent to the problem.

From what you're gathering, do you imagine an eventuality in which a medical device can take a form factor similar to these popular consumer choices and provide a signal of diagnostic quality? Or do you think PPG-style sensing is limited to a fundamental degree of uncertainty with respect to arrhythmias?

Yeah! Absolutely. In healthcare we call it "patient compliance," but it's really just usability.

My guess is that by the time we get to the Apple Watch or Android Wear 3 or 4, it'll include a built-in ECG sensor. There are some devices out there already (like Nymi) that have built ECG into a wristband -- if the user touches a wristband worn on one hand with a finger on the other hand, that completes a circuit through the heart and gives you a 1-lead ECG.

I think the downstream implication is that, just like smart phones replaced dedicated GPSes and music players, many medical use cases will become "apps" that run on general purpose wristbands. The big advantage is that since they'll be always-on and non-intrusive, we'll be able to catch conditions much earlier.

As somebody who has both A. had a heart attack and B. is working on a machine learning project now, this is a subject that's very, erm, near to my heart ( no pun intended really).

Do you have an email address you can be reached at brandonb? I'd love to chat with you sometime if you're willing. I'm definitely interested in ways to apply machine learning to health-care.

Sure! Hit me up at brandon@cardiogr.am.

Do you have any plans for integrating other sensors, such as many HRV capable bluetooth smart HR straps on the market?

edit: and other platforms, such as android

Yes! We plan to support other devices like Fitbit and Android Wear. My co-founder and I both worked on Android at Google, so we'd love to get back to Android Studio.

I downloaded the app, logged in with Facebook, and then went to mRhythmStudy.org only to find out that signing up for the study wants an email/password login. I don't have that since I used Facebook. And if I Sign Up for a new account, I don't see any way in the app to log out and back in with the new email/password account.

Oh sorry—that's confusing. You'll actually have two accounts. One for the Cardiogram app using your choice of Facebook / Twitter / Email, and one for Health eHeart study (our partners at UCSF) using email.

When you sign up for mRhythm on health-eheartstudy.org, create a new account, and then you'll be guided through a few survey questions and have the ability to link your Cardiogram account with your Health eHeart account. That link is what lets us do the research.

Does that make more sense?

Ok yeah that does make sense.

Could you point to a UCSF .edu page validating the mRhythm study and this app's connection to it?

Our academic partner at UCSF Cardiology hosts their study on their own domain (https://www.health-eheartstudy.org/). If you sign up through the links to Health eHeart provided at mRhythmStudy.org, you'll see a customized flow for the mRhythmStudy which lets you link your Cardiogram account.

Our principle investigator, Dr. Greg Marcus, also spoke briefly with with Fortune when the mRhythm study launched:

  https://fortune.com/2016/03/16/apple-watch-cardiogram-stroke-study/

  http://profiles.ucsf.edu/gregory.marcus

Wouldn't it make more sense to use a device that's not locked in to a specific vendor?

We do plan to support other devices like Fitbit and Android Wear. The reason we started with Apple Watch is that 1) the heart rate sensor is quite good, and 2) we can target millions of people with one platform (iOS / Watch OS).

I thought the HR detection on consumer fitness gear wasn't very accurate, yet we are able to base medical decisions off it? Have I been misled?

They're not very accurate. But some heart rate symptoms are obvious enough to make even poor quality measurements useful enough to make a medical decision.

Like atrial fibrillation, which I UST found out I have. I've been capturing my heart rate from a polar h7 chest band to an Android app, then visualising with grafanas stddev. The pattern is strikingly obvious

This is covered in the article.

Yeah, the article mentions that they only cover pulse tracking, not arrhythmias, but I was under the impression that even the pulse tracking was somewhat iffy due to the nature of how the tracking is performed.

In fact, the article says there's a class action lawsuit because of it, which means using this data in a medical context could have gone very badly potentially.

Given that they were seeing if a precondition is true (arrhythmia started within last 48 hours), and the data showed that there was a significant change over time, it seems somewhat reasonable to act on if the patient wasn't totally sure.

Physicians usually want to know what is different or regular for the specific patient over time, whether or not a value seems high for one person doesn't mean that action needs to be taken--but when that value changes significantly or shows a trend in a particular direction over time, that is information relevant to what procedure they decide next.

It is entirely possible that the fitbit started recording data differently at or after the seizure, supposing it got shifted in the event. However, that does not invalidate the trend of the data as a whole. In this case, the patient was seemingly able to help acquire such data. If the patient said they were one way or the other, or if the patient depended on a device to fulfill that judgement, is it really the physician's fault to choose what the patient provides? IANAL

I may have been told wrong, but I'm sure I recall being told that some consumer heart rate monitors (the kind with a chest strap) are approximately as accurate as what's used in a doctor's office. Now something that doesn't use a chest strap... I don't know. Probably those are less accurate? And/or maybe I'm the one who's been misled.

According to this article http://www.tomsguide.com/us/heart-rate-monitor,review-2885.h..., several consumer products are accurate to within 1 beat of an EKG.

This is certainly true for the optical sensors. I wonder what the long term effect is of shining a bright LED at a small patch of skin like that?

I'm not convinced here. Sure, they may have wanted to look at the FitBit for confirmation, but surely the patient could tell them it had been only three hours since he started to have trouble? You're not going to confuse three and twenty four hours.

I'd have to agree with you. I cant see how the fitbit data was important in the decision making process.. it was mostly just an extra confirmation. Definitely something the patient can verbally provide to the care team in most situations.

From the article: "It wasn't clear, though, when it had started"

Like I said, I'm skeptical. When your heart rate starts to soar you know it.

I recommend using https://www.alivecor.com/ (I have one) - this one actually performs ECG (can be used on hands, but the best results are when used on one's chest). Just measuring pulse gives a very indirect, and unspecific, information of arrhythmia. (Still, it is great that even such data can be used to save health or live.)

Reply on news.ycombinator.com