Heartbeats and heart attacks (2016) | Hacker News Reader

Heartbeats and heart attacks (2016)

tabletopwhale.com

86 pointsTomte17 comments

Funny story. I have occasional irregular heartbeats. A number of years ago I was given a monitor to take home and use periodically when I was getting episodes of skipped beats.

This was a while ago, and after I ran it for about an hour I was supposed to call a phone number and basically modem in the data.

I dutifully uploaded my episodes for a few weeks, and apparently the doctor or specialist who was reviewing the data saw nothing out of the ordinary.

With a friend of mine we decided to reverse-engineer the signal the device was making to the telephone. It turns out it was a simple frequency modulation of the EKG Trace.

Long story short, we found the skipped beats, printed out the traces, presented our own printouts to my doctor and confirmed a diagnosis of premature atrial contraction.

There's something about routine diagnostics that causes docs to not put a lot of effort into interpreting the data. I had a sleep study plus MSLT done because my doctor suspected narcolepsy. The MSLT results came back just shy of the textbook definition of narcolepsy. Basically you take 5 15-minute naps and if you fall into REM sleep during 3 or more of the naps, insurance will accept a narcolepsy diagnosis.

When I read the summary though, it was obvious there were at least some mistakes in interpreting the data. I can't remember the details, but my doctor wasn't impressed and wanted to look at the data himself. Long story short, the place I got the test done at had no protocol for transferring that data anywhere. Which, btw, is a requirement under HIPAA. I have a legal right to have access to that data and to have it sent to another medical professional. From what I understand, the software they use has a proprietary file type and my doctor couldn't open it with anything he had. They could print it out, but it would be hundreds of pages and I'm not convinced they actually would do that unless I threatened to file a HIPAA report.

I ended up having to go to a different facility for another sleep study so that my doctor could actually see what was going on. If you've ever had one done, you know exactly how pleasant they are. 2 within 2 months of each other because the doc at the first place phoned it in is just ridiculous.

The data from the last study was clear enough that my doc was able to diagnose me with narcolepsy. That diagnosis allowed me to have Xyrem covered by my insurance. Between that and nuvigil, I'm still tired sometimes, but not the way I used to be. It took 8 years of being exhausted and miserable to finally find a doc that had the knowledge and the motivation to figure out what the hell was wrong with me. Even with a good doc, I still had to deal with the lack of effort some docs put into interpreting routine diagnostic tests.

Data mobility is a huge problem in the sciences and medical field. For years, companies have been leeching off researchers by trapping them into their proprietary data types which forces them to purchase expensive software licences and prevents sharing due to the need for again, expensive software

This is why open source and open data is important

Can you elaborate on this? I was working on few expensive, proprietary software for the apnea & ECG analysis and they had rather simple method for exporting to the EDF format.

> my doc was able to diagnose me with narcolepsy. That diagnosis allowed me to have Xyrem covered by my insurance.

That’s a very effective but highly controlled substance.

My introduction to Xyrem was via a simple statement: “Sometimes people sleep but don’t get rested. Those people can take gamma hydroxybutyrate, which is found in passionfruit.” I haven’t found any confirmation of passionfruit having this substance. I got a friend some passionflower tea, and she found it remarkably effective for helping her drift off to sleep.

It’s a safe simple compound, easy to synthesize with lab chemicals. But it also has nefarious uses, so it got scheduled.

“Sodium oxybate is the sodium salt of γ-hydroxybutyric acid (GHB).”

Honestly I'm surprised it's even on the market. For as long as it's taking to get fairly harmless marijuana approved for medical use, I'm not sure if any actual rational reason why Xyrem isn't facing the same issue. GHB can definitely kill you if you misuse it or combine it with other CNS depressants and has recreational value.

GHB is also used in some European countries to help people quit alcohol. That's not why I started taking it, but it's largely replaced my alcohol consumption even on the weekends and to my knowledge I'm healthier for it. I can't take both on account of it can kill you and Xyrem knocks me out without giving me a hangover the next morning. I'm finding myself less and less interested in drinking even on days where I don't have to function in the morning.

The new monitors are great - I had a Zio patch stuck on me for two weeks, also due to some ectopic heartbeats (especially when exercising) with a button to press and a notepad to log symptoms when button was pressed.

The cardiologist had the entire two weeks of EKG trace, with the times I'd pressed the button (and a minute either side) highlighted, as well as other times it had detected something out of the ordinary.

Luckily in my case turns out I have fewer ectopic beats than the average person, I'm just more aware of them.

Maybe your doctor further explained this but PACs are extremely common and not considered out of ordinary. Not sure if you were implying that the first doc missed it or not.

The kardiamobile 6-lead EKG device is down to $100 now, it's rather fascinating to use daily especially if you have heart concerns.

It's actually a very simple device, recording the electrical impulses between the hands and a third point like a knee contact to make six signals which are broadcasted as positive/negative numbers describing the curves/spikes over bluetooth.

It just makes simple small files on the mobile device with those list of numbers, which someone could write a program to intercept and interpret far beyond what they offer for free.

For legal reasons between most device makers and the FDA it appears they are not allowed to diagnose more than a few kinds of arrhythmias like afib but there is no reason why a bio-hacker collective couldn't figure out more automated interpretations.

adding: note you can not only analyze HRV info from EKG but also detect several different kinds of electrolyte disorders (too much or too little of potassium, calcium, magnesium etc)

https://en.ecgpedia.org/wiki/Electrolyte_Disorders

https://ecgwaves.com/topic/ecg-electrolyte-imbalance-electro...

Just want to point out that Atrial Fibrillation and Atrial Flutter are (usually) very benign, chronic conditions. Many people have for years without symptoms.

Ventricular fibrillation, ventricular tachycardia and torsades are acutely deadly. You have seconds to minutes to get a normal rhythm back.

The charts colors made me feel like they were implying a-fib was the worst rhythm listed. (Though they list the severity in the descriptions in small text)

While atrial fibrillation does not present anything close to the immediate and critical threat of the three you mentioned, it should not be ignored, as it increases the risk of stroke and embolism through blood pooling and clotting in the left atrial appendage of the heart.

When it is not asymptomatic, it is difficult to ignore anyway.

There is the problem that the irregularities may only appear infrequently. Often data collection may be done for a day or multiple days to capture irregular patterns and try to understand how various activities impact those. Seems unlikely that one would be able to self diagnose with an Apple Watch due to limited data. Unless the condition was very dominate.

I find the infographic terrifying.

Though I went through a phase in uni (probably due to finals stress + drugs) where I had heart palpitation and would feel palid out of nowhere, and just how scary/futile it feels when your heart is acting up.

Anyways, I like to think that my daily exercise does something to counteract my drug use. The waveforms in this infographic will surely pop into my head next time I'm passing around a baggy of coke and notice my chest a tad tighter than usual.

Don't mess with the ticker.

Can I use these charts with the Apple watch single probe ECG graph as seen on the iPhone after a 30sec ECG capture?

Interesting but would have been more useful if the charts showed slight variations you could look for in the ekg that were suggestive of developing problems. anyone who’s seen any amount of tv would be able to tell looking at any of the abnormal charts there that “you either bugs bunny or you ded”

I think sports chest straps get overlooked even talking about this space.

DCraimaker is a very active and does deep statistical analysis and recording for comparing many types of chest straps and wrist straps. Worth checking out. [3]

The polar H10 has been compared to medical grade heart rate monitors and has is as good if not better. https://pubmed.ncbi.nlm.nih.gov/31004219/ shows an evaluation for EKG. Polar also has sdk to develop apps.

It's the "gold standard" but there have been more and more advances in the space. But the polar h10 has stood the test of time.

If you're into fitness, and happen to have a polar H10 you can get tons of data from just wearing the strap overnight. Via this recommended android app. It supports mqtt as well. You can see live or you can record to file. There's no cloud involved and you can export directly from app.

https://play.google.com/store/apps/details?id=com.j_ware.pol... [1]

My tried and true app I have been using for workouts that's also gives me a Pulseometer RR. It doesn't do EKG but I have recording of over 18 hours no problem. Its a little out of date, in terms of not getting updates. But it is solid. It also does hrv and again local exports to csv.

https://play.google.com/store/apps/details?id=com.beetlesoft... [2]

Chest strap = EKG, wrist! = ekg as it's optical

*excluding the apple watch, however chest/electrodes will always be accurate than wrist/optical sensory not to mention battery life.

[1] Details SUPPORTED: Verity Sense: - HR, PPi, Accelerometer, Gyro, Magnetometer and PPG

OH1: wrist - HR, PPi, Accelerometer and PPG

H10: Chest - HR, RR, ECG and Accelerometer

H7/H9:CHEST - HR and RR

Application also supports sensor data forwarding using MQTT-protocol.

[3] https://www.dcrainmaker.com/2021/12/polar-h10-heart-rate-mon...

-- I did leave off garmin products intentionally. They don't share ekg and at that price range I would suggest a Polar H10. Polar has a an sdk available for their products.

I found one but study but there's plenty to validate that it is indeed medical grade if you search. I know many sports fitness studies that have used polar H10.

You need bluetooth for ekg data. Ant+ only does hrv (with app), RR and basic HR data.

If anyone has a windows PC app that logs and shows live EKG data for the polar H10, I would love to your suggestions. I have Golden cheetah and have experimented with android emulators but most seem to lack a proper bluetooth stack.

I've never seen an HRM strap attempt EKG data, at best they only track HRV which would be a single dimension instead of the two needed, basically fine resolution of timing between beats but not intensity/duration of beats.

But just to add to the knowledgebank, Wahoo makes probably the most interesting HRM strap right now that can not only store the HRV data for days but unlike Garmin straps the tickerX strap can record without any phone or watch starting or connected to it. If someone could hack it to also do 2-lead EKG for days that would be fascinating and probably very useful.

Then there is the Firstbeat Bodyguard single node device which is too expensive to play with but fascinating: https://shop.firstbeatsports.us/collections/bodyguard-3-sens...

> The Exporter software provides access to raw data files for ECG, RR-I and ACC in CSV format

The fact firstbeat can do it virtually proves HRM straps could do EKG, the makers just don't want to support it and there are probably FDA complications with certifying that.

Reply on news.ycombinator.com