Wearing A Homemade EKG Whilst Base Jumping!


[Andrew Wilson] is a pretty extreme guy. He base jumps for fun, and is also a hacker. And while you can try to explain the awesome adrenaline rush that comes with this kind of extreme hobby, it’d be nice if you could show it off, you know, quantitatively. So, he decided to make his own EKG, pair it with his GoPro, and go for a jump!

An EKG is an electrocardiogram — a fancy term for a heart rate monitor — and [Andrew’s] has built his own using a small instrument amplifier circuit. This circuit amplifies the differential signal put out by your heart. The data are fed through an ADC on an Arduino Uno, and then saved to a SD card. He also added a piezo buzzer to try to help sync the data to the video — unfortunately it was too quiet for the GoPro to pick up. So for now he’s stuck with pressing record and start on his EKG at the same time.

Once he was satisfied with a few safe tests, he decided to take it for a base jump. For our viewing pleasure, he’s taken the data collected from the EKG and post-processed it into a nice scrolling graph overlay for the video.

We guarantee your hands will get sweaty as his heart rate goes up as he prepares to make the plunge.

36 thoughts on “Wearing A Homemade EKG Whilst Base Jumping!

  1. We made an EKG as a project in my Electrical Engineering circuits lab. There’s a lot of noise in his signal, after he jumps, no heartbeats a recognizable and the signal jumps around everywhere. Right before he jumps, however, is very very cool.

    1. ALEX says: “…of what use is this data?”

      Back off man… it’s ART!

      Just kidding of course, but as it says in the write-up, Andrew wanted a way to quantify his actual experience and an EKG visual overlay on a video is one way to do so.

      He accomplished his task quite well from what I see, and that is the use of the data.

      And reading between the lines, it seems from the esoteric, to me, prelude to your rhetorical question, you feel that because it doesn’t have a scientific or more professional bent that it’s not worthy of some imaginary threshold of credibility, so perhaps you might follow up with something of more substance that I/we can understand?

  2. While the EKG trace is cool and all I think the heart rate show the “Holy S**t” factor just as well, as Mister X said an Oximeter might get cleaner data (built into a glove or shoe)? But as for me I wouldn’t need fancy sensors to show my reaction to a jump like that, just a moisture monitor in my underware, this guy has balls!

  3. When the trace is clean it looks like you have pretty consistent S-T segment elevation(Read: Bad). Though, this could just be from low voltage and having the trace amplified. As stated in another post lead II is a pretty common view, and it seems to work the well with movement. Lead II gives you the best picture of your left ventricle, the part of the heart that is the strongest/most muscle.

    1. Two things to add, which I tell all my students (and get a little frustrated when I get calls in the middle do the night)…

      Firstly, You can’t make a diagnosis off a rhythm strip, especially about ST segments. You need a diagnostic 12 lead to make any judgement. For the academic exercise though, I’d interpret the QRS morphology as suggestive of a right bundle branch pattern, though probably not fulfilling the criteria for a block. This would make ST segments uninterpretable anyway, explaining why it has a biphasic appearance, more so suggestive of early repolarisation. This could be a totally normal variant – my ECG had a somewhat similar appearance when I was in my late teens/early 20s. There is an R wave, followed by a larger S Wave, and the T wave is predominantly upright (as is the P wave). I think he probably has the leads around the correct way, and has a “plumb line heart” with some physiological right axis deviation. Again, need a proper 12 lead without pectoralis interference to make the diagnosis. You could capture this the old-school way, one lead at a time using your current setup at rest!

      Secondly, I’m not sure of the time scale. Andrew, you may have had the trace at 25mm/sec (for what it’s worth when the screen size is scalable). But this plus the interference makes it even harder to make medical sense of.

      In any case, I realise that your intention was to make a cool project, which you succeeded at without a doubt in this humble author’s mind! Well done! People will try and interpret things off an interference-laden rhythm strip, but that should not detract from the issue that I think you succeeded.

      My two cents… Maybe try using your RL lead to generate lead II (an *inferior* lead! for the record, but often easier to assess PR morphology, and will make your R wave more positive to satisfy the armchair critics. Also, maybe try placing electrodes over bony prominences (ie acromioclavicular joint(s) and shins/anterior superior iliac spines). This may improve your interference – it sometimes works.

      Thanks for a great project, and looking forward to the updates!

      1. Drj,

        Thanks for your response!

        I agree, in the video it is nearly impossible to decipher the timescale. I have an alternate output that is closer to conventional EKG sheets here: http://derefnull.org/ekg/dataset.031.pdf . The larger boxes are 0.2 sec, while the smaller boxes are 0.04 sec. The scale is very close to the 25mm/sec feed rate you mention (25.4mm/sec in this case).

        Also, in the PDF I inverted the signal from what was used in the video overlay. This gives a more conventional looking waveform (as per the image I found on wikipedia), though I admit I have no experience reading a medical EKG.

        I will look into attaching electrodes to my bony prominences — that is an interesting idea!


        1. Nicely done. The timescale makes things a little easier to interpret :-).

          I learned the trick of reducing interference with bony prominences as a university student when I worked on the weekends for a pathology company at a private hospital. Between taking blood samples, I would take ECGs of inpatients. One of the technicians taught me that trick, and it often seems to make a difference.

          I think my previous comments still stand on the interpretation of this dataset – partial right bundle branch pattern, with high ST take-off, all of which in a non-12 lead ECG on a rhythm strip would be considered a normal variation in someone fit enough to go base jumping! Interesting to note there is a rate-related change in the S-wave of the QRS complex towards the end of the dataset, when your heart rate increases to around/just over the 150 mark – the S-wave becomes deeper and a little wider, supporting the previous theory about right bundle pattern. You often see rate-related changes in conduction intervals.

          Interesting physiology, (call it a physiology “hack” if one would indulge), but likely of absolutely zero clinical significance. I still think it would be interesting to use your hardware to capture a true 12 lead ECG – probably would only take minimal modification!

          Anyhow, nice to have an interesting topic/idea on this website that I can make a meaningful contribution to! :-) It’s easy for trolls in glass houses to hurl the pebbles from the banks of the river flowing under their bridge.

  4. Funny how his heart rate slows from 140 to about 110 the first time he looks over the edge.

    it’s unfortunate the reading is mostly useless during first part of the jump, any idea what caused it?

    The numbers are a bit hard to read, you might want to increase the contrast a little, or add some dark shadows around the numbers or something.

  5. Cool project!
    A heart is a cool thing, huh?
    As a med student I always wanted to build my own EKG as well but have never gotten around to do it. Especially the noise filtering part is what would require some investigation on my side.

    Now to the interpretation of your EKG:
    It’s not useable :-D
    It’s true that it kinda looks like a S-T elevation (which would be really bad) but in no way can you use such a noisy graph with pads not stuck to the anatomically correct places to interpret anything about the hearts muscles. So no worries!
    Plus, when we do this in the hospital the patient is asked to not move at all. Even shivering or scratching your nose results in noise that makes the readout basically unusable. I guess in your case it’s a bit more extreme ;-)

    But it does what it’s supposed to! I liked watching the video.

    Greetings Shah

  6. “He base jumps for fun, and is also a hacker.”

    OK, then WHERE’s the “hack”???

    He made an EKG -> Has been done a million times before, this was a lab project in my EE lab a looooong time ago.

    He made the EKG beep -> Big deal

    He put the project near the GoPro so it could hear the beep -> Big deal

    Where is the novel idea here? Where is the “hack”??? Does the target audience of this site lack so much creativity that recording a beep from a common project on video is called a “hack”???? He didn’t even write his own routines to access the SD card. This is the equivalent of a 4 year old playing with lego blocks that someone else made. I’m not bashing him – Good for you kid, if you want to learn about electronics that is absolutely awesome. I’m bashing HaD for making the idiotic statement that this guy is some sort of “hacker” – That is a very stupid statement.

    1. I’d say the term “hacker” qualifies. Basic definition: a person who uses an unconventional or unintended solution to a technological problem. His problem was: how do you show a more objective measurement of how exciting a base jump is? His answer, hack an EKG together with a camera to show the physiological response at various points of the jump.

      Is it an “elite” sort of hack? Perhaps not. We can’t all be MIT valedictorians. But it does meet the basic definition.

      I’ve said it once, and I’ll say it again, if you’re not going to meaningfully contribute (submit your own hack, show us how it’s done!), the maker movement doesn’t need or want you.

        1. You know, if you could take a bowel movement (which is a necessary part of life) and make it something more useful, that would be pretty good for developing countries. Green tech is a growth industry these days.

          I love how you didn’t bring a single real point up for discussion. I’m guessing you don’t have anything redeeming to say for yourself. Enjoy your perfectionist utopia, where nothing is as good as it should be, if only *you* were the one doing it.

          Those of us makers who are actually getting things done (as opposed to the bureaucrats and managers who like to say “make it so” and take credit for other peoples’ work) might prefer naysayers to find their own forum to bring down, though.

    2. You people are annoying as all hell, get over yourselves or find another website similar to Hackaday, except with “real” (in your opinion) hacks… Just because it’s been done before doesn’t make it any less of a hack… Seriously, grow up!

      1. You don’t get it. This guy doesn’t say the word “hack” on his writeup even once because he KNOWS it isn’t a “hack”. Did he make the GoPro behave as an EKG? No. Did he make an EKG behave as a GoPro? No. Did he modify the GoPro to record EKG data? No. He had two separate systems, and the GoPro recorded beeps so he could later syncrhonize his EKG data to the video. He has no real “hacks” on his website and makes absolutely no claim to be a hacker. This is factual information gleaned from his website.

        It would be fine to have written this up as something cool that someone did (which it is). It is idiotic to write it up as a “hack”. I am not bashing the creator of this project – I am bashing the clueless moron who called him a “hacker” in the first place.

  7. Felix Baumgartner had a monitor for his record-setting jump, and when he went into a spin, you can see that either there was a hole in the data, or that dude’s heart flat out stopped for about a half a minute.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.