Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Biomed pong game interface.

  1. Dec 5, 2006 #1
    Hey guys how's it going this is my first post so take it easy on me :wink: . I'm in my last year of electrical engineering technology, and we need to design and construct a final project. On the list of possible topics was the idea of a bio-med game interface. It would be biomed probes detecting muscle activity to control a video game. Rather than getting an open source game, rewriting the code, building a cable to connect it to the computer, and etc I found an old think geek DIY pong game.

    http://www.thinkgeek.com/geektoys/science/8546/
    (avaible via partsexpress.com)

    So I want to replace the buttons with connector probes attached to circuitry. I figured I would have the biomed sensors on the arms going through a high gain op-amp, and that through a ADC. Whichever output became high when then arm was flexed I would tie to a relay or something to close the switch. Simulating someone pushing the buttons.

    I'm great with the amplificiation of volages, transitors, and such. However if anyone can any point me to where I would purchase biomed probes and any documentation you might know of it would be much appreciated. I know nothing about biomedical stuff. How an arm would react when flexing versus not flexing and reading signals from it. Any input would be much appreciated. Googling has gotten me nowheres and I'm really lloking forward to getting started on the project.
     
    Last edited: Dec 5, 2006
  2. jcsd
  3. Dec 5, 2006 #2

    Danger

    User Avatar
    Gold Member

    My next Hallowe'en costume (if I can get it done in time) incorporates myoelectric controls, so I've checked into it a bit. My first stop was a local prosthetics distributor. Unfortunately, the surface electrodes in use these days are made of gold and cost over $700 Canadian per.
    I have an idea that I am looking into, but haven't had time to get very far. I'm thinking of using metal acupuncture needles to infiltrate the flesh and rest up beside the muscles. Nothing approaching the sensitivity of a transcutaneous pick-up should be necessary because of the proximity to the source. On the other hand, I can foresee it being somewhat uncomfortable, and I'll possibly have to wrap my arms in anti-biotic bandages. :rolleyes:
     
  4. Dec 5, 2006 #3

    berkeman

    User Avatar

    Staff: Mentor

    I googled medical supply electrode +emg, and got a bzillion good hits.

    I'll tell you from personal experience, though, electrodes (which you can get cheap disposable versions of, Danger) are pretty messy and uncomfortable to put on and wear. They would certainly put me off of playing your game. Plus, there are some pretty strict medical safety standards that you would have to design to (I think UL510? here in the US) to reduce your liablity risk. Whenever you hook electrodes to the body, you are somewhat bypassing the normal insulation level that your skin provides, so shock risks go up significantly. Danger would probably get a kick out of the experiment I did with a 9V batter and a couple EEG electrodes may years ago -- just about shot myself across the room.....

    So I'd suggest that you re-think your sensor approach to use some other angle. How about some optical pickup or exercise device (play and work out at the same time)?
     
  5. Dec 5, 2006 #4

    dlgoff

    User Avatar
    Science Advisor
    Gold Member

    Berkeman,

    I have to disagree a little with you on not hooking up electrodes to your body. Sure you need to make sure that there will be no leakage currents involved with your circuit; but using a good instrument op-amp with a medical grade power supply should be okay. The OP should check the UL510? requirements however to understand how to achieve and test for low leakage currents. e.g. hospital grade power cables and plugs.

    BTW How did you get jolted with a 9 volt battery? You must have been switching it through some big coils or something?

    Regards
     
  6. Dec 5, 2006 #5

    Danger

    User Avatar
    Gold Member

    Berkeman, I'd certainly appreciate anything that you can steer me toward in the line of disposable myoelectric surface pick-ups.
    Now, as much as I appreciate that you're one of the most 'electric savvy' guys on PF, I have to agree with digoff about the feedback thing. I have absolutely never heard of anyone getting jolted by a prosthetic limb. You must have screwed up your connections somehow. At the very least, you must have been running your 9 volts through a frequency generator of some sort, which might have resulted in a Taser-like effect if that frequency interfered with neurological impulses. The main thing to remember in this context is that the electrodes are a passive sensor system. No power is applied to the electrodes. They provide the power to fire a Schmitt trigger.
     
    Last edited: Dec 5, 2006
  7. Dec 6, 2006 #6

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

  8. Dec 6, 2006 #7

    Danger

    User Avatar
    Gold Member

    The thing here is that those electrodes are EKG units. They wouldn't be discriminatory enough for prosthetic control, but maybe for this application.
     
  9. Dec 6, 2006 #8

    berkeman

    User Avatar

    Staff: Mentor

    Nope, just a straight 9V. Okay, nobody try this at home, okay? I was designing and prototyping a biofeedback EEG device back in my consulting days (it turned out pretty cool in the end), and I obviously used myself for a lot of the EEG testing. So I had EEG electrodes on my head pretty often. One day I was working on some part of the circuit (I don't remember what), and I ended up putting a small voltage across a couple of the electrodes on my forehead. It caused a little tingle and twitch which surprised me.

    So I uttered a line from the movie Real Genius, "Up the voltage!" I pulled out a 9V battery and touched the electrode lead ends to the buttons, and WHAM!! That shot me backwards out of my chair (and luckily made me let go of the battery) and gave me a hell of a shock to my head. :eek:

    So I learned my lesson about how electrodes drop the skin resistance a lot. I mean, it's common to use your tounge to battery test 9V batteries, right? And all that does is give a little tingle for fully charged batteries. And think how wet and low-Z your tongue is compared to your dry skin. :tongue2:
     
  10. Dec 6, 2006 #9

    Danger

    User Avatar
    Gold Member

    That doesn't sound right, Berkeman.
    I'm not denying that it happened, but I have an alternate explanation that seems to make more sense. I'm thinking that the placement of the electrodes was such that you accidentally triggered neuronal firing in your motor cortex. We'd need somebody like Moonbear to confirm or deny the possibility of that, though; I don't know much biology.
     
  11. Dec 6, 2006 #10

    berkeman

    User Avatar

    Staff: Mentor

    Oh it triggerd some firing all right! Yeah, I wasn't saying the forehead electrodes directly coupled to any major muscles. That was my brain complaining!
     
  12. Dec 7, 2006 #11

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

    Don't know about that.
    As I understand it, muscles generate small electric currents when they move.
    Most muscles are fairly large compared to the size of the electrode.
    So I might think there should be fair discrimination.
    The operating principal of these electrodes seems to only be to reduce skin resistance.
    Proper placement would be important.

    The only other posibility I can think of is nerve implant electrodes.
    That would require surgery.
     
  13. Dec 7, 2006 #12

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

    Electoshock therapy might be a relavent topic.
     
  14. Dec 7, 2006 #13

    Danger

    User Avatar
    Gold Member

    You're both right and wrong there, NoTime. Transcutaneous electrodes can't do anything about reducing skin resistance; nothing can. The point is to isolate the impulses produced by a particular muscle. You're right that placement is important, but so is 'focus' (which is what I meant by 'descrimination'). Do you realize how close together the muscles are that control your index finger and your bird finger or your thumb, never mind the individual joints of each? A disposable EKG electrode wouldn't be able to tell the difference.
    And neurological implants today go beyond simple surgery; it's almost a mating process. The original nerve tissue is prompted to grow around the coiled platinum (or similar) electrode in order to create a full fusion of the two. It's still no easy task to translate an electrochemical impulse into a strictly electronic one.
     
  15. Dec 7, 2006 #14

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

    The electrodes do decrease skin resistance, by a few orders of magnitude.
    I never measured what a good electrode can obtain, but just wetting your fingers and holding the ohm meter probes will show that.
    The medical ones have salts that increase the effect.

    The fine control muscles are close together, stacked even.
    Don't recall anyone trying to read those, just the major muscles.
    But, I've never researched ths, just miscellaneous sutuff I'be read.
     
  16. Dec 7, 2006 #15

    Danger

    User Avatar
    Gold Member

    I see where you're going about the resistance, but that's the paste at work, not the electrodes themselves, and it's a surface effect. It's more about ensuring continuity between the skin and the sensor, rather than making the skin more conductive.
    The range of motion of new designs in hands requires very sophisticated control systems, and my costume needs flex and extend pickups on all ten fingers. (That's muscular intent, by the way; my fingers won't actually be able to move, so switches won't work.)
     
  17. Dec 7, 2006 #16

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

    Not entirely, the skin surface is composed of dead dry cells.
    The skin is permeable, quite a problem with some chemicals.
    Plain water and body salts or the special paste bypass the dry layer insulation properties and provide electrical connection to the body interior.

    Since your fingers and arms still can exert force even when constrained, have you considered pressure transducers for your costume control.
    Sounds simpler to me.

    Or perhaps smearing your arm with goop and wrapping it with a contact sensor array, using a computer with pattern recognition to determine muscle activity.
    Unfortunately, pattern recognition is still a very immature technology.
    Gold would only be important for long term corrosion in the contact array.
     
  18. Dec 7, 2006 #17

    Danger

    User Avatar
    Gold Member

    Okay, I'll concede the point about the conductance. I've never heard of that as an issue, but it makes sense. As for the use of gold, I never bothered asking why they're made out of it; I quit the conversation when I heard the price... and that's the only kind they sell.
    As for transducers, I don't think that they'd work. There's a possibility, so I'll do some figuring. Thanks for the idea. The reason that switches (or overly sensitive transducers) won't work is because both hands will have a firm grip on something at all times. If I can find transducers with the proper trigger pressure, though, I might get by with them. They'd have to be double-acting, with up to about 0-10 lbs. of pressure unreactive, 10-20 lbs. giving a positive output, and 20+ lbs. giving a negative one (to make up for having only flex rather than flex/extend). Learing to operate it that way will be tough, but it's going to be anyhow.
     
  19. Dec 7, 2006 #18

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

    How complex is your control requirement.
    Could you for instance, use one or both little fingers to switch command sequences (along the line of morse code)

    How about your toes. Apparently some people have enough toe control to tie knots.

    Just throwing out ideas here.
     
  20. Dec 8, 2006 #19

    Danger

    User Avatar
    Gold Member

    My feet are going to be somewhat tied up as well, so no go there. (And I have barely enough toe control to walk. :biggrin: )
    As for complexity, the outfit has 10 major bi-directional servo control channels, plus 2 momentary contact solenoids. (Hmmm... just got a thought there; I can put the solenoids on a timer. There's 2 switches gone.)
     
  21. Dec 9, 2006 #20

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

    Are you building an exoskeleton?
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Biomed pong game interface.
  1. PIC to PC interface (Replies: 0)

Loading...