Implant Stimulation? Extended hacks?

I'm sure most of you guys have read about the bottlenose project, and possibly read the paper. I see the rough design of the coil they use to stimulate the magnet in the picture, but was wondering if anyone has any specific experience with extending that or using a similar setup.

Also, has anyone tried anything like setting up an open cube of magnetometers to get a 3D coordinate of your finger in space, such that you could use it as a PC control surface or something of the sort?

Generally, what kind of hacks have you guys attempted with your magnets?

Comments

  • edited May 2014
    @mothball,

    As far as the coil goes, what are you looking to do? The easiest way I've found to get started with that is to buy one of these: http://www.ebay.com/itm/271288743429

    Your magnetometer idea is particularly interesting to me. I've got a bunch of old Android phones (with magnetometers!) sitting around gathering dust, and I've been thinking for a while of making a cube or something out of them for input purposes. I could totally do the software, but I'm really, really unsure on where to start as far as the math goes.

    If we got it working, though, I'm imagining that being able to interact with 3D environments could be very exciting.

    ~Drew
  • Software could be super complex with more than 3 magnetometers, but with that small number you could detrct x,y, and z axis (all the same axes realities to the individual magnetometers) by relative falloff of magnetic field if the magnetometers are all at right angles to each other in a sort of pyramid formation. You could read those by serial or some other method and then derive a 3D coordinate to do with as you please with software emulation.
  • Also, I'm looking for a much smaller inductor - finger diameter or so, such that I can make a 'socket' of sorts out of low temp shape lock polymer or maybe even a finger from a glove. I would assume I can just make a small inductor coil and hook it up to my arduino and play with frequencies and power. I was just wondering if anyone had a baseline of a good starting point other than the 200ish Hz mentioned in the paper.
  • edited May 2014
    http://www.melexis.com/mlx90333
    ^Hall effect sensor designed to interpret XYZ coordinates of magnets.

    I've been thinking about getting some kind of hall effect sensor for a while now, but I'm really not sure what strength my magnets are vs what strength these can pick up...
  • i too am curious if we can develop an interface device for pc specifically for pc. I would be highly interested in this. 
    perhaps something along the lines of a waccom tablet for design but instead for magnets not for stylus. theoretically if you could get surface of panels that detect magnetic field across it in differing strengths you could have a pretty high fidelity interface. this would not be unlike what they are doing with 3d interfaces and ultrasound.
    anyone know how expensive it would be to get started doing that

  • Well I just ordered a single magnetometer from Adafruit for about $15 (note I haven't tested or even recieved this in the mail yet, so order at your own risk/YMMV).



    I'm going to set it up with my arduino and see what the resolution, jitter, etc. is. I'll of course keep a thread going on here. If I determine that range and signal strength is practical, I'll go talk to my EM physics professor about what the most sensible setup for 3D sensing would be (open cube of multiple sensors ($$$), three sensors at right angles, etc).



    To get started from nothing, though, an arduino, breadboard, few jumpers can be gotten along with servos, resistors, etc in a kit for about $100 (Take a look here). Programming knowledge is a big plus, but not vital, for learning an Arduino - you'll definitely have an easier time handling more complex tasks like converting serial data from the arduino to a usable HID device on the computer, as well as a bit of math for triangulation. So, from square one, an Arduino kit + magnetometer would be about $100.



    As I get working prototypes and develop things, I'll keep everything open source and chuck up code, pinouts, etc. as I go along so it should be decently simple for someone with comparatively basic coding knowledge to plug'n'play with an Arduino and a parts list. The only real variable between grinders that might be implementing things is magnetic field power which could affect how practical certain sensors are, so I'll do my best to do some math to figure out my field strength and include that with code. I'm a few weeks away from finals (schedule gets tight), though, and even one week away from my actual implant date, but I'll be setting things up prior to my implant and able to work with magnetometers prior to total healing, albeit with slightly less resolution (through bandages).



    One of my other long term goals, as I mentioned, is to make a socket that I can put my fingertip into that has coils, a magnetometer, and a small amount of wiggle room for my most distal joint so I can move it slightly up and down in the socket, hopefully enough for the magnetometer to discriminate (dipole strength falls off as a function of 1/radius^3 so I'm not super concerned). Hopefully, I can use the coils for output and the magnetometer for output, turning off the sensor when the coil is pulsed, so I can have a coherent and self contained IO device - that's a way's down the road though.



    $10 says my implant rejects and I have to wait like a month to try any of this - that would majorly suck.
  • edited May 2014
    @mothball I'm interested in seeing how that works, I don't know what the code looks like but make sure your GN bits are set to 111 for max. sensitivity and range when you're testing. The math makes it look like it will take about 4000 different measurements between 1/2" to 4.5" distance if you have an N52 neodymium that hasn't been overheated for coating. The math is rarely near correct in reality though so definitely let us know how it goes!

    From LSM303DLHC datasheet @ GN = 111:
    -range +/- 8 gauss
    -magnetic gain = 205-230 least significant bit / gauss
    -12 bit output per axis

    https://www.kjmagnetics.com/calculator.asp used to gauss @ distance calcs.
  • No need to rush. Give your body time to heal before you start taking reading and messing around to much. It's better in the long run.

    One word of advice, making things that cross the skin barrier (like a socket) is really difficult. The body is a sealed system. Cross skin implants have a hugely high failure rate and can lead to some really nasty infections. I know you said it's long term, i'm just giving you the heads up now. 
  • Oh yeah - when I say socket, I don't mean as a cross skin implant. Just a tube/fingertip I stick my finger into that would have an inductor and sensor that works with the magnet.
Sign In or Register to comment.