Hackaday Prize Semifinalist: Low Cost Radiography

For the past year, [Adam] has been working full-time on developing a low-cost x-ray system for developing nations. He has more than 3,500 hours into the project. A few months ago, we announced the 2015 Hackaday Prize, with a theme of, ‘build something that matters.’ A low-cost x-ray would certainly matter to the two-thirds of the world’s population that does not have access to medical radiography, making this project a great entry for The Hackaday Prize.

[Adam]’s portable x-ray system consists of an x-ray tube encased in an epoxied, 3D printed enclosure filled with dialectric oil. This tube is tucked away inside a beautiful case with just a single 12VDC input and an easy to understand user manual. This is just very high voltages and x-rays, nothing [Adam] hasn’t handled (safely) before. The real trick is in the imaging, and for this, [Adam] is using a phosphor screen to turn that x-ray exposure into something visible, an off the shelf x-ray sensor, and a prism to adapt the sensor to the phosphor screen.

The results are incredible. After taking a few pictures of what he had on hand, [Adam] can see the bond wires inside the microprocessor of a calculator. That’s more than sufficient for medical imaging – the goal of the project – and cheap enough to send it to the far-flung reaches of the planet.

The 2015 Hackaday Prize is sponsored by:

37 thoughts on “Hackaday Prize Semifinalist: Low Cost Radiography

  1. The actual x-ray sensor seems to be on the order of $4000 USD. This may be much cheaper than all other alternatives, but it is still too expensive for widespead hobby use. Anyone have ideas for a lower cost sensor alternative?

    1. Holy crap, that is awesome!

      I don’t have time at the moment, but after the contest I think I’ll offer to join and help his project.

      Here’s a random quip from his build logs:

      “When I came back, I was greeted with a smashed automobile, and glass scattered about the parking lot. The box was gone, my computer was gone. Even the flash drive in the center console.
      I lost all my work. Everything.”

    2. If you do not care about dose that studied object would receive, than any sdram chip should fit.
      Without refresh, data bits would have a lifetime inversely proportionally to the dose it receive, so periodically filing it with 0xFF (or 0x00) and reading it out, one could get a xray sensor with few mm size and micron resolution.
      Lifetime of such sensor would be another issue.
      Also any cmos image sensor with attached fluorescent screen would work, but you need to put something like leadglass in between in order to shield the sensor from xray, and that would spoil the resolution.
      But anyway fluorescent screen, mirror (to get the camera away from xray) and any camera would work.

    3. Possibly. I’ve only just skimmed his project, but it occurs to me that Zinc sulfide spin coated onto a glass substrate would probably glow under X-rays, and be visible to a normal CCD camera.

      This is only off the top of my head, I would have to do a bunch of research (read a whole lot of research papers), see what energy photons he uses, and so on.

      But I’ll bet that something like that, perhaps with a substrate different than Zinc sulfide, would work.

      As a quick test, I wonder if an EL panel would show the image under X-rays?

      (And then the math part of my brain kicked in and thought about gaining higher resolution by using multiple exposures and the synthetic aperture stuff that Peter Jansen is working on. Hmmmmm… https://hackaday.io/project/5030-low-field-mri)

      1. Of course while cheap (to purchase), film has the annoying property of requiring temperature-stabilized chemical development which takes time to process (about 5-20 minutes for any film I’ve developed) and consumables (chemicals and film),

        While you can develop film without temperature stabilization, not having consistency wreaks both the contrast and brightness of the exposure, making it very crude for any diagnostic purposes.

        So it becomes a business / engineering trade-off, high purchase price (solid state x-ray) versus more complex and time-consuming process (film, lab to develop said film). I suspect that the higher one-off cost of a solid state X-ray machine would be more affordable for emergencies and 3rd-world countries.

    4. If you read his build log, this is one of his goals as well. He’s on his second design, using nothing more than a goofy looking prism he made himself, and and phosphor screen and webcam. He’s got a pretty good looking test image up from it as well, and that is very cheap. See his entry here:
      https://hackaday.io/project/7225-low-cost-x-ray-systems-for-developing-nations/log/22932-april-may-2015-designing-a-low-cost-x-ray-sensor-second-attempt

  2. I have two *childrens* books from the 1930s. One tells you how to build your own xray machine, after having purchased the tube. The other describes fun things to do with it, but warns that you shouldn’t use it so much that your skin reddens.

    The fluorescent detector didn’t cost anything near $4k!

  3. Just wondering, can a tube from a CRT screen be used to make a x-ray tube? (I’m guessing with a high enough voltage it might emit the rays) I’d love to see someone design medical equipment out of old junk that can be found in poor countries/

    1. If i recall correctly they had either a lead coating or were made with lead glass to prevent this kind of thing from accidentally occurring. old vacuum rectifier tubes work though.

  4. I don’t know much about using fluorescent screens to turn xrays into visible wavelength photons, but it occurs to me that you could use a much hinkier and cheaper kind of screen if you put an image intensifier in the loop. I bought an image intensifier scope for less than 100 pounds sterling a while ago. II tubes are noisy, but in this application you could signal average to a cleaner output.

    1. The description has been mangled a bit. The expensive direct X-ray sensor, 4k US is replaced by a phosphor screen, a webcam and something to keep the CCD out of the line of the X-rays right?

  5. An incredible project, and having read the guys project logs, he has an incredible story in there about his prototype being stolen, and it randomly being returned to him through the oddest of circumstances!

    Worth a read.

  6. It’s actually the cost of a xray sensors that makes DIY xray images so difficult.

    Xray tube on ebay is like 50-100$. A Cockcroft-Walton generator powered from a DIY flyback isn’t that expensive either and easily gives you 50-120kV @ a few mA DC. But these damn sensors are expensive as hell.

    Sure you can also try long time exposure of a fluorescent screen with DSLR, but this won’t give you high quality images, since it’s distorted and limited by camera optics. That setup is also difficult to shield and I don’t want to die from cancer later on because I played with Xrays as a kid. Flatbed scanning the fluorescent screen (disable scanner light) could also work, but might require really strong xray source to get bright fluorescence, since all scanner I tested had poor light sensitivity.

    BTW, that kid has no idea of xray backscatter and probably doesn’t even own a geiger counter for basic xray measurement, because if he would, he wouldn’t be that careless with xrays. The table (mostly light elements) prolly gives him bad xray backscatter and he stands next to it. And it’s the relatively low energy backscatter that gets mostly absorbed in your body (high energy xrays go right trough it, with much less damage) and gives you cancer on the long run. So build yourself a concrete or lead box if you plan to operate a xray machine on the long run.

    1. I’m glad to see other comments about how fantastically dangerous this is on here. Coming from someone who has a PhD in science and did a little medical imaging in graduate school (though I’m no means an expert), this is the most dangerous, reckless, and (in most places) illegal project I’ve seen in a long time. The comments above remind me of the dangerous excitement after x-rays were discovered in the 1890s — everyone was building their own machines and x-raying all sorts of things without care to dose or shielding, and a great many of them became very ill or died. Yes, x-ray pictures look incredible, and yes, seeing inside of things is one of the most fascinating things you can do to study it. But this is genuinely incredibly dangerous, the danger of doing something like this outside of the proper facilities (like a lead room), without validated and calibrated safe equipment, or a really thorough knowledge of radiation safety (and a healthy fear of death) really can’t be overestimated.

      Inexpensive medical technology is clearly a really noble goal, but building it safely (for yourself and others) is paramount, and publishing the schematics for what could essentially become an x-ray death ray is really very questionable to me. If someone were to stick this in their knapsack, and sit down in any public place (like a coffee shop), how long would it take to give the person beside them a completely silent and invisible lethal dose of radiation? Seconds? Minutes? And who would ever know? They literally convicted someone of /attempting/ to build such a device for this purpose to kill Muslims only days ago ( http://bigstory.ap.org/article/6476b3fec9734230a278b3712f164f41/jury-gets-case-man-accused-plot-build-x-ray-weapon ). To say nothing of the well-intentioned makers building something like this and irradiating the people in their house, or their next door neighbours. Some extremely dangerous knowledge and pieces of technology have well-crafted barriers to entry (like, qualifications) for public safety.

      1. The project is intended to reduce the cost of the digital imaging system. X-ray devices have been public knowledge for far longer than I have been alive, and are easily constructed by someone of moderate skill. I remember reading a very old Scientific American “Amateur Scientist” article many years ago about making your own X-ray source at home.
        http://www.sciencemadness.org/library/books/projects_for_the_amateur_scientist.pdf

        The manufacture of a source, for personal use or otherwise, even from readily available materials, requires licensing by the NRC (if in the US). So the creator of the project would be wise to contact their licensing branch (they may be in an agreement state) for licensing information- Or, they may just show up at his doorstep.

        An interesting possibility would be the creator of the project to do some gamma radiography using sealed check sources. They are legal to posses in small quantities in the US, and the radioisotope can be selected to obtain different gamma energies. The additional challenge of obtaining images with a low activity source presents quite an engineering challenge.

  7. Just telling, because you were apparently standing next to the table while xraying things and because that DIY xray box doesn’t really look safe to me, as it is unshielded and has no apparent lock preventing it from being operated inadvertently.

    I operated a DIY experimental xray setup for a while, before I decided that I will only continue, if I have a shielded box. The backscatter from the wooden wall (were the xray tube was directed at) was filling the room with xrays (in the 10’s of milliroentgens/h range, measured with old soviet DP66M, probably higher, since a geiger counter only measures the higher energy parts of xray spectrum).

  8. One should get smaller, more directed, and more efficient in detecting.

    Smaller beam aperture: look at the work of building an x-ray emitter from carbon nanotube coated metal surfaces by
    University of Northern Carolina: https://www.med.unc.edu/radonc/research/nano-papers
    Direction and slit beam forming could be done by x-ray lenses that are crafted by deep ion etching from silicon processes in fabs.

    I do not see why some ccd line would be inefficient or the optics in a flat bed scanner.
    Perhaps one could etch away the color coating on the ccd of a color flat bed scanner.
    Or use a monochrome one right from the start.

    Schott has various geometries of light conducting and x-ray absorbing plates and stripes that consist of stacked light conducting fibres, similar to endoscopic imaging transferring light conducting fiber (arrays), only shorter and thicker ;)
    http://www.schott.com/magazine/german/sol213/sol213_10_faceplates.html

    A good place to do things like that is at universities, or researching hospitals or x-ray treatment centers that have x-ray labs, and the proper infrastructure like dosimeters for every one running around there ;)

    Stay healthy ;)

    Andi

Leave a Reply to seanrhinehart Cancel reply

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