CT scanning and x-rays

[Anonymous No.2541144]

One of my clients is looking at spending about a million dollars on a CT scanning machine, with a lead time of about a year. In the interest of saving some time and money, I've been thinking about building a DIY machine; even if we end up ordering the "real" machine, it would be useful to have something usable - if somewhat inferior - several months prior. I'm a software/robotics engineer and it seems like a pretty simple system (aside from the xrays). The hardware (turntable + vertical actuator) is trivial and the software/algorithm used to compute slices (inverse radon transform) is available online for free. However, the x-ray element is intimidating; I'm not even sure if I can get one as a pleb with no special licenses.

I should mention - this is being used for research (non-medical) purposes, so the requirements are somewhat relaxed. Ideally, we're looking to scan specimens anywhere from 50x50x50mm to 50x50x50cm, but we'll take what we can get.

I have three questions:
1) Is this a really stupid idea? What sort of enclosure would this device need to be safe?
2) From what I can tell, I'll need some special phosphor plate(s) for the xrays to excite, which a camera then takes photos of. It's unclear to me if these plates are one-time use, reusable for a limitation duration, or reusable indefinitely. I suspect it's the second one (reusable for only for N exposures).
3) Where does one find a suitable xray emitter (and phosphor plates, see 2)? I had a brief look on ebay and I see some dental xrays there. Would those be usable?

[Anonymous No.2541156]

Follow-up question: I've noticed that dental xrays like pic related appear to operate without any backing phosphor plate; presumably there is some image sensor (or equivalent assembly) in the emitter and it's measuring particles that bounce back rather than particles that go through the subject. Is this accurate? Would this be suitable for a DIY CT scanner, provided the subjects being scanned are similar scale to a human jaw (what a dental xray is designed for)? If so, how does one obtain the actual image captured by a dental xray?

[Anonymous No.2541157]

>>2541144
>1) Is this a really stupid idea

yes, you would probably kill yourself from the crazy high voltage requirements or radiation.

commercial xray use radioactive material for the anode you wount be able to get. furthermore, a basic bitch medical xray machine has insane voltage requirements in the hundreds of thousands. each room has giant ass transformers,

you dont sound like you even understand the basics of the technology.

t. former xray tech that had to learn all this shit.

[Anonymous No.2541161]

>>2541156
Actually, scratch that - I did some digging and I see that the dental xrays do use a phosphor plate/film, but it's placed in the mouth so I couldn't see it before. With that in mind, it seems like these dental xrays would be quite ideal for small subjects. Is there a "standard" type of phosphor plates (that would be suitable for a dental xrays) or are there different types and I need to find a matching plate for my emitter?

[Anonymous No.2541164]

>>2541157
>yes, you would probably kill yourself from the crazy high voltage requirements or radiation. commercial xray use radioactive material for the anode you wount be able to get

My plan is to buy all the xray equipment off the shelf; my only concern is what kind of enclosure I need to operate this safely.
Again, to be clear - I want to buy an off the shelf xray emitter, like pic in >>2541156
Does a device like that need an additional power supply, or is that what's in the wall-mounted enclosure?

[Anonymous No.2541171]

>>2541164
industrial radiography wont use "phospher" plates because they exist for the sole purpose of lowering radiation dose for the patient and lower resolution.

digital imaging technology is extremely expensive. and good luck trying to make a fucking diy ct machine. frankly an absurd notion. even doing xrays straight to film requires extensive knowledge on several toxic chemicals

you could figure out xrays but ct is so insane i dont even want to give you the time of day.

industrial radiography requires high voltage which creates high scatter radiation. a basic medical room requires lead lined walls, a s you would need a multiple if that.

[Anonymous No.2541175]

>>2541171
>industrial radiography wont use "phospher" plates
Why do you say "phospher" when every one of my posts says "phosphor" (and that's the correct spelling)?

because they exist for the sole purpose of lowering radiation dose for the patient and lower resolution.

>digital imaging technology is extremely expensive. and good luck trying to make a fucking diy ct machine. frankly an absurd notion. even doing xrays straight to film requires extensive knowledge on several toxic chemicals

This guy did it exactly how I describe and it works just fine: https://www.youtube.com/watch?v=hF3V-GHiJ78

>you could figure out xrays but ct is so insane i dont even want to give you the time of day.

CT is just xrays + controlled motion + some post-processing.

>industrial radiography requires high voltage which creates high scatter radiation. a basic medical room requires lead lined walls, a s you would need a multiple if that.

You're hallucinating a far more powerful & large-scale system than I'm discussing here. There room in the video above is hardly lead lined. Dental xrays are used in rooms that are not lead lined. You're being an unhelpful blowhard.

[Anonymous No.2541205]

>>2541175
>There room in the video above is hardly lead lined.

yeah, and the guy is probably irradiating himself like a dumbass. Dental rooms are lead lined.


>you're being an unhelpful blowhard.

yeah because you soundkind of clueless. Have fun with the cataracts and liability from making your DIY xray machine. You may want to look into umbrella insurance.

[Anonymous No.2541224]

>>2541205
>yeah, and the guy is probably irradiating himself like a dumbass. Dental rooms are lead lined.

I doubt it. He mentions backscatter in the video and that he left the garage while the machine was in operation. If you look at the other videos in this channel you will see the guy is an extremely proficient engineer and making top-tier videos for over a decade. Another recent xray video (albeit not CT): https://www.youtube.com/watch?v=j-FHbHoiwNk

>yeah because you soundkind of clueless.

so far only one of us has managed to spell "phosphor" correctly

> Have fun with the cataracts and liability from making your DIY xray machine.

It seems you can't read properly. As I've said several times, I'm buying an off the shelf xray. I just wanted advise about an enclosure. A fucking box. For the rest of the setup I can do exactly what is in these videos. Instead of helping me, you've chased one red herring after another. Thanks.

[Anonymous No.2541231]

>>2541144
If you do make one please keep us updated, this sounds cool
God speed OP
Since it's not for medical and only research I don't think it'd be a bad idea as long as you know wtf you are doing

[Anonymous No.2541236]

>>2541231

Well I know what I'm doing when it comes to everything except the x-rays, but I figured I can (literally) encapsulate that problem away by enclosing the entire assembly (xray emitter + subject + imaging plate) in a lead enclosure. As long xrays can't leak out of the enclosure, I think it can be done safely.

[Anonymous No.2541238]

>>2541224
>If you look at the other videos in this channel you will see the guy is an extremely proficient engineer

the guy seems like a super genius. maybe he isn't. maybe he is. I don't see him wearing a dosimeter.

>so far only one of us has managed to spell "phosphor" correctly
>looks like your grammatical mistake invalidates your argument. I win

Back to pleddit faggot.

>It seems you can't read properly. As I've said several times, I'm buying an off the shelf xray. I just wanted advise about an enclosure. A fucking box.

lol, maybe you can't read. existing x-ray equipment is frequently tested and everyone around the machines is constantly monitored with dosimeter badges. Do you know what ionizing radiation means? Do you know what low KVP radiation does to other atoms?

I'll admit I didn't know diy xray had the software support. the image acquisition was pretty interesting and was like delayed fluoroscopy.

Regardless if you can make this or not, taking on the liability alone makes you a dumbass. If anything goes wrong, the legal system will rake you over the fucking coals because of high tightly regulated xray imaging is. Have fun telling a judge you thought it was ok to play around with ionizing radiation without following standards of practice. you may want to see how how exposure changes with increments of seconds alone.

[Anonymous No.2541242]

>>2541238
>the guy seems like a super genius. maybe he isn't. maybe he is. I don't see him wearing a dosimeter.

He's got a geiger counter and addresses this at 5:00 in the second video I posted

>Do you know what ionizing radiation means?

Why are you asking such idiotic questions? How many times do I have to state that my intention is to build a lead enclosure to stop all ionizing radiation? If the entire unit is inside a suitable enclosure, WHAT IS THE FUCKING PROBLEM?

[Anonymous No.2541247]

>>2541242
>Why are you asking such idiotic questions?

>why do you keep being a blowhard when I don't care or know about the consequences or how radiation even works.

[Anonymous No.2541249]

>>2541247

> jerking off about vague unspecified dangers

I repeat:

How many times do I have to state that my intention is to build a lead enclosure to stop all ionizing radiation? If the entire unit is inside a suitable enclosure, WHAT IS THE FUCKING PROBLEM?

Why is it so fucking difficult for you to have a straightforward conversation? All of your concerns seem to be predicated on hallucinations. It's as though I've said I want to buy a microwave and you're going on a tirade about how dangerous magnetrons are. There's an enclosure that blocks the dangerous radiation being emitted. In this case, it's a lead enclosure instead of a faraday cage/shield.

[Anonymous No.2541254]

>>2541249
this is the last time i'm answering you. I'm being vague because i'm fishing for answers that show you know what the fuck you are doing. you don't seem to get it.

standard xray exposures are fraction of a second. you are going to make all these wank DIY controllers for something that has a small margin of error. ionizing radation shot at lead can create beta radiation which is worse than fucking photons.

smarter people than you have killed themselves from irradiating themselves.

go have fun simping for a client that will sue your ass, and they will laugh at you in the courthouse. Like imagine how dumb you would feel. lol.

[Anonymous No.2541257]

>>2541157
>commercial xray use radioactive material for the anode you wount be able to get

Using radioactive materials as x-ray sources has been a dead meme for decades. Handling and disposal is a nightmare and there's no reason to do it when modern, electronic sources do a better job, anyway. There are probably still a handful of old machines around that use a radioactive source (that story of a "mysterious pretty rock" killing multiple people in Brazil or wherever comes to mind), but, far as I'm aware, they're a rarity.

[Anonymous No.2541258]

>>2541254
>standard xray exposures are fraction of a second. you are going to make all these wank DIY controllers for something that has a small margin of error

So your concern basically amounts to skepticism that I can reliably activate a MOSFET for a controlled period of time? Do I understand correctly?
If this is the extent of the issue (erroneously long exposure inducing significant beta radiation that escapes the lead enclosure) then I'm not worried.

[Anonymous No.2541262]

>>2541257

You seem knowledgeable. Would a dental xray like >>2541156 be suitable for the system I described in OP? I realize the field of view is quite small so it would only be usable for small specimens, but that's fine as a proof of concept. It looks like there's a handset on the left side, presumably a trigger (with a coiled cord so you can do it semi-remotely). I suspect that actuating this thing remotely would be as simple as cutting open that cord and splicing in an opto to emulate a button press. The knob (presumably intensity control) can stay manual, or I could electronically control that with a digital potentiometer.

[Anonymous No.2541263]

>>2541257
>for decades.

wrong

[Anonymous No.2541266]

>>2541262
>You seem knowledgeable.

I am not. I'm good enough with electronics and dangerous equipment to know this isn't that risky, and good enough with law and bureaucracy to know that it's too risky to be worth it, I'll put it that way.

While the other guy you're arguing with is autistic, retarded, or both, he's right on one point: This is a dumb idea. Not for technical reasons, of course. You could easily...well, easily-ish, maybe...get a working x-ray machine and just use it, as-is, in your ghetto CT rig. Any mass will eventually block radiation, it's just that denser materials are a lot better at it. Lead is pretty effective for that reason, but concrete, particularly high-density concrete made specifically for blocking radiation, might end up cheaper if you have the space. Either use a phosphor and take pics like Krasnow did, or, if you're feeling spicy, there are sensors that detect x-rays directly. Mad expensive, obviously, but you already knew that.

Back to the point: You should not do this. The amount of liability involved for anything involving ionizing radiation in general is nuts. Even if what you build is perfectly safe in terms of construction and loaded with multiple redundant failsafes and interlocks, whatever general insurance this business has would drop you the day they found out this thing existed. Then they'd probably sic the NRC on you, and I have no idea what kind of legal assraping they'd hand out.

Even if you were to profit off this directly, it's a bad idea. If you aren't, triply so.

[Anonymous No.2541272]

>>2541144
What are you taking CT scans of? If it's anything living forget about it, wht ayou're describing is giving a million lawyers a million erections.

[Anonymous No.2541275]

>>2541144
>The hardware (turntable + vertical actuator) is trivial
>the software/algorithm used to compute slices (inverse radon transform) is available online for free
You SEVERELY underestimate the difficulty of these aspects
Why not build all the rest of it before you start looking for an X-ray emitter and plate? You'd be surprised how much you can learn from nothing more than a very bright light and a camera.
Don't believe me? Hold you hand over a bright flashlight and move it around. There's visible structure in there.

[Anonymous No.2541278]

>>2541266
Now this is a thoughtful and useful answer. Thank you. I'm going to take your advice and pass on the project, with one caveat - does all of the above apply even if I put the assembly into a proper commercial x-ray cabinet (and I'm using a commercial xray emitter)? That's likely what I would do anyway, rather than trying to cobble something together myself. It looks like they cost a few thousand, and that's no problem at all if the company is already writing a million dollar cheque for the "real" unit.

[Anonymous No.2541279]

>>2541275
>You SEVERELY underestimate the difficulty of these aspects

The turntable and vertical actuator are truly trivial, at least for me. I build stuff like that routinely (as do many others). I'm aware that things like precision, backlash, and alignment matter a lot here.

As for the algorithm, it seems like complicated stuff, but since there are libraries and examples available, it doesn't seem like a big hurdle to me. Not compared to the challenges imposed by involving xrays, anyway.

[Anonymous No.2541290]

>>2541278
Not OP but I think you would still face those same legal and insurance problems. The parts may be commercial but they're still being put together by YOU.
So it would still count as an X-ray machine made by an individual without the required permissions.

[Anonymous No.2541294]

>>2541290
>made by an individual without the required permissions.
OP says it's a testing device for use in a laboratory. There are no permissions required.

There's so much misinformation and stupidity in this thread.

OP, you can purchase a suitable x ray source from ebay like William Osman did. Check his video. It's easy enough to construct adequate shielding and test for leaks.

[Anonymous No.2541295]

>>2541294
Well shit, guess I'm a retard then. My bad.

[Anonymous No.2541301]

You can rent time at a national lab lbnl or Argonne synchrotron based CT imaging.

[Anonymous No.2541303]

>>2541278
>does all of the above apply even if

I don't know. If you don't actually build anything, and just put together stuff that has whatever relevant certification it needs, you may be fine. The exact wording of any applicable regulation would matter here. I don't know what that would be, and you'd have to consult an authority better than a Polynesian balloon animal forum like /diy/.

>>2541290
>The parts may be commercial but they're still being put together by YOU

Again, depends entirely on regulation. NEC, example, generally forbids using non-listed or "unapproved" fixtures, boxes, conduit, and whatever, regardless of whether or not you're a licensed electrician or not. It doesn't prohibit doing electrical work yourself with approved materials. Requiring pre-approved equipment but not requiring licensing (usually with stipulations) is pretty common in various building codes.

Whether or not that's true here is still a question for someone who specializes in that sort of thing, though.

[Anonymous No.2541309]

>>2541294
>There's so much misinformation and stupidity in this thread.

>be retarded op
>don't understand much about ionizing radiation, xray equipment, testing of equipment, standards of practice.
>nigger rig my DIY CT Machine.
>Accidentally expose several people from scatter, leakage, malfunction
>Get sued
>at court
> be asked, were you trained in radiation, radiation biology, engineering of xray equipment, operation of xray equipment, standards of dose protocols, safety standards of equipment, industry certifications, OSHA Law
>answer no to all those questions.
>be laughed at by the court
>ass raped into bankruptcy.
>face criminal OSHA violation

https://www.osha.gov/laws-regs/regulations/standardnumber/1926/1926.53

1926.53(a)
In construction and related activities involving the use of sources of ionizing radiation, the pertinent provisions of the Nuclear Regulatory Commission's Standards for Protection Against Radiation (10 CFR part 20), relating to protection against occupational radiation exposure, shall apply.

1926.53(b)
Any activity which involves the use of radioactive materials or X-rays, whether or not under license from the Nuclear Regulatory Commission, shall be performed by competent persons specially trained in the proper and safe operation of such equipment. In the case of materials used under Commission license, only persons actually licensed, or competent persons under direction and supervision of the licensee, shall perform such work.

[Anonymous No.2541314]

>>2541242
>He's got a geiger counter and addresses this at 5:00 in the second video I posted

DIY xray genius casually compares ionizing xray radiation to non-ionizing UV radiation.

>it's less than a plane flight across the country.

This is why ionizing radiation is heavily regulated. People who think they are smart fuck up.


Maybe go look up what happed to all the people they blasted with xrays to clear up skin, or what happed to peoples feet that had fluoroscopy to see how shoes fit.

[Anonymous No.2541315]

>>2541314

You're a fucking idiot, dude.
1) He's talking about cosmic rays, which are ionizing radiation. The only one talking about UV is you. You're hallucinating again. You're the obnoxious, toxic sort of person who invents things to object to so you can get on your soap box and denounce your target. It's pathetic and impotent.

>Maybe go look up what happed to all the people they blasted with xrays to clear up skin, or what happed to peoples feet that had fluoroscopy to see how shoes fit.
This has absolutely no relevance to the topic at hand, which is the construction of an appropriate enclosure to safely operate a small DIY CT scanner for the purpose of research on non-living specimens. Again, you're an idiot. It's like someone is asking about what gun safe to buy and you're telling people DUDE GO LOOK AT PHOTOS OF GUNSHOT WOUNDS BRO, YOU ARE GOING TO KILL SOMEONE BRO

[Anonymous No.2541322]

>>2541144
>) Is this a really stupid idea?

obviously.

[Anonymous No.2541346]

>>2541309
>shall be performed by competent persons specially trained in the proper and safe operation of such equipment

So are there any specifics regarding what training is required for materials not under license from the NRC, or is that deliberately left vague?

[Anonymous No.2541377]

>>2541309
>https://www.osha.gov/laws-regs/regulations/standardnumber/1926/1926.53
a banana is a source of ionizing radiation too. I'm sure OSHA is going to be kicking doors down left and right after they realize that

[Anonymous No.2541435]

>>2541279
Well then. Get building. At the very least you can have a high precision 3d acanner, right?

[Anonymous No.2541451]

>>2541377
I'm sure if you worked in a food processing plant handling large quantities of bananas, yes, there would be training in the safe handling of them. It might just be a bullet point in a powerpoint presentation, but if there was a workplace incident and the OSHA didn't see it somewhere in the training materials, then asses would be on the line.

[Anonymous No.2541475]

>>2541257
wrong, radiography for welding still uses cobalt sources all the time. if you’re talking only for medical use you might be correct

[Anonymous No.2541477]

Actual nuclear engineer here. Well trolled post, OP. Made me reply. First, the million dollar industrial CT scanner is worth it if you want actual CT images. This includes not only the x-ray hardware and enclosure, but the operating software that computes the tomographs (the "C" and "T"). It is arguably the most sophisticated component. It takes complex linear algebra and quaternion manipulation to combine the images into one coherent image model. Bonus is your customer can also charge a fee for the service to perform industrial radiography as a 3rd party with that kind of instrument, too.

If you just want 2D images to look inside objects there are hand portable x-ray generators that plug into mains voltage. They are not big; essentially an x-ray tube on one end, control electronics in the middle, and a power supply at the back. They have a long cable that is attached to a switch so you exit the room and can control the exposure. They expose film in a case cartridge that is then removed and developed on the spot like a Polaroid instant picture. Very inefficient and emits more radiation but good for taking x-rays of objects you can't move, valves, conduits, etc.

I don't want to discourage you from DIY exploration, but this is very seriously dangerous. Reading your level of Q&A in this thread I would strongly advise you to consult a competent health physicist about shielding and x-ray exposures. You are working in systems with high voltages and photon energy regimes that can absolutely injure, permanently disable, or kill you and your coworkers. It requires specialized physics and training to understand and mitigate those hazards. The world does not need another Therac-25.

[Anonymous No.2541672]

>>2541477
>Actual nuclear engineer here. Well trolled post, OP. Made me reply.

Thanks for the reply!

> First, the million dollar industrial CT scanner is worth it if you want actual CT images.

Yeah, as I said, I don't expect to be able to produce something as good as a commercial machine; it's mostly about getting something functional while we wait for the "real" machine to be delivered.

> This includes not only the x-ray hardware and enclosure, but the operating software that computes the tomographs (the "C" and "T"). It is arguably the most sophisticated component. It takes complex linear algebra and quaternion manipulation to combine the images into one coherent image model.

These algorithms (and various implementation) are publicly available as well, and I'm a software dev with 20+ years experience.

> Bonus is your customer can also charge a fee for the service to perform industrial radiography as a 3rd party with that kind of instrument, too.

We've thought about that but it's a very niche market and we don't expect to see much action.

>If you just want 2D images to look inside objects there are hand portable x-ray generators that plug into mains voltage.

It needs to be a CT scan, unfortunately.

>I don't want to discourage you from DIY exploration, but this is very seriously dangerous. Reading your level of Q&A in this thread I would strongly advise you to consult a competent health physicist about shielding and x-ray exposures.

> You are working in systems with high voltages and photon energy regimes that can absolutely injure, permanently disable, or kill you and your coworkers. It requires specialized physics and training to understand and mitigate those hazards. The world does not need another Therac-25.

The voltages are irrelevant, and this comment makes me question the applicability of your entire post because it seems you've failed to grasp what I'm doing here. I'm buying an off the shelf xray unit, power supply and all.

[Anonymous No.2541721]

Well, have you decided on which X-ray source to buy yet?

[Anonymous No.2541766]

>>2541721

I'm pretty flexible (this isn't something I'm rushing into) but the unit in pic related >>2541156 is a GE 700 Model 46-197360G1 on sale for a great price and fully functional. Since it's designed for dental work, it would only be suitable for small scans, but that's perfectly fine. As I mentioned in a previous post, the manual trigger is a big plus for me since it means I can easily splice in an opto to trigger it (and since I'm replacing a human-actuated button, I know the system isn't finicky about the timing/duration of the trigger). The knobs can also be replaced with digital potentiometers (although that would be further down the road once the system is working when manually operated).

It's a shame there have been so few useful replies to this thread. What I was hoping for was responses like:
- yes, it's dangerous, but here's how you mitigate those dangers
- if your x-ray is a class X device, you'll need an enclosure like Y, etc
- you'll want this specific geiger counter or sensor to validate the effectiveness of your enclosure
- here are some sensible fail-safes to implement

Instead I got 30+ variations of "DUDE do you even KNOW that xrays are DANGEROUS??? bro do you know how much VOLTAGE is involved?

[Anonymous No.2541807]

retard here, geiger counters may not be ideal for ensuring safety. they're not awful, but iirc geiger tubes aren't that sensitive to x-rays. might be cost effective to cover photodiodes with alfoil to make arrays of sensors scattered all over the place. see what the backscatter is like.

also an off-the-shelf dental x-ray not designed for ct use may not be rated for continuous use. x-ray tubes have thermal limitations, worst case it could take hours to take enough single images for a full 3d scan. also maybe the firmware is "smart" enough to forcibly prevent you from using it that often.

[Anonymous No.2541814]

>>2541672
You can DIY this anon, just be open minded to expert opinion. My advice is free and carries no warranty. Post your results when you're done.

>you've failed to grasp what I'm doing here.
No, I understand your intent. You are over confident.

>I'm a software dev with 20+ years experience.
Ok, your professional experience is significant. I know as much software engineering as you do nuclear engineering. For a groundbreaking paper on software and x-ray machines I direct you to Nancy Leveson's paper on the Therac-25. Please, don't fuck around.

>I'm buying an off the shelf xray unit, power supply and all.
These x-ray units are designed for single exposures and cool-down time. Your use case is outside of the intent and capability of the general system specification. Try it for your experiment and it might work on your intended scale for short runs. Making x-rays for each image creates a large amount of energy in the form of heat in the x-ray head that must be removed. These heads have a thermal cutoff to prevent damage from short duty cycles.

By all means purchase the x-ray unit and find a suitable flat panel detector to pair it with. Take lessons from amateur photography, analog and digital. Make some 2D test exposures. Build your turntable and rotate your object around and take x-rays through only one polar axis at a time until you figure out what you're doing. These x-rays are shielded by drywall of a well-constructed room (no gaps or holes in walls or joints). X-rays are photons just the same as visible light, they will reflect from surfaces but not bend around corners. You and your coworkers gtfo the room before energizing the machine. Put interlocks on all egresses. And above all, don't fuck around.

[Anonymous No.2541815]

>>2541814
Forgot pic for this image-less thread.

[Anonymous No.2541830]

>>2541807
>retard here, geiger counters may not be ideal for ensuring safety. they're not awful, but iirc geiger tubes aren't that sensitive to x-rays.
> might be cost effective to cover photodiodes with alfoil to make arrays of sensors scattered all over the place. see what the backscatter is like.

I was looking at buying a geiger counter but this is an interesting idea and easy enough to execute at scale - I could design an all-in-one PCB with some photodiodes + microcontroller and have a bunch of them made (JLC etc).

>also an off-the-shelf dental x-ray not designed for ct use may not be rated for continuous use. x-ray tubes have thermal limitations, worst case it could take hours to take enough single images for a full 3d scan. also maybe the firmware is "smart" enough to forcibly prevent you from using it that often.

Thanks - this is exactly the kind of information I posted this thread for. To find out what I don't know I don't know. What class of xray should I be looking at, and what is a realistic price range for a suitable second hand unit?

>No, I understand your intent. You are over confident.
If I was overconfident I would have gone ahead and done it without asking internet people for advice.

>These x-ray units are designed for single exposures and cool-down time. Your use case is outside of the intent and capability of the general system specification. Try it for your experiment and it might work on your intended scale for short runs. Making x-rays for each image creates a large amount of energy in the form of heat in the x-ray head that must be removed. These heads have a thermal cutoff to prevent damage from short duty cycles.

As with the other poster - thanks, this is useful information. I wonder, how much heat are we talking about? Would it be reasonable to introduce some active cooling to reduce cycle times? Air, water, thermoelectric, etc. I even have some phase change coolers (-40C) but that's probably overkill.

[Anonymous No.2541833]

>>2541144
This is a really fucking stupid idea. It'd be really fucking stupid if you were doing it to /diy/ CT scan shit in your own home, because you'd be giving yourself cancer, but to bring in a /diy/ CT scanner to work, give cancer to everyone you're working with, then get found out that you've been running a highly radioactive, uncertified machine at work, your employer isn't going to be happy, your employer is going to get sued to hell and back, you will get fired, and you will get sued. Unless you live in a third world country, but then your employer wouldn't have $1 million to buy a CT. Even if you could build a safe machine, which you clearly can't since you're already trying to weasel around the safety requirements "because some guy on jewtube did it and said it's fine", just the fact that it's not certified is enough to completely fuck you in the ass.

[Anonymous No.2541836]

>>2541830
>have a bunch of them made
Honestly I was thinking of one board, with a bunch of SMD resistors, and all these pairs of wires going off to free photodiodes. Maybe you'd want phototransistors or avalanche photodiodes though. Best read up on x-ray detection techniques. And I guess you also want to measure for beta rays, if what the somewhat irate anon said was correct. Metal-walled geiger tubes are bad at low-energy beta (which I assume you'll have) so you'd want an end-window geiger-tube instead, if not an ionisation chamber. Ionisation chambers are really easy to make, so I'd definitely consider using them for beta.

The other option is using photographic paper like classic dosimeter badges, which I assume will work fairly similarly to silicon photo-detection. IIRC they use aluminium foil over the gamma detector and paper over the beta detector.

>What class of xray should I be looking at, and what is a realistic price range for a suitable second hand unit?
The only information I know about x-rays is from the Osman and Applied Science videos, plus some undergrad physics labs on radiation. I can only suggest that you look for information on what particular tube a machine uses, and then find that tube's datasheet. It's an option to copy the setup used by Applied Science, but IIRC that would make it a lot less safe. Both in terms of HV and radiation.

>>2541833
someone already made that post, try again anon

[Anonymous No.2541840]

/diy/-spirit and all, but engineering such a machine from scratch is extremely hard. You would need to study the science behind the radiation mechanisms, connect them with electronics and then you would need to develop software to evaluate and display everything properly.

You would need a team of several blue and white collar specialists plus a lot of time and money.

[Anonymous No.2541853]

>>2541766
>Instead I got 30+ variations of "DUDE do you even KNOW that xrays are DANGEROUS??? bro do you know how much VOLTAGE is involved?

because you are acting like a fucking retard in more than one way, and fail to appreciate the science.

half a million dollar machines, installed by trained professional fucking malfunction. a team of technologists, physicists, and intense protocol is required to mitigate the consequences. and still people are over exposed.

you clearly dont know what you are doing.

imaging technology is some of the most sophisticated technology known to man. there are several nobel prize winners in the field.

you have a fucking nuclear engineer in the thread and you are arguing with him. fucking christ.

[Anonymous No.2541860]

>>2541144
Jesus christ my dude, ive worked on some pretty batshit projects before so i know my way around but this idea takes the cake for sure...

1st of all fuckwit; you are a software engineer. Wth are you thinking about building a ct machine when even based on this post you know fuck all?
2nd think about what you are even doing. High voltage. Dangerous wavelengths. Off the fucking shelf equipment. "I just need to know how to make an enclosure"
3rd this is either well made bait or a dead man posting

Fuck man. Grow a brain cell and admit that this is a stupid fucking idea not worth doing, much less asking for help from an online forum, and on /diy/ of all places.

[Anonymous No.2541880]

>>2541853
>because you are acting like a fucking retard in more than one way, and fail to appreciate the science

Nah it's because most people don't read more than a sentence or two and imagine I'm intending to build a machine from scratch, ie I build a high voltage power supply and I'm playing with x-ray tubes in my garage like an idiot. They love to get on their pedestal and talk down to somebody they imagine as an inept and foolish based on their own misinterpretation of intentions.

In reality what I'm looking to do is conceptually similar to an automated photogrammetry setup, only with dangerous radiation which needs to be contained via enclosure (in this case, an off the shelf x-ray cabinet). The cabinet in this case would be far too small for a person to climb inside; thus, it seems to me that the only real retirement make such a system safe is to ensure that it only ever powers on if the door is closed (can be done mechanically and electronically).

[Anonymous No.2541881]

>>2541144
https://www.youtube.com/watch?v=IiJAq53knwc

[Anonymous No.2541901]

>>2541144
it depends on your required resolution there are cheap diy builds with shit resolution. The problem is always the fucking magnet if you had a friend with a SEM who knows a big about 3d (nano-)printing you could probably build a high end CT scanner for under 50k. Google turns up lots of shit
https://hackaday.com/2011/09/15/build-your-own-ct-scanner/
https://hackaday.com/2015/05/29/hackaday-prize-entry-a-better-diy-ct-scanner/
https://hackaday.com/2019/02/20/diy-x-ray-machine-becomes-ct-scanner/

[Anonymous No.2542138]

>>2541901
The magnet? Are you thinking of an MRI? Thanks for the links though.

[Anonymous No.2542151]

no insurance company is going to work with your company if you're using a homemade ct scanner. no patients are going to pay your company out of pocket for scans from your homemade scanner.

[Anonymous No.2542163]

>>2542151
>no insurance company is going to work with your company if you're using a homemade ct scanner. no patients are going to pay your company out of pocket for scans from your homemade scanner.

> t. retard who doesn't even read the first post and thus thinks this is being used for medical applications

[Anonymous No.2542191]

>>2541901
>it depends on your required resolution there are cheap diy builds with shit resolution.

This is one of the big unknowns for me as I assess the feasibility of the project - will a diy solution inevitably be crappy resolution, or is it possible to get decent results without a high end setup?

[Anonymous No.2542216]

>>2541475
Cobalt is a gamma ray source, not the same as X-ray.
Do you have any qualifications for offering advice on ionizng radiation?

[Anonymous No.2542250]

What parts are you trying to CT? By that i mean what material is it made of?

Most dental wont go high in voltage, like 100kV (im not from that field so that might be incorrect). With that you wont be able to scan even thin metal parts. Also dental tubes might not be suitable for long term operation. Commercial machines have water/oil cooled anodes (only 1% of the input energy is converted into x-ray, 99% is lost and becomes heat in the anode). I guess its an okay compromise if a DIY scan takes 2 hours instead of 20 minutes

Most of the phosphor plates out there need a reading machine, actual light emitting ones are rare, try looking for scintillator plates. But the quality would be garbage with these plates. Radiation would hit the camera recording the plate and you would get tons of speckle, can be filtered tho.

If i wanted to build one id go on ebay and buy used parts, there are tons of high voltage generators, tubes and detectors out there. It would be possible to whip up a machine for fraction of the price of a commercial new machine.

These things are regulated. You can still do it but if someone finds out you can get fucked easily. I wouldnt trust other people to operate a DIY machine at all.

Commercial machines rarely use lead. Just steel and a lot of it. The resort to lead only where space is tight like gaps and small openings.

Doses with a dental x-ray arent that menacing. The radiation goes down by the square of the distance from the source.

The voltages involved are lethal (100kV can arc across 1 meter!) but if the cabling and insulation (lots of plastic and grease) is sound it is not a problem. Of course you wouldnt want to tinker inside a generator right after you turned it off.

Commercial CT reconstruction (and analysis) software is pretty expensive but there are free options.

I love how 90% of the posters have no idea what this involves and says that it is crazy because of fear and superstition about radiation and high voltages.

[Anonymous No.2542257]

>>2541157
>commercial xray use radioactive material for the anode
>former xray tech
Biggest load of bullshit ive ever seen.

Get it right next time:
The anode (or target) in an X-ray tube is tungsten or copper most of the time and they are not isotopes so they do not emit radiation by themselves.They are not radioactive. High velocity electrons hitting the target generate radiation, if you turn off the power the electron flow stops and the radiation stops. X-ray tubes are used for radiography and CT.

Isotopes of iridium, cobalt and whatever are used as an x-ray source, they are radioactive (they are not an anode because they are not in a vacuum tube or in an electrical circuit). These emit radiation all the time, they cannot be turned off. If you want to "turn it off" you "put it away" in a lead container. These are used mostly in radiography of large structures, and never for CT.

[Anonymous No.2542284]

>>2542250
First of all - thanks for this response. Very informative and helpful.

>What parts are you trying to CT? By that i mean what material is it made of?
> Most dental wont go high in voltage, like 100kV (im not from that field so that might be incorrect). With that you wont be able to scan even thin metal parts. Also dental tubes might not be suitable for long term operation. Commercial machines have water/oil cooled anodes (only 1% of the input energy is converted into x-ray, 99% is lost and becomes heat in the anode). I guess its an okay compromise if a DIY scan takes 2 hours instead of 20 minutes

Plants and produce; leaves, sticks/branches, nuts, fruits, etc. I assume leaves will require the least power since they're so thin and already translucent. With this in mind, I suspect a dental xray would be sufficient.

>Most of the phosphor plates out there need a reading machine, actual light emitting ones are rare, try looking for scintillator plates. But the quality would be garbage with these plates. Radiation would hit the camera recording the plate and you would get tons of speckle, can be filtered tho.

What would be a more optimal approach, in terms of resolution?

>If i wanted to build one id go on ebay and buy used parts, there are tons of high voltage generators, tubes and detectors out there. It would be possible to whip up a machine for fraction of the price of a commercial new machine.

That's a bit too granular for me; I'd rather buy a commercial (xray) machine and know it's well built and functions properly if I operate it correctly.

Suppose I were to place a dental xray machine entirely inside of a commercial xray cabinet like pic related. As long as I have a properly functioning interlock that prevents the xray from ever energizing unless the door is properly closed, is there really any other significant safety concern? It doesn't seem like it to me, but again, I don't know what I don't know.

[Anonymous No.2542300]

>>2542284
>What would be a more optimal approach, in terms of resolution?

Just to follow up on this - I'm currently reading https://www.scintillator-crylink.com/industrial-ct/ and it's mostly answering this question for me. If there's anything important left out, let me know. I haven't gotten all the way through it yet, but I wonder if I could position a lens after the scintillator plate in order to shrink the image and focus it on a (visible-light) CMOS sensor.

On that note - pic related is basically what I want to put in the cabinet. Since my source/sensor (or perhaps subject, we'll see) will be moved vertically for each slice, my sensor array can be a thin strip (as depicted) rather than a conventional squarish image sensor. Can I pack photodiodes on that strip/array, put some scintillator in front of the array, and basically call it a day?

[Anonymous No.2542305]

>>2542300
>I wonder if I could position a lens after the scintillator plate in order to shrink the image and focus it on a (visible-light) CMOS sensor.

Did a bit more digging and it looks like indeed I can (pic related). This approach makes sense for cone beam, but I suspect a vertical scanning approach with a thin sensor strip (as outlined above) would produce better results.

[Anonymous No.2542310]

Looking at xray panels now, something like https://www.ebay.ca/itm/195320037238 looks pretty good. At ~$8k usd it's not super cheap but it's well within the budget if it performs reasonably well.

It looks like there are a lot of veterinary xrays available (both new and used) but the dental one I mentioned earlier seems like a better fit to me because:
1) lower energy
2) likely easier to automate

[Anonymous No.2542315]

>>2542284
>plants n shit
Yeah dental is fine
>resolution
Depends on the focal spot of the source and the resolution/quality of your detector.
I cant tell whats the focal spot of a dental xray but looking at datasheets will get you started. Optimally a micro-focus system would be your best bet (focal spot ~>10 microns) Regarding the detector a >0.04mm detector pitch would suffice. If you get used stuff this is what to look for.
The problem with the phosphor screen plus camera is that the light emitted from one spot will illuminate every pixel of the camera. Not sure how to describe it correctly, hopefully you get it. Filtering for that is almost impossible. In a real detector the light emitting part (the scintillator) is right on top of the light detecting device so theres no possible light spilled over from other detector element. Also you need to calibrate for distortions. and correct them.
I have acquaintances who work at a nuclear facility, they have this sort of system except they are using neutrons from the reactor not x-rays. Sadly i cant recall the quality of CT images that system puts out.

>prevents the xray from ever energizing unless the door is properly closed, is there really any other significant safety concern
I dont think so.

>>2542300
>>2542305
ZEISS Xradia Versa does just that.

[Anonymous No.2542432]

>>2542305
>a thin sensor strip
Like the ones in a document scanner? Where do you get them from and how do you drive them?

>>2542315
>the light emitted from one spot will illuminate every pixel of the camera
Maybe that limitation applies to using a raw CMOS sensor, but in an existing camera that's what the aperture is for. Use of a lens before the aperture will allow you to increase the light intake without making the image blurrier.
>distortions
I suspect this will be a very real problem.

[Anonymous No.2542531]

>>2542250
>These things are regulated. You can still do it but if someone finds out you can get fucked easily. I wouldnt trust other people to operate a DIY machine at all.

>I love how 90% of the posters have no idea what this involves and says that it is crazy because of fear and superstition about radiation and high voltages.

shut the fuck up dude. the shear cognitive dissidence.

It's a shit idea for many reasons. You even say you wouldn't trust anyone DIY machine. OP doesn't know jack shit about radiation, the law, or anything really.

go waste your money being a dumbass I guess.

[Anonymous No.2542537]

>>2541144
This clown made a shitty diy xray so i have faith you can to OP.
>https://www.youtube.com/watch?v=IiJAq53knwc

[Anonymous No.2542610]

>>2542531
Who hurt you?

[Anonymous No.2542620]

>>2542531
>cognitive dissidence
Wanna try again, illiterate boy? The word you were retardely grasping for is "dissonance." As in "cognitive dissonance".

And, this is not an example of it. So you're even more retarded than you appear at first glance, because you're trying to use words to insult someone, that you don't know the meaning of.

Also, the other word you were retardely grasping for with your tard helmet on is "sheer", "shear" means cutting something.

Let's all point and laugh at this fucking retard.

[Anonymous No.2542621]

>>2542610
A dictionary, apparently.

[Anonymous No.2542676]

>>2541144
this is a dumb idea for several reasons not withstanding you're apparent ignorance.

you don't seem to give a fuck about the law or liability and you absolutely should.

a dental x-ray machine takes low voltage xrays that are like 1''x1''. trying to image 50cm cubed isn't gonna work. you didn't say what you are trying to image which is a crucial aspect.

you seem way to Calloway and that's the reason half the people in this thread are calling you a dumbass.

[Anonymous No.2542681]

>>2542620
...because you're trying to insult someone, but you don't know the meaning of the words that you're using.

Ftfy

[Anonymous No.2542689]

>>2542681
Wow.

Just...wow. You are that fucking retarded. How do you exist? You're so dumb they probably have to retrain you in how to breathe every morning.

[Anonymous No.2542694]

>>2542689
He did a good job of editing the awkward phrasing of the other guy.

[Anonymous No.2542695]

>>2542689
Seriously? Incorrect comma usage, as well as ending a sentence with a preposition, are things that you think should just be IGNORED?! Are you fucking SERIOUS? What kind of animal are you?

[Anonymous No.2542713]

>>2542676
>you're apparent ignorance.
You cant even spell, retard.

[Anonymous No.2542752]

>>2542676

Another fucking idiot marches in and hallucinates 90% of the information he's "responding to".

>this is a dumb idea for several reasons not withstanding you're apparent ignorance.

your*

It's interesting that you call me ignorant - on what basis do you make this assessment? Be specific. As you review the thread, you may discover that all of your assumptions about what I do or do not know are just that - assumptions.

>a dental x-ray machine takes low voltage xrays

it's 70kvp, that's lower than an industrial machine that scans metal but it's right in the middle of biological xrays and actually probably overkill for what I need

> that are like 1''x1''. trying to image 50cm cubed isn't gonna work

Yeah I know, that's why I said:
> Would a dental xray like >>2541156 be suitable for the system I described in OP? I realize the field of view is quite small so it would only be usable for small specimens, but that's fine as a proof of concept


> >you didn't say what you are trying to image which is a crucial aspect.
yes I did

>you seem way to Calloway

definitely too calloway

In reality, I made this thread precisely because I know xrays are dangerous and I wanted some pointers as to what class of enclosure I would need to operate this device safely. The whole point of the enclosure is to eliminate the risks, which multiple knowledgeable anons have confirmed to be a perfectly sensible approach. Somehow, imbeciles you like read this exchange as "I don't think xrays are dangerous" and take it upon yourselves to lecture me about high voltages (irrelevant) and xray radiation (the protection from which was the whole point of me creating this discussion). KYS

[Anonymous No.2542762]

>>2542752
cope and seethe more you fucking idiot.

[Anonymous No.2542767]

>>2541144
https://www.youtube.com/watch?v=IiJAq53knwc

[Anonymous No.2542769]

>>2542762

> get demolished

> uhh... c-c-cope and s-s-eethe!

GOT EM!

[Anonymous No.2542771]

>>2541144
>I've noticed that dental xrays like pic related appear to operate without any backing phosphor plate; presumably there is some image sensor (or equivalent assembly) in the emitter and it's measuring particles that bounce back rather than particles that go through the subject. Is this accurate? Would this be suitable for a DIY CT scanner, provided the subjects being scanned are similar scale to a human jaw

imagine writing this non-sense and then getting offended when people call you ignorant.

[Anonymous No.2542775]

>>2542771

Yeah, and 12 minutes later I knew the answer. Congratulations on discovering the process of finding out new information. Notice that this has nothing to do with the dangers posed by xrays.

[Anonymous No.2542790]

Go for it. The software aspects seem difficult and intimidating for me as I'm not a software engineer. The xray part is something that I went through and it wasn't particularly hard at all, by far not the most challenging hobby project. I'll repeat some of the details the few, very few, competent anons mentioned here.

A dental xray unit is fine in terms of xray energy for plants and shit. Focus may be a bit fuzzy but probably alright. Typical spot size is about 0.7-1mm for a dental tube. Go and find a microfocus source like that one anon said.

Bigger issue is the low duty cycle. Tube is designed to provide a relatively high dose of xrays briefly while the copper anode acts as a heat sink. That heat is then slowly rejected to the dielectric oil that fills the head. I don't have the datasheet for one of my tubes handy now but all of this is called out and it's fairly limiting in terms of what you can do. I had to go with a custom power supply as I wanted continuous operation at lower doses. You could find a more modern dental xray head that a switching power supply. The one I took apart had the ferrite transformer and multiplier inside the head and used a separate transformer for the filament. Control electronics were external and the head had divider resistors and a shunt for measuring tube voltage and current, respectively. Gotta make your own driver. Run the tube at your desired voltage which could be well under 70kV for optimal contrast with plants. Adjust your filament drive to get a low current through the tube such that it can operate continuously.

[Anonymous No.2542791]

Part 2 of my diary

Older xray cassettes are still available on eBay. Get the ones with rare earth phosphors for better efficiency. You can build up a box and image them with a camera. Chest sized cassettes exist, harder to get anything beyond that. Even the veterinary cassettes are about chest sized. Better yet get an xray intensifier. That was a game changer for me though you're typically limited to a 6-9" diameter sensor. Output is bright, easily imaged with a camera, and you need ridiculously low xray intensities to get a good image.

Making this as a product seems like liability hell but you do you. Shielding isn't particularly hard. If you run at lower xray energies which for plants may be better then your shielding becomes even easier. Something around 20keV sticks out in my mind for plants and other applications. I vaguely recall titty xrays run around 20keV since it's all soft tissue. For detecting xrays and knowing if you have a leak somewhere, get a mica window geiger tube. Be mindful that geiger tubes saturate easily and if it stops counting that is indicative of a bunch of xrays coming out. Get a dosimeter just to be safe. Some dosimeters stop responding around that 20keV level so if that's where you're fucking around, it may be useless and soft xrays like that are particularly nasty.

[Anonymous No.2542829]

>>2542790
>Go for it. The software aspects seem difficult and intimidating for me as I'm not a software engineer.

I don't want to get ahead of myself because I know this is a deep field, but it seems like the simplest algorithms (backprojected & filtered back projections) are actually quite simple, and already implemented in many libraries.

>Better yet get an xray intensifier. That was a game changer for me though you're typically limited to a 6-9" diameter sensor. Output is bright, easily imaged with a camera, and you need ridiculously low xray intensities to get a good image.

Yeah, an intensifier + camera was my initial intention (as I was looking to copy the setup in the first video I posted) but after doing more digging it seems like spending ~$10k on a commercial xray panel (https://www.ebay.ca/itm/195320037238)) would be worth the investment; presumably the resolution alone would dwarf whatever system I could cobble together, and at $10k it's not a huge investment. There's also the time savings (instant solution) and the knowledge that it works properly right away. Would a panel like this be a fit match for the dental xray machine I mentioned in my second post?

>Making this as a product seems like liability hell but you do you
It will never be product (never sold to third parties), and there's a good chance I'll be the only one operating it. For what it's worth, though, once it's in a suitable xray cabinet with a properly operating interlock, I don't think it actually matters who operates it. Maybe that view will change as I learn more, but that is my impression right now.

>Be mindful that geiger tubes saturate easily and if it stops counting that is indicative of a bunch of xrays coming out. Get a dosimeter just to be safe.

Yeah, it goes without saying that I'm going to have multiple redundant sensor systems rather than just assuming it's all working fine.

[Anonymous No.2542966]

OP, why don't you try looking up a C-arm? They provide constat x-ray exposure and their own image output, which can be easily extracted (usually) through a BNC connector. Something like a GE OEC 9800

[Anonymous No.2543046]

OP here. Spent a couple hours building a (simple) simulated CT scanner and a reconstruction algorithm and I think the results are pretty good. Pic related. I definitely need to try this with some real world datasets, and also adjust it to account for the fact that I'll likely be using a machine with cone beam geometry rather than parallel beams (and my simulation & reconstruction here assumes parallel beams), but to me it's an encouraging result so far. Thanks to all anons who posted helpful information.

[Anonymous No.2543056]

Improved version that tries to recover the opacity (density) information instead of just a black & white image

[Anonymous No.2543119]

I did software testing in college for a module, one of the examples used in a book was someone being cooked by a CT-scan machine because someone accidentally coded something somewhere so that an answer was -1 instead of 1. So good luck with that.

[Anonymous No.2543425]

>>2541144
Ahh yes, let's make x rays and shoot it at stuff that reflects it in all directions.

[Anonymous No.2543432]

>>2541156
Have you never had you teeth xrayed or did you simply forget about the sharp pieces of plastic they shove in your mouth?

[Anonymous No.2543449]

>>2541377
>>2541451
>fall into vat of mashed banana
>become the next superhero because of mutations
>stop crimes by shooting banana peels from my ass

[Anonymous No.2543454]

>>2543449
I see you too are a man of culture

[Anonymous No.2543480]

>>2543046
>>2543056
Try out astra toolbox. I got good results and it is very flexible. Works on GPU aswell.

[Anonymous No.2543488]

>>2543425
>>2543119
>i dont know shit about the topic
>but ill chime anyway in with urban legends, misconceptions and irrelevant information

[Anonymous No.2543532]

>>2542191
>will a diy solution inevitably be crappy resolution
yes simple as

[Anonymous No.2543555]

>>2543119
There's was a machine that accidentally shot full doses of radiation if the person operating the controls went too fast.

[Anonymous No.2543594]

>>2542767
I mean, he does have a real engineering degree

[Anonymous No.2543816]

>>2543056
Even then it doesn't look like it's picking up the opacity of the ring around the centre, maybe that's just an artefact that you don't have enough degrees of freedom to remove.
I'd try adding distortions to the images to see what happens. Like a bit of random jitter about the 1° increment, or a bit of per-slice or beam-angle-dependant brightness modulation, or random brightness noise, or what happens as a result of non-collimated x-rays (i.e. diverging). I imagine there must be existing correction algorithms for beam-angle and divergence errors.

>>2543594
He is a shitty engineer.

[Anonymous No.2543831]

>>2542791
Do dosimeters or rf detectors exist that aren't hippy garbage?

I work around cell phone antennas and no one in my area knows what PPE is.

[Anonymous No.2543835]

>>2543816
>Even then it doesn't look like it's picking up the opacity of the ring around the centre, maybe that's just an artefact that you don't have enough degrees of freedom to remove.

The "raw" reconstruction (third image from left) picks up the semi-transparent center ring correctly, but yeah, the output (which has been masked) hides most of it (which is incorrect). There are a lot of settings you can play with that affect the outcome. Bear in mind this is a toy algorithm that I threw together, I haven't implemented anything sophisticated at all (and I'm not using an existing CT library).

In terms of next steps, I'm doing two things:

1) building some fairly simple software that takes an existing 3d model (point cloud or vertices from an obj etc) and produces "ground truth" horizontal slice images like the input images I made in mspaint for my software demos above (using the vertex colors to assign per-pixel "material density"); I will then feed them into my simulated ct scanner, reconstruct the slices, and build a 3d model from those slices. It will be interesting to see how well the result matches the input 3d model.
2) build a visible light ct scanner and feed the resulting photos into my reconstruction algorithm to see what kind of results I get with translucent/transparent objects like plastic bottles etc

[Anonymous No.2543846]

>>2543831
"Radiation detection" basically always refers to ionising radiation, RF power meters are what you're after. There may be inline coaxial ones also, you probably want one that uses an antenna. What frequencies do you care about? Could tune an SDR with a bunch of input attenuation and have it screech when it picks up too high a signal strength.

>>2543835
>building some fairly simple software that takes an existing 3d model
Sounds like a pain, tried doing geometric intersection code before and it didn't turn out well. But that's probably because I was a brainlet who was trying to reinvent the wheel. In python.
>build a visible light ct scanner
That's a really fucking cool idea for a project. But you'll likely run into significant problems with refraction and reflection. Actually I bet X-ray reflection is significant if you're imaging someone with metal implants. X-ray refraction though, not sure.

[Anonymous No.2543886]

>>2543846
Yeah I looked at rf power meters for a a minute and it's either real power meters for aiming antennas or stuff for EMI sensitive hippies. Fucknit. Maybe some of the stuff marketed for the EMI stuff will actually do a thing

[Anonymous No.2544081]

>>2543835
Maybe you didnt see my post, Astra-toolbox is very good.
It can do forward projection on a 3D or 2D array, to simulate a CT scan of an object and you will get the simulated projected images, what you would capture with a detector. Of course the physics interactions are not there (Compton scattering, beam hardening, accurate attenuation). And then it can do backprojection from projections to get the reconstructed 2D 3D array of your thing. It can also do reconstruction from non uniform angles. Lets say your system is not perfect, but you can

>visible light ct scanner
As other anon said, its a pain, too much reflection. Other than laying down the control foundations for a CT scanner i see no point dealing with visible light like that.
I can provide you with real projections of scans if you want.

>>2543846
>reflection is significant if you're imaging someone with metal implants
Its is a problem, even metals in polymer are a pain. Reflection has a small effect, the problem is the density difference and the mathematics behind it. In short, the bigger the density difference the worse the interface will be between the densities. Pic related is a screw in a plastic pipe and wood. Detail is lost around the metal part completely.

[Anonymous No.2545211]

>>2543846
>>build a visible light ct scanner
>That's a really fucking cool idea for a project. But you'll likely run into significant problems with refraction and reflection

>>2544081
>As other anon said, its a pain, too much reflection

Is anyone doing this? Suppose I found a way to do visible light CT scanning (glass/plastic objects). Would it be a big deal?

[Anonymous No.2545237]

>>2541157
When was the last time you were an x-ray tech, 1951? Hospital x-ray machines are portable, many are battery powered and if not plug into 110v, and no larger than a drink cart.

[Anonymous No.2547165]

>>2545237
My dentist's xray source is handheld and battery powered

Fucming dumb bitch didn't have the target synced to her computer and wanted to keep.shooting me after it didn't register 3 times.

[Anonymous No.2547181]

https://www.youtube.com/watch?v=IiJAq53knwc
William Osman was able to build this in a garage with a box of scraps.

[Anonymous No.2547183]

>>2547181
and then there's this thing https://www.youtube.com/watch?v=gqZ6PyIGJ_c

[Anonymous No.2547184]

>>2547183
and this https://www.youtube.com/watch?v=xdpDd7dyU00

[Anonymous No.2547615]

>>2547181
>literally the 4th time this link has been posted
why the fuck is osman's shitty contraption getting posted more than applied science's actual ct-scanner?

and read the damn thread, op obviously doesn't want to hack together something from a tube and hv power supply

[Anonymous No.2548800]

Bump for progress update?