4 ms·

High speed X-ray video: jumping beans, wind-up toys and more

61 points by surprisetalk 1y ago

echelon 1y ago
I got nervous watching him put his hand in front of the machine. You don't want to subject yourself needlessly to ionizing radiation. A little here and there is probably okay at small doses and for valid reasons, but it all adds up over a lifetime. He didn't mention how much X-ray radiation this puts out in the first half or so of the video. I worry it's more than a medical photograph as it's continuous high sample rate video, but I'm not an expert. Would be curious to know. Also curious about the shielding and leaking. Don't damage your DNA if you don't have to. This is a cool, semi-educational video. I don't think I'd take the same risk though.
- zzlk 1y ago
  I think it's probably not true that it all adds up over a lifetime. That's the model that is widely used but there's some good evidence to suggest it's not accurate. see: https://en.wikipedia.org/wiki/Linear_no-threshold_model https://en.wikipedia.org/wiki/Linear_no-threshold_model
- eig 1y ago
  Be rest assured Ben’s previous job was in the medial imaging industry. While he worked on MRI machines rather than ionizing radiation, I think he’s very well aware of the dangers of X rays and has many projects dealing with ionizing radiation. There’s a lot of bad safety science YouTubers, Ben isn’t one of them :) Funny thing: it’s actually rare to get radiation damage to human hands and feet since there’s not too much growing tissue there!
  - scythe 1y ago
    >there’s not too much growing tissue there! On the contrary, I was told stories in school that old IR doctors used to lose the hair on their hands after using the fluoro for years. The fingernails are also radiosensitive. The main reason that X-rays of the hands and feet are usually very low risk is because the beam intensity (dose) required to penetrate the small amount of tissue is very low. Because the video uses a high-sensitivity detector (photon counter) the dose may be even less than usual. However, it would still be a regulatory violation if you did it in a hospital.
- scythe 1y ago
  >I worry it's more than a medical photograph as it's continuous high sample rate video, but I'm not an expert. Would be curious to know. Typical fluoro skin entrance exposure rate to go through someone's hand is on the order of 1-3 millisieverts per minute. With a more advanced detector (like this one) it may be lower, but increasing the frame rate or resolution will tend to require higher dose rate. The associated risk of skin cancer is quite small. But please be aware that unnecessary X-Ray imaging of living things is against the law in most jurisdictions.
  - SoftTalker 1y ago
    I've wondered how much the average X-ray dosage dropped since they stopped using film and switched to detector-based imaging. Anyone know?
    - scythe 1y ago
      You can compare diagnostic reference levels (DRLs) from different years for the same exam; these correspond to the 75th percentile patient dose for a particular imaging study. It can be a little annoying to find older DRLs (I just tried). NCRP 172 (published 2015) provides a direct comparison of digital to film for some examinations. For upper GI fluoroscopy without oral contrast the DRL for film is 3.9 mSv and for digital is 1.5 mSv. I think this is roughly typical.
- rdtsc 1y ago
  > He didn't mention how much X-ray radiation this puts out in the first half or so of the video. I worry it's more than a medical photograph as it's continuous high sample rate video, but I'm not an expert. Would be curious to know. > Also curious about the shielding and leaking. If you'd watch the video the last part explains exactly that. The energy levels involved, how the beam is stopped, how backscatter is checked. He is definitely no dummy who just found an x-ray machine in a dump and just powered it up for clicks and giggles.
bkraz 1y ago
I may have the opportunity to use a much higher framerate camera in the future. Please leave suggestions for things that are fairly transparent to x-rays (no metal, or very thin small internal parts), and demonstrates the utility of high speed capture (typically things breaking, bouncing, shattering). The last requirement is that the object be very small, just 1 to 2 cm long. Since learning about ptycography, I may try that as well since x-ray ptycography is one of the real world uses for these sensors.
- s0rce 1y ago
  I thought the insect breathing work at the synchrotron was cool https://www.science.org/doi/10.1126/science.1078008 https://www.science.org/doi/10.1126/science.1078008
- jcims 1y ago
  I wonder if you could blow some fine mesh aluminum/iron powder through a small whistle to see any of the airflow effects. Spurious xray emission from triboelectric effect. Phase change of various materials changing absorption characteristics. Capacitor discharge.
- shellfishgene 1y ago
  This may be a bit hard, but jumping insects like locusts and crickets?
- Python3267 1y ago
  Here's an interesting one, if the equipment is sensitive enough you could test if you can see the difference in air densities as a .22 caliber bullet passes by.
- SweetSoftPillow 1y ago
  Hummingbird
- SweetSoftPillow 1y ago
  Brookesia nana aka nano-chameleon.
- 1y ago
lacoolj 1y ago
Not only is this incredibly cool, but it bridges my knowledge of x-ray (CR/XR, in this case, DR) even further. I work for a medical imaging company and we actually sell and promote our digital radiography machine that will do this, but for humans (at a much, much slower rate, but enough to be useful in medical diagnostics). I had never considered an application of use outside of medicine. So, my day is now accounted for. In addition to "overdue training" on things I already know, I will be youtubing all the x-ray, MRI, ultrasound fun things that you would never see otherwise. Thank you OP!
- cycomanic 1y ago
  Please post a link to your YT channel so we all can take part in the fun.
rdtsc 1y ago
That's wild they have a high speed 8K photon counting device for x-rays now. It's not the one he tested but it's the latest one the company has https://www.dectris.com/en/detectors/x-ray-detectors/eiger2/ https://www.dectris.com/en/detectors/x-ray-detectors/eiger2/ It's also neat how he describes how the acquisition works: 32bit tiffs, pci card, real-time linux, and a special rack-mounted server.
world2vec 1y ago
One of the most underrated channels on YouTube.
pmontra 1y ago
A Motif interface from the early 90s at about 10:00 in the video. That X-Ray detector has probably a very long history.

61 points by surprisetalk 1y ago

I got nervous watching him put his hand in front of the machine. You don't want to subject yourself needlessly to ionizing radiation. A little here and there is probably okay at small doses and for valid reasons, but it all adds up over a lifetime. He didn't mention how much X-ray radiation this puts out in the first half or so of the video. I worry it's more than a medical photograph as it's continuous high sample rate video, but I'm not an expert. Would be curious to know. Also curious about the shielding and leaking. Don't damage your DNA if you don't have to. This is a cool, semi-educational video. I don't think I'd take the same risk though.

zzlk 1y ago
I think it's probably not true that it all adds up over a lifetime. That's the model that is widely used but there's some good evidence to suggest it's not accurate. see: https://en.wikipedia.org/wiki/Linear_no-threshold_model https://en.wikipedia.org/wiki/Linear_no-threshold_model
eig 1y ago
Be rest assured Ben’s previous job was in the medial imaging industry. While he worked on MRI machines rather than ionizing radiation, I think he’s very well aware of the dangers of X rays and has many projects dealing with ionizing radiation. There’s a lot of bad safety science YouTubers, Ben isn’t one of them :) Funny thing: it’s actually rare to get radiation damage to human hands and feet since there’s not too much growing tissue there!
- scythe 1y ago
  >there’s not too much growing tissue there! On the contrary, I was told stories in school that old IR doctors used to lose the hair on their hands after using the fluoro for years. The fingernails are also radiosensitive. The main reason that X-rays of the hands and feet are usually very low risk is because the beam intensity (dose) required to penetrate the small amount of tissue is very low. Because the video uses a high-sensitivity detector (photon counter) the dose may be even less than usual. However, it would still be a regulatory violation if you did it in a hospital.
scythe 1y ago
>I worry it's more than a medical photograph as it's continuous high sample rate video, but I'm not an expert. Would be curious to know. Typical fluoro skin entrance exposure rate to go through someone's hand is on the order of 1-3 millisieverts per minute. With a more advanced detector (like this one) it may be lower, but increasing the frame rate or resolution will tend to require higher dose rate. The associated risk of skin cancer is quite small. But please be aware that unnecessary X-Ray imaging of living things is against the law in most jurisdictions.
- SoftTalker 1y ago
  I've wondered how much the average X-ray dosage dropped since they stopped using film and switched to detector-based imaging. Anyone know?
  - scythe 1y ago
    You can compare diagnostic reference levels (DRLs) from different years for the same exam; these correspond to the 75th percentile patient dose for a particular imaging study. It can be a little annoying to find older DRLs (I just tried). NCRP 172 (published 2015) provides a direct comparison of digital to film for some examinations. For upper GI fluoroscopy without oral contrast the DRL for film is 3.9 mSv and for digital is 1.5 mSv. I think this is roughly typical.
rdtsc 1y ago
> He didn't mention how much X-ray radiation this puts out in the first half or so of the video. I worry it's more than a medical photograph as it's continuous high sample rate video, but I'm not an expert. Would be curious to know. > Also curious about the shielding and leaking. If you'd watch the video the last part explains exactly that. The energy levels involved, how the beam is stopped, how backscatter is checked. He is definitely no dummy who just found an x-ray machine in a dump and just powered it up for clicks and giggles.

bkraz 1y ago

I may have the opportunity to use a much higher framerate camera in the future. Please leave suggestions for things that are fairly transparent to x-rays (no metal, or very thin small internal parts), and demonstrates the utility of high speed capture (typically things breaking, bouncing, shattering). The last requirement is that the object be very small, just 1 to 2 cm long. Since learning about ptycography, I may try that as well since x-ray ptycography is one of the real world uses for these sensors.

s0rce 1y ago
I thought the insect breathing work at the synchrotron was cool https://www.science.org/doi/10.1126/science.1078008 https://www.science.org/doi/10.1126/science.1078008
jcims 1y ago
I wonder if you could blow some fine mesh aluminum/iron powder through a small whistle to see any of the airflow effects. Spurious xray emission from triboelectric effect. Phase change of various materials changing absorption characteristics. Capacitor discharge.
shellfishgene 1y ago
This may be a bit hard, but jumping insects like locusts and crickets?
Python3267 1y ago
Here's an interesting one, if the equipment is sensitive enough you could test if you can see the difference in air densities as a .22 caliber bullet passes by.
SweetSoftPillow 1y ago
Hummingbird
SweetSoftPillow 1y ago
Brookesia nana aka nano-chameleon.
1y ago

lacoolj 1y ago

Not only is this incredibly cool, but it bridges my knowledge of x-ray (CR/XR, in this case, DR) even further. I work for a medical imaging company and we actually sell and promote our digital radiography machine that will do this, but for humans (at a much, much slower rate, but enough to be useful in medical diagnostics). I had never considered an application of use outside of medicine. So, my day is now accounted for. In addition to "overdue training" on things I already know, I will be youtubing all the x-ray, MRI, ultrasound fun things that you would never see otherwise. Thank you OP!

cycomanic 1y ago
Please post a link to your YT channel so we all can take part in the fun.

rdtsc 1y ago

That's wild they have a high speed 8K photon counting device for x-rays now. It's not the one he tested but it's the latest one the company has https://www.dectris.com/en/detectors/x-ray-detectors/eiger2/ https://www.dectris.com/en/detectors/x-ray-detectors/eiger2/ It's also neat how he describes how the acquisition works: 32bit tiffs, pci card, real-time linux, and a special rack-mounted server.

world2vec 1y ago

One of the most underrated channels on YouTube.

pmontra 1y ago

A Motif interface from the early 90s at about 10:00 in the video. That X-Ray detector has probably a very long history.