@mustardofdoom

5 days ago

I'm a researcher who has done work on Dual Energy Computed Tomography. Specifically for rock classification in CT scans of drilled rock cores. As soon as you started setting up the problem, I suspected that DECT was how these scanners were achieving their results. It escaped me that airport detectors are now doing CT, rather than standard 2D scans. That's a huge leap in complexity and maintenance. Very impressive the reconstruction of image data is done in real time with the Dual energy calculation. That's extremely advanced! At 4:48, you mention the atomic number. DECT work tends to instead deal with "Zeff" which means effective atomic number (Z). You mention this later at 6:10, but it is not a 1:1 thing. There are very weird X-ray behaviors at high energies (>1 MeV) not intuitive if you think only about atomic number. One thing that surprises me about these machines is that the DECT theory generally assumes a monochromatic X-ray source. Yes, X-rays are polychromatic and have a sort of rainbow of X-rays rather than just one kind. To get a monochromatic X-ray typically requires a linear accelerator or particle accelerator followed by filtering out undesired X-ray intensities. There's just no way these machines do that, as such systems are huge and very expensive. So they somehow solved the problem of standardization to a degree that they feel confident in assigning material identity. That's a big technical challenge that impedes the general use of DECT, so kudos to the Smiths Detection team if they did indeed solve this problem.

2 |