Computed Tomography (CT)
Computed Tomography (CT) scanning is a well-established process that has been widely used many decades. Essentially, it is a three-dimensional X-ray passed through an object slice by slice. A source generates the X-Ray and a collector takes in the signal for processing. For each slice an X-Ray source is either rotated around the object with a collector on the opposite side also rotating, or there are multiple sources and collectors distributed around the object.
The different absorption rates of the materials inside and on the surface of the object will be received as different levels in the signal received by a source. Bones will typically look different than a muscle. Each slice is then like a poker chip of whatever thickness you select to scan at. The more slices, the more time and memory involved, but also the thinner the poker chips and the more accurate the resulting models.
Typical medical CT systems will do a maximum of 0.5 or 1.0 mm slices. Micro CT systems can do slices down to 7 microns. These poker chips are then exported as individual but associated files and these must be merged using expensive software into three-dimensional solid models. This can be a complex process as equivalent structures on each slice must be associated and used to build each element in the final product, which includes internal data. For example, if doing a CT of a human leg, the femur will be inside and to generate a whole bone, the part of the cross-section that is the femur must be identified in each slice.
Medical scanners can be relatively weak because the object being scanned is sometimes alive and it is beneficial if it stays that way. Industrial CT scanners, adapted for non-living objects, can use three or four times as much power (a motor will not die if overexposed) and can get much finer resolution. We can also work with data from other medical type scanners (e.g., MRI).