How does a micro-CT scanner work?
What is Micro Computed Tomography (micro-CT)?
Micro-CT, and the higher resolution Nano-CT, is like having X-ray vision, only better. It allows you to see the inside of something without having to destroy the object itself. What we typically think of as X-ray vision is similar to planar X-ray images that you get in a hospital when you break an arm. Micro-CT / Nano-CT are more like the medical CT systems where you get slice-by-slice information, but without having to cut up the sample.
With 2D X-ray systems you can see what is in an object, while with 3D X-ray systems, such as the Micro-CT, you can see where those things are located. This is useful for nondestructively visualizing and analyzing the internal structure of materials (composites, metals, bones, soft tissues, geological cores, manufacture objections, etc) and living animals (mice, rats, rabbits, etc).
General Basis of MicroCT
At Micro Photonics, our Micro-CT experts help researchers to select and customize an appropriate instrument for their labs. We provide information, instruments, training, and support to help advance research using these transformative technologies. For more information on our products, lab services, and support, visit www.microphotonics.com/products/micro-ct/.
The following provides a broad overview of Micro-CT:
MicroComputed tomography is an X-ray transmission image technique. X-rays are emitted from an X-ray generator, travel through a sample, and are recorded by a detector on the other side to produce radiograph (known as projection image- Figure 1). The sample is then rotated by a fraction of a degree and another projection image is taken at the new position. This procedure is iterated until the sample has rotated 180 or 360 degrees producing a series of projection images.
The projection images are then processed using computer software (typically based on a modified Feldkamp Cone Beam reconstruction algorithm) to show the internal structure of the object nondestructively. This series of images is typically called the reconstructed images or cross sections (figure 2).
The reconstructed images can then be taken and modeled into 3D volumetric objects for quantitative analysis or visualization (Figure 3).
The process of Acquiring a MicroCT Scan
If you would like to see how Micro-CT can support your research, we invite you to submit a sample to our lab for a FREE first scan.
If we look at the Micro-CT process in more depth and examine some of the physical aspects, the process can be expanded to four steps:
1) Generate X-rays
2) Transmit X-rays Through the Sample
3) Rotate the Sample to Acquire a Series of Projection Images
4) Reconstruct the Projection Images into Virtual Slices
Let’s first start by examine the X-ray source. For laboratory CT systems, X-rays are generated by directing electrons produced in a cathode at a target (Tungsten, copper, etc). The target emits x-rays which are then transmitted to the sample. Below is a depiction of the basic anatomy of an X-ray source. For more details on how X-rays are generated see (How Are X-rays Generated). The finer the electron beam can be focused on the target, the smaller the ‘spot’ size of the x-rays will be which leads to higher resolution images.
The emitted X-rays are in the shape of a cone where the origination point is a spot on the target and the emitted beam diverges out in a conical shape. Prior to Feldkamp, Davis, and Kress’s famous 1984 publication on Practical Cone-Beam Algorithm, X-ray geometries were mostly limited to point or a fan beam. The advantage of a cone is the ability to capture a larger volume in a single scan rotation. (Why is a Large Field of View Important?)
X-ray Absorption Through the Sample
Micro-CT requires that there are both: Partial Absorption, meaning some X-ray photons are absorbed in the material while others are transmitted to the detector, and Differential Absorption, meaning that different materials within the object have different absorption characteristics to give contrast. If there is no differential absorption, the sample result comes out as a uniform gray level.
The X-rays propagate through the sample where some of the X-ray photons are absorbed and others are transmitted to the detector. The general form of X-ray attenuation is:
I0 = X-ray intensity before reaching object
I1 = X-ray intensity after passing through object
e = the exponential coefficient (2.7182818……….)
μ = the x-ray attenuation coefficient
t = the thickness of the absorbing material, in chosen distance units e.g. mm
The unabsorbed X-rays are recorded by the detector. This produces a single radiographic image similar to an X-ray you would get for a broken bone at the doctor. Just like the doctor’s X-ray, denser material (such as a bone) will absorb more X-rays than less dense material (soft tissue). Thicker materials will also absorb more x-rays, which is why 2D x-ray systems aren’t used for measuring densities, unless they use more than one energy peak.
If samples are too high in atomic number, the x-rays won’t have enough power to pass through the sample and reach the detector. For example, lead stops x-rays so well it is used as a shielding material for the systems but it isn’t useful to scan more and a mm or so of lead in an x-ray system. Also, the sample has to be dense enough though to stop some x-rays otherwise it is transparent. Low z materials such as pure Beryllium can be difficult to image due to their low attenuation rates.
Projection image of a Cricket (Image of the Month January 2015)
Rotate the Sample to Acquire a Series of Projection Images
After a projection image is taken, the sample is rotated a fraction of a degree, typical 0.5 degrees or less. (For in vivo scanning, the X-ray source and detector pair are actually rotated. See What is the difference between in vivo and ex vivo scanning). At each step a new projection image is taken. This is done throughout the 360 degree rotation. 180 degrees can be used to shorten the time of the scan as the projection images from 0-180 degrees are the mirror images of the project images from 180-360 degrees. Typically, if you take finer step sizes, the resultant cross-section will be finer as well.
Reconstruct the Projection Images into Virtual Slices
The process of computing the internal structural information from the projection images is known as reconstruction. This procedure results in a stack of reconstruction images (also referred to as “cross-sectional images” or “slices”). The most prolific reconstruction algorithm is the Feldkamp, Davis, and Kemp (FDK) cone beam reconstruction algorithm which is a form of filtered backprojection (FDP). These cross-sections can then be used to view the internal features, analyzed, reconstructed into virtual 3D models, made into movies, printed into 3D models and more.
Cross sectional images of snow leopard skull (see Image of the Month April 2015)