X-ray/CT technique

  • X-ray

  • CT
     

X-ray

X-ray tube

X-ray beams are generated in a vacuum x-ray tube. In the x-ray tube are a negatively charged cathode and a positively charged anode. The cathode generally consists of a tungsten spiral. Passing an electrical current through the cathode will cause powerful heating (≥2200°C) of the spiral. The heating causes emission of electrons. The potential difference between the anode and cathode (= tube charge) will cause the electrons to shoot towards the positively charged anode (= focus/target). 
As the electron flow (=tube current) in the anode decelerates, the kinetic energy of the electrons is converted into x-rays (fig. 1).

Figure 1. X-rays are generated in the x-ray tube. The x-ray beam passes through the body part and hits a phosphorus plate/detector.

The tube charge is expressed in kilovolts (kV) and the tube current is expressed in milliampere (mA).

X-ray image

When an x-ray is made, an x-ray beam leaves the x-ray tube, passes through the body and hits a phosphorus plate/detector. The whiteness (=density) depends on the amount of X-ray radiation passing through the tissue (fig. 2) .

Figure 2. X-ray densities (=whiteness).

The more X-rays are obstructed (absorption or dispersion) and do not reach the phosphorous plate/detector, the denser (=whiter) the image. Highly absorbent materials, such as metal, will be imaged as dense. Another example: X-rays pass more easily through the air-filled lungs (black) than bone (white). The information received on the plate is converted into a digital image.
Correctly imaged, an X-ray provides information on the ossal structures, fluid, air, soft tissue contours and prostheses/osteosynthetic material.

Comments:

  • Each X-ray is evaluated as if you are standing in front of the patient; so the right side of the image is the patient's left side and vice versa. 
  • Importantly, the X-ray beam has a divergent property. This means it widens as the distance to the X-ray tube increases. A drawback of this phenomenon is that tissues/structures farther from the plate are imaged larger. This is relevant when evaluating the size of the heart in a chest x-ray (see fig. 3).

Figure 3. Effect of the divergent x-ray on the size of the heart (a = posterior-anterior technique, b = anterior-posterior technique).

Indication/request form

Despite the fact that CT (computer tomography) or MRI (magnetic resonance imaging) generally provides more information on the ossal structures and soft tissues, conventional x-rays do have a number of benefits. An x-ray is a relatively quick, non-expensive and non-invasive technique. A chest x-ray, for example, can quickly provide much useful information in the trauma center. Also think of imaging prostheses/osteosynthesis material, which frequently generate undesired artifacts in CT and MRI.    
In order to assess an x-ray, it is important that the radiologist receive adequate information about the patient and the request.  Relevant history (surgery/treatments and malignancies in particular), relevant clinical information (including fever, location of pain symptoms and trauma mechanism) and a specific and clear request are essential to an adequate radiologic assessment. Without the above, certain findings on the x-ray may be interpreted incorrectly.  Additionally, it always helps for the radiologist to be directed to the actual problem in order for him/her to pay specific attention to this (especially if there are subtle abnormalities). 

CT

CT stands for computer tomography and like conventional x-rays uses x-ray beams. The CT scan is a test where the interior of the human body is imaged in three dimensions. 
The x-ray beam passes through the human body in a thin axial slice, which is repeated in various directions (fig. 4) .

Figure 4. General technique of a CT scanner.

The detectors on the opposite side measure the radiation transmission through the patient.  This enables the computer to determine the degree of absorption in very small volume elements, the so-called voxels. The size of the voxel depends, among other things, on the matrix size (number of pixels) and slice thickness. The information on the voxels is then converted into ‘CT numbers’, better known as Hounsfield unit (HU). More about this later.
When interpreting a CT scan, you should picture yourself standing at the patient's feet looking at his or her head; top is the side of the abdomen, bottom the side of the back (examination table) and left & right have been inverted. 

Nowadays 3rd-generation CT scanners are mostly used (1st, 2nd and 4th-generation CT scanners will not be discussed in this course). In 3rd-generation CT scanners, the x-ray tube and detectors synchronously rotate around the patient.  The detector row covers the full width of the fan-shaped x-ray beam (fig. 5).

Figure 5. Third-generation CT scanner. 

Multislice CT

The development of the multislice CT scanner (also known as multidetector CT or volume CT) significantly reduced scan time. As opposed to standard systems, multislice CT uses multiple detector rows. In this setting, not just one slice is scanned per rotation, but multiple slices simultaneously (fig. 6).

Figure 6. Single-slice versus multislice.

Depending on the number of detector rows, we use the term 4, 6, 8, 10, 16 or 64-slice CT (the term multidetector CT is also used regularly in the literature). Significant benefits of multislice CT include shorter scan time (with fewer movement artifacts), thinner slices and longer scan range in CT angiography tests.

Spiral CT

In conventional CT technique, first a slice is made of the desired area, after which the table moves up a little.  In this way the patient is imaged slice by slice (step-by-step).
Around 1990, the ‘slip ring’ technique was developed where the x-ray tube and detector ring rotate and continue scanning without interruption.  This led to the so-called spiral CT where the scanner table moves with constant speed through the ring with the rotating x-ray tube and detectors. This generates a helix/spiral-shaped pattern (fig. 7).

Figure 7. Spiral CT technique.

The significant benefit of spiral CT is the shorter scan time. The patient can be scanned during one breathing instruction.  Another benefit are the overlapping intervals, improving the visualization of small lesions and counteracting the undesired partial volume effect.  In the partial volume effect, two different structures are located in the same voxel; the average of both densities will then be converted into a gray tone (particularly thicker slices). Very subtle abnormalities will have only negligible impact on mean density, rendering them undetectable.
A drawback of spiral CT is the longer time needed to create image reconstructions.  New CT scanners, however, are becoming increasingly fast in processing the information obtained. 
Another drawback of spiral CT are the specific spiral CT artifacts.  

The spiral technique is used frequently in the above-described multislice CT scanners.  In some cases conventional step-by-step technique is still used, as in HRCT tests or intervention procedures.  HRCT stands for high resolution CT and is used to create thin CT slices (1-2 mm) of the chest which are reconstructed with high resolution and powerful enlargement (this is used for interstitial lung disease in particular).

Contrast agents

Contrast agents improve the imaging of an organ or vessel. The contrast agent is injected intravenously (usually through a vein at the front of the elbow) and then spreads through the circulation to the entire body.
When using an intravenous contrast agent, the question to be answered is important. For adequate assessment, the contrast agent should be located in the target organ/vessel. 
Example: when asked to confirm pulmonary embolisms, the scan is made at the moment the contrast agent is in the pulmonary artery (9-15 sec following injection, fig. 8). If you are more interested in the status of the carotids, you will need to scan 16-24 seconds after injection.

Figure 8. High contrast in the pulmonary artery (through elbow – superior caval vein – right side of the heart). Note the contrast agent has not yet reached the aorta.

Oral and rectal barium contrast can be used to evaluate the intestines and help distinguish intestines from surrounding tissues. 
Iodinated contrast fluid may damage the kidneys. Hydration is essential to prevent contrast nephropathy. Refer to the protocol of the hospital where you are working for more details on preventive measures (including pre/posthydration) and the risk factors. 

Hounsfield units

The degree of x-ray attenuation depends on the tissue type. These differences are converted into ‘CT numbers’, better known as Hounsfield units (HU). A spectrum of gray tones is generated, from -1000 to +3000 (note: the upper limit is determined by the scanner type). Tissues with low attenuation (such as air and fat) have a low HU number. Tissues with high x-ray attenuation (such as bone and contrast fluid) have a high HU value (fig. 9). Water has an HU value of 0.

Figure 9. Hounsfield units (HU) of various tissues.

Humans can distinguish a limited number of gray tones only.  If the entire spectrum of figure 9 were to be imaged, many structures cannot be distinguished. In order to increase contrast between tissues with similar HU values, a certain part of the spectrum can be enlarged as it were. 
The upper and lower limit of the enlarged section is termed the window width. The middle of the window width is the window level
A frequently used option is the soft tissue setting. The soft tissue setting generally has a window level of 40 (note: soft tissues have an HU value around +40 - +80) and a window width of 400 (fig. 10). Everything above the window width upper limit, in this case +240, is projected as white. Everything under the lower limit of the window width, in this case -160, is projected as black. A soft tissue setting therefore provides virtually no information about the air-filled lungs (HU of lungs around -500).

Figure 10. Soft tissue setting with a window width of 400 and a window level of 40. Note that all structures above +240 and under -160 are projected as white and black respectively. 

Other frequently used settings include the lung setting and bone setting (fig. 11/12). These standard settings can usually be activated with a programmed key on the keyboard. 
Note that subtle density differences are best distinguished with a narrow window width.

Figure 11. Lung setting with a window width of 1500 and window level of -650. Note that all structures above +100 and under -1400 are projected as white and black respectively.

Figure 12. Bone setting with a window width of 2000 and  level of +400. Note that all structures above +1400 and under -600 are projected as white and black respectively.

Note: changing the window/level is a software manipulation. Patients therefore do not need to be scanned again in order to change the window level. Window leveling can be used in each test. Be aware that window leveling has its limits in terms of image quality and evaluation.  A CT scan focusing on a certain organ will not be ideal for assessment of another organ (note: the test is based on the question to be answered!). The scanner/technique, scan protocol and kV/mA settings impact the final result.

Sources

  • M. Prokop; Spiral and Multislice Computed Tomography of the body (2003)
  • A. Lemmens; Praktische radiologie (april 2005)
  • J. Rydberg, et al; Multisection CT: Scanning Techniques and Clinical Applications. RadioGraphics 2000. 
  • P. Allisy-Roberts, J. Williams; Farr’s Physics for Medical Imaging (sec. edition 2008)

Author

  • Annelies van der Plas, resident radiology LUMC.

17/03/2014 (translated 23/08/2016)