Volume Averaging

Certain factors affect the accuracy of the Hounsfield units and the accuracy of the image. The degree of linear attenuation and the resultant CT number are determined by the average density of the material within that voxel. Therefore, when a voxel is filled with several structures (or the structure of interest is smaller than the voxel), the voxel will be assigned a CT number or HU that is subject to volume averaging artifact. If the resolution of a very small structure is important, decreasing the scan thickness will improve accuracy. However, the benefits of thinner sections must be considered along with the risk of a higher radiation dose to the patient with the increased number of slices. Scanning protocols are designed to balance image resolution with acceptable radiation dose.

Although thinner sections increase image resolution, some of the advantages are lost due to increased image noise, also known as quantum mottle. Image noise is due to an insufficient number of x-ray photons reaching the detectors, resulting in a grainy image. Image noise can be reduced by increasing the radiation dose, but, again, a compromise between radiation exposure and image noise determines the quantitative x-ray dose.

Window width and window level are display settings that can be manipulated to optimize the appearance of the image. Selecting a window width determines the range of CT numbers that will be represented on a specific image. The computer assigns different shades of gray to CT numbers that fall within the selected range. CT numbers that are above the chosen range will appear white, while numbers below the chosen range will appear black. By increasing the window width, a greater range of CT numbers are each assigned to a shade of gray. This technique is used when it is desirable to view a variety of tissues that vary greatly in density (e.g., lung). The disadvantage to wider windows (400 to 2000 HU) is that subtle differences in density will not be visualized.

When the goal is to visualize a section of anatomy with minor discrepancies in tissue density, narrow window width is chosen (50 to 400 HU). A good example of the use of narrow window width is in the brain, where there is little difference in tissue densities; yet, with the higher contrast achieved with a narrow window, white and gray matter can be differentiated.

Window level determines the CT number that will be the center of the gray scale. The window level is generally set at the same value as the average attenuation number of the tissue of interest.

The image from a single data set can be displayed at different window widths and levels to highlight various tissues of interest. For example, in chest CT, settings that optimize the image of the soft tissue mediastinal structures use a moderate window width (e.g., 350) centered at a level just above water (e.g., +30). However, this setting does not allow visualization of the details of the lungs ( Fig...n 296:2.^.). On the other hand, lung settings use a wide window (e.g., 1400) and a low level (e.g., -600) to center the gray scale so it includes air densities ( Fig..n i296-2.S). To optimize the image of a bony structure, a wide window (e.g., 1500) and a higher level (e.g.,

Peripheral Neuropathy Natural Treatment Options

Peripheral Neuropathy Natural Treatment Options

This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.

Get My Free Ebook


Post a comment