Computed Tomography Image Quality and Dose Management

How to Use This Resource

Table of Contents and List of Topics

Computed Tomography Image Quality
and Dose Management

Mind Map

Outline
and
Learning Guide

Objectives
and
Learning Activities

Visuals for
Class Discussions

Self-study
Module

Online
Textbook

Self-study Module

To begin module,
CLICK HERE.

The Five Principle Characteristics of CT Images

BACK

Like all medical images, CT images have five principle image quality characteristics:
Contrast Sensitivity
Detail (Blurring)
Visual Noise
Spatial Characteristics (Views, FOV, etc)
Artifacts

Image quality is determined by a combination of factors, including:
1. Development and design that has changed over the years
2. General performance and maintance of the equipment
3. How the equipment is operated.

Image Quality Depends On The Selection of Protocol Factors

How the equipment is operated is what accounts for the greatest variation in Image Quality.

Appropriate protocol factor values should be selected to optimize the image quality characteristics for specific clinical objectives.

Optimizing image quality is the process of achieving a balance among the various characteristics (such as detail and noise) and adjusting the image quality to appropriate levels in order to manage the radiation dose to the patient.

The Factors That Affect Blurring and Image Detail

Blurring which effects image detail occurs both in the scanning and image reconstruction phases.

The amount of blurring is determined by a variety of factors, some are fixed by the design of the equipment, but many are adjustable protocol factors.

During the scanning phase the focal spot size and the detector dimensions determine the size of each ray within the x-ray beam. Small rays produce scan data with "better detail".
Increasing the pitch has the effect of reducing the data detail in the direction of patient motion.

All anatomical detail within each voxel is "Blurred together" and represented by one CT number value. Therefore, small voxels produce images with less blurring and better detail.

Some of the filter algorithms (like used to reduce noise) blur the image.

The Dimensions of Each Ray Is Determined by the Size of the Detectors

The size of the detectors have an effect on the size of the individual rays within the x-ray beam.

It is not possible to have a ray or x-ray beam width that is smaller than the detectors. Therefore, detector size will generally limit the detail that can be achieved.

Spiral Scanning

Spiral scanning moves the body through the x-ray beam in a continuous manner.

This creates a continuous data set that can later be sliced in many ways during the reconstruction phase.

The selected pitch value determines how fast the body is moved and how much the x-ray beam is "spread" over the body.

Increasing the pitch value can limit the detail that can be achieved in the direction of movement.

Increasing the Pitch Increases the Effective Width of the Beam and the Associated Blurring.

The detail that is required, in the slice thickness direction, limits how much the pitch can be increased.

The Filter Algorithm Used To Decrease Noise Also Blurs The Image.

Digital image processing generally reduces noise by blurring the image.

Therefore, when the noise reducing filter is selected, it will reduce image detail.

The Factors That Affect Noise in a CT Image

There are several adjustable protocol factors that have an effect on the image noise.

Reducing voxel size (to increase detail) increases the noise because fewer protons are absorbed or captured in each voxel.

Noise can be decreased by increasing the MAS, but this increases the dose to the patient.

The selected filter algorithms can either decrease or increase the noise, it depends on which is selected.

The Calculation of CT Numbers from Tissue Attenuation Coefficient Values

The CT number for each pixel is calculated from the attenuation coefficient values measured for each voxel by the CT process.

Photons Absorbed in Tissue Voxels

The (quantum) noise is determined by the number of photons captured or absorbed in each tissue voxel. This is directly related to the dose to the tissue.

Reducing the Dose Increases the Noise

The noise is produced by the random variation or difference in the number of photons from one voxel to another.

We recall from a previous module (Image Noise) that the statistical variation increases as the number of photons, exposure and dose, is decreased.

Noise Results From the Random Variation in the Number of Photons Absorbed in the Individual Voxels

Small voxels, as used for better detail, capture less photons and result in more noise.

Random Distribution of CT Numbers for a Region in an Image of Water

This represents an area (Region of Interest) within an image of water.

Here we see that all of the pixels do not have a CT number value of zero, as might be expected. This variation in the CT number values is what we see in the image as noise.

The amount of variation, or spread, can be calculated and expressed by the statistical parameter, Standard Deviation (SD).

All CT machines are programmed to calculate the SD within a ROI setup by the operator.

This makes it easy to measure the level of noise in CT images.

Scanning a Water Phantom Produces an Image Showing the Noise

The procedure for measuring noise in CT is to:

scan a water phantom (a plastic milk jug works fine)

setup a region of interest (ROI)

select Standard Deviation from the CT machine functions.

Selecting The Filter Algorithm to Enhance Detail Increases the Noise

Noise in an image is actually an undesirable relatively high detail characteristic. Therefore, when images are processed to increase or enhance detail, the processing also increases the visibility of the noise.

Dose is the Radiation Quantity that Expresses the Concentration of Radiation Energy Absorbed in the Tissue

The radiation dose to patients undergoing CT procedures is an issue that must be considered. The protocol factors should be adjusted to produce the necessary image quality without delivering unnecessary radiation to the patient.

As we now consider the process of achieving this, let's recall that dose is the quantity that expresses the concentration of radiation energy absorbed in a specific tissue location.

The CTDI is the Special Quantity Used to Express Radiation Dose in CT

The Computed Tomography Dose Index is the special quantity used to express radiation dose in CT.

One reason it is used is because it is not easy to measure the actual dose delivered to the internal body regions.

The CTDI can be measured with a dosimeter inserted into a phantom that represents a patient's body. The dose is measured by scanning one slice, but the CTDI is defined and calculated to account for contributions from scattered radiation that occur when multiple slices are scanned.

When the appropriate factors are applied to convert the measured phantom CTDI to an actual patient scan, the CTDI is a reasonable estimation of the actual dose to the patient.

The Total Radiation Energy Deposited in the Body Depends on the Volume of Tissue Scanned

A CTDI value expresses the concentration of radiation to a specific tissue location. It tells nothing about the total radiation energy delivered to a patient.

The total radiation delivered is proportional to the volume of tissue scanned.

The Dose-Length-Product is a Practical Quantity for Expressing the Total Radiation Energy Deposited in the Body

The Dose-Length-Product is a Practical Quantity for Expressing the Total Radiation Energy Deposited in the Body.

If the CTDI for a specific procedure is 3 rad, and we scan a length of 25 cm, the DLP will be 75 rad-cm.

Decreasing Slice Thickness (To Improve Detail) Will Require an Increased Dose to Maintain the Same Noise Level

When the slice thickness is decreased to improve image detail, the noise level will increase because of the smaller voxels as we have just observed.

If the MAS is then increased (either manually or automatically by the equipment) to maintain the same noise level, the radiation dose will be increased.

This is why thin slices should only be used when necessary from a clinical perspective.

This is good dose management.

Increasing Pitch Will Decrease Dose but Can Also Increase Blurring and Reduce Detail

Changing pitch has several effects.

Two general advantages of increasing pitch are:
1. faster scanning
2. reduced dose (the radiation is less concentrated)

The major limiting factor associated with increased pitch is reduced image detail in the direction the body is moved.

Several Types of CT Image Artifacts

Artifacts in CT can be of several forms.

Some are streaks as produced by metal and patient motion.

Others are in the form of regions with incorrect CT numbers.

The beam harding artifact is generally corrected by processing during the reconstruction process.

The partial volume artifact occurs when a voxel contains two very different materials, like bone and soft tissue. The resulting CT number will be somewhere between the correct values for the different materials, but not correct for either. Depending on how the window is set, a structure such as bone, can appear either thinner or thicker than it's actual dimension.

23	End of Module To return to the beginning, CLICK HERE

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

Notes.....

NEXT

Notes.....