CT Image Production Study Guide

Comprehensive review flashcards covering differential attenuation, linear attenuation coefficients, CT numbers, tube physics, scan modes, parameters, reconstructions, and quality control concepts from the CT Image Production Study Guide.

What is differential attenuation?

Differential attenuation is the difference in how much the body's tissues absorb or deflect (scatter) x-ray photons passing through them. A large portion of photons are absorbed or deflected out of the scanner, while remaining photons reach the detector array. Different tissues attenuate the beam by different amounts (e.g., bone attenuates more radiation than air), which helps create the digital image.

What is the Linear Attenuation Coefficient (LAC)?

The Linear Attenuation Coefficient (LAC) is a numerical value representing the fraction of the x-ray beam absorbed or scattered (attenuated) as it passes through a material. Materials that attenuate more radiation have higher attenuation coefficients. The CT scanner calculates LAC values and converts them into CT numbers (Hounsfield units).

What are CT numbers (Hounsfield units)?

CT numbers, also called Hounsfield units ($\text{HU}$), are numerical values that represent the linear attenuation properties of tissues calibrated to water ($0\,\text{HU}$). Materials that attenuate more radiation than water have positive CT numbers, while those that attenuate less have negative CT numbers.

Will a tissue with a high CT number appear white or black on a displayed CT image?

A high CT number appears white or light gray because it represents greater x-ray attenuation. Low CT numbers appear darker or black (the displayed appearance also depends on selected window width and window level).

What are the reference CT numbers (in $\text{HU}$) for water, bone, fat, and air?

Water: $0\,\text{HU}$; Bone: approximately $+1{,}000\,\text{HU}$; Fat: approximately $-50\,\text{HU}$; Air: $-1{,}000\,\text{HU}$.

How do CT x-ray tubes differ from general radiography x-ray tubes?

CT tubes must withstand significantly greater heat loads because CT requires continuous x-ray exposure for several seconds. They feature thicker anodes, rotate rapidly, use tungsten alloys with higher heat capacities, and can reach temperatures of approximately $3{,}600^\circ\text{F}$. Proper tube warm-up helps prevent anode damage.

What is meant by the intensity of the x-ray beam, and what factors affect it?

Intensity is the quantity or total number of photons in the x-ray beam. Increasing $\text{mA}$ increases photon number; increasing exposure time increases total photons produced; and increasing $\text{kVp}$ in the CT range also increases total photon number. Photon quantity is proportional to patient dose and detector exposure.

What is meant by the energy of the x-ray beam, and how is it controlled?

Energy refers to the penetrating ability (quality) of individual photons. It is controlled primarily by $\text{kVp}$. The CT beam is polychromatic (polyenergetic), containing photons of various energies, with the average energy expressed in $\text{keV}$. Higher-energy photons travel farther through matter and penetrate dense materials more effectively.

What is Automatic Dose Modulation in CT?

Automatic Dose Modulation is a technique in which the CT scanner automatically varies $\text{mA}$ based on the thickness and attenuation of the anatomy being scanned. Thicker anatomy receives more radiation to maintain adequate detector exposure, while thinner anatomy receives less, optimizing patient dose while maintaining image quality.

If the tube current ($\text{mA}$) is doubled, how does it impact patient radiation dose?

Holding all other factors constant, doubling the $\text{mA}$ doubles beam intensity, patient radiation dose, and detector exposure. Decreasing $\text{mA}$ by $50\%$ decreases dose by $50\%$.

What is rotation time in CT?

Rotation time is the total time required for one complete ($360^\circ$) rotation of the x-ray tube around the patient, typically ranging from $0.1$ to $1.0\,\text{seconds}$. Together with $\text{mA}$, it determines tube output per rotation via the formula: $\text{mAs} = \text{mA} \times \text{time (seconds)}$.

How does an increase in rotation time impact patient dose?

Patient dose and detector exposure increase by the exact same fraction as the rotation time increase, assuming other factors remain constant (e.g., increasing rotation time from $0.5\,\text{seconds}$ to $1.0\,\text{seconds}$ at the same $\text{mA}$ doubles $\text{mAs}$, dose, and detector exposure).

If the rotation time is reduced by $1/2$, what adjustment must be made to the $\text{mA}$ to maintain the same photon intensity?

The $\text{mA}$ must be doubled, because $\text{mA}$ and exposure time share a reciprocal relationship (e.g., $100\,\text{mA} \times 0.6\,\text{s} = 60\,\text{mAs}$, and $200\,\text{mA} \times 0.3\,\text{s} = 60\,\text{mAs}$).

What does $\text{kVp}$ control in CT scanning?

$\text{kVp}$ controls x-ray tube potential and beam energy. Increasing $\text{kVp}$ generates higher-energy photons, increases beam intensity, increases patient dose and detector exposure, improves penetration, and decreases subject contrast.

How does increasing $\text{kVp}$ affect patient dose and image contrast in CT?

Increasing $\text{kVp}$ increases patient dose, photon energy, beam intensity, and detector exposure. Because it creates more uniform tissue penetration, it decreases image contrast.

What are the common $\text{kVp}$ stations used in CT?

Common $\text{kVp}$ stations are $80\,\text{kVp}$, $100\,\text{kVp}$, $120\,\text{kVp}$, and $140\,\text{kVp}$.

What is the typical $\text{mA}$ operating range in CT?

The typical operating range listed is $20\text{--}500\,\text{mA}$.

What is the difference between uniform and nonuniform CT detector arrays?

Uniform detector arrays consist of rows that are all equal in thickness, producing very thin slices across all channels but requiring higher radiation dose to control noise. Nonuniform detector arrays have rows of varying thicknesses (thinner in the center, thicker peripherally), requiring less radiation dose but yielding thicker slices with less spatial detail.

How is total beam width calculated, and what is the beam width for a $16$-slice scanner with $1.25\,\text{mm}$ detector rows?

$\text{Beam width} = \text{number of active detector rows} \times \text{row width}$. For a $16$-slice scanner using $1.25\,\text{mm}$ detector rows: $16 \times 1.25\,\text{mm} = 20\,\text{mm}$.

What are the key characteristics of axial, helical, and volume scan types?

Axial: step-and-shoot sequential scanning where data is collected section by section, pausing exposure while the table moves. Helical: continuous tube rotation synchronized with continuous table movement, shortening scan times and breath holds. Volume: the entire organ dataset is acquired in a single tube rotation without table movement using large detector arrays ($250+$ rows).

What is Window Width (WW) and what does it control?

Window Width is the total range of CT numbers ($\text{HU}$) displayed as shades of gray. It controls displayed image contrast. A wide WW displays a broader range of tissues with more intermediate gray levels and less stark contrast, while a narrow WW displays higher contrast.

What is Window Level (WL) and what does it control?

Window Level is the midpoint (center) CT number ($\text{HU}$) of the displayed grayscale range. It controls overall image brightness. Increasing the WL shifts the range toward higher CT numbers, making more lower-attenuation pixels appear black and resulting in an overall darker image.

What is contrast resolution, and what factors influence it in CT?

Contrast resolution is the ability to distinguish between tissues that share very similar linear attenuation coefficients. Factors affecting it include differential attenuation, $\text{kVp}$ (higher $\text{kVp}$ decreases contrast), image noise (excessive noise obscures subtle density differences), reconstruction kernels, and window settings.

What is spatial resolution, and what factors increase or decrease it in CT?

Spatial resolution is the ability to resolve small objects, sharp edges, and small spaces between structures. It increases with smaller DFOV (smaller pixel size), thinner reconstruction slices, thin slices for MPR, and edge-enhancing algorithms (e.g., bone kernel). It decreases with larger DFOV, thicker slices, high pitch, and reconstructing MPRs from thick slices.

What is image noise in CT, and what factors increase or decrease it?

Image noise is the mottled or patchy appearance of an image primarily caused by quantum mottle (inadequate detector exposure). Noise increases with reduced $\text{mA}$, shorter rotation times without $\text{mA}$ compensation, thinner slices, lower photon flux, and edge-enhancing filters. Noise decreases with increased $\text{mA}$, longer rotation times, higher $\text{kVp}$, thicker slices, smoothing algorithms, and iterative reconstruction.

What are the roles of the pre-patient and post-patient (detector) collimators?

Pre-patient collimators are positioned before the patient to shape the beam, restrict exposure, and match the beam width to the active detector rows. Post-patient (detector) collimators sit between the patient and detectors to shield inactive rows, reduce incoming scatter radiation, and improve subject contrast.

What are the roles of bowtie filters, the DAS, and the array processor?

Bowtie filters shape beam intensity across the field to attenuate beam edges and match patient contour. The Data Acquisition System (DAS) measures remnant detector signals, amplifies them, and converts them to digital raw data. The array processor is a dedicated high-speed computational processor that performs rapid mathematical reconstructions.

How do the axial, sagittal, and coronal planes relate to the X, Y, and Z Cartesian axes in CT?

Axial plane: divides the body into superior/inferior portions along the Z-axis. Sagittal plane: divides the body into right/left portions along the X-axis. Coronal plane: divides the body into anterior/posterior portions along the Y-axis. The CT isocenter is the intersection/midpoint of the X and Y axes.

What is the Scan Field of View (SFOV)?

SFOV is the circular area within the gantry opening from which raw transmission data is collected. Anatomy situated outside the SFOV cannot be recorded, reconstructed, or recovered. Technologists choose between preset sizes such as $25\,\text{cm}$ (small) and $50\,\text{cm}$ (large).

What is the Display Field of View (DFOV)?

DFOV is the targeted section of acquired raw data selected by the technologist to be reconstructed into an image matrix for display. Changing the DFOV alters pixel size ($\text{Pixel size} = \text{DFOV} \div \text{matrix dimension}$). Decreasing DFOV yields smaller pixels and higher spatial resolution.

What is retrospective reconstruction?

Retrospective reconstruction is the process of using saved raw data after scanning is complete to produce new images with altered parameters, such as modified slice thicknesses, slice intervals, DFOV, reconstruction kernels, or start/end anatomical positions.

What is prospective reconstruction?

Prospective reconstruction is the initial image generation programmed prior to initiating the scan. The scanner automatically produces the specified image sets (e.g., standard soft tissue and lung algorithms) directly as data acquisition occurs.

What is interpolation in CT imaging?

Interpolation is a mathematical estimation technique used in helical CT to calculate unknown values from known data points above and below a plane, allowing the continuous spiral dataset to be reconstructed into individual flat, axial slices.

How is pitch defined in helical CT scanning?

Pitch is the ratio of table travel distance per complete tube rotation to the total x-ray beam width: $\text{Pitch} = \text{Table movement per rotation} \div \text{Beam width}$.

What does a pitch of $1$ indicate?

A pitch of $1$ means the patient table moves a distance exactly equal to the beam width during each $360^\circ$ rotation of the x-ray tube, resulting in contiguous data collection with no overlap and no missing anatomical views.

What does a pitch of $< 1$ (oversampling) mean?

A pitch of $< 1$ means the table travels less than the beam width per rotation, causing helical loops to overlap. This collects duplicate projection data over each section, improving z-axis resolution at the expense of increased scan time and higher patient dose.

What does pitch undersampling ($\text{pitch} > 1$) represent?

Undersampling occurs when pitch is greater than $1$: the table moves farther than the beam width during each rotation. Gaps exist between spiral loops, shortening total scan time and reducing dose, but slightly degrading spatial resolution.

If the beam width is $5\,\text{mm}$ and the pitch is $1.5$, what is the table movement per rotation?

$\text{Table movement} = \text{beam width} \times \text{pitch} = 5\,\text{mm} \times 1.5 = 7.5\,\text{mm}$ per tube rotation.

What is the reconstruction slice interval?

Reconstruction slice interval is the spacing or center-to-center distance from one reconstructed image to the next along the z-axis. When it matches slice thickness, slices are contiguous; overlapping intervals (interval smaller than thickness) are preferred for multiplanar reformations.

What is Multiplanar Reconstruction (MPR)?

MPR is a post-processing technique that takes a contiguous stack of axial image data and reformats it into coronal, sagittal, or oblique planar perspectives without subjecting the patient to additional radiation exposure. Slices should be thin to prevent step-like blurring.

What is temporal resolution and how is it improved in cardiac CT?

Temporal resolution is the ability of the imaging system to freeze the motion of moving structures (such as cardiac anatomy). It is improved through faster gantry rotation speeds, half-scan reconstruction algorithms ($180^\circ$ of data), and EKG gating to synchronize data acquisition with cardiac cycles.

What is convolution (reconstruction kernel/filter)?

Convolution is a mathematical filtering algorithm applied to raw projection data prior to backprojection to enhance specific image characteristics. Bone (edge-enhancing) algorithms accentuate high-spatial-frequency structures, while standard/soft-tissue smoothing algorithms optimize low-contrast resolution.

What is Filtered Back Projection (FBP)?

Filtered Back Projection is an analytic reconstruction technique that projects attenuated ray profiles back across a coordinate matrix, using mathematical filtration kernels to remove the characteristic star-like blurring artifact inherent to simple backprojection.

What is partial volume averaging?

Partial volume averaging occurs when multiple tissue types with different linear attenuation values occupy the same individual voxel. The computer averages their attenuation into a single CT number, which can obscure small structures or falsely alter tissue density. Using thinner slices minimizes this artifact.

What is the purpose of the Region of Interest (ROI) tool?

The ROI tool is a graphic cursor placed over a specific area of a reconstructed image to measure diagnostic metrics, such as the mean Hounsfield unit ($\text{HU}$) value and standard deviation, or to evaluate scanner calibration and CT number uniformity across phantoms.

What is a pixel, and how is pixel size calculated?

A pixel (picture element) is a two-dimensional square cell within a digital display matrix assigned a single CT number and grayscale shade. Pixel size is calculated as: $\text{Pixel size} = \text{DFOV} \div \text{matrix dimension}$.

What is a voxel?

A voxel (volume pixel) represents a three-dimensional volume of patient tissue. Its width and height correspond to the two-dimensional pixel dimensions, while its depth corresponds to the reconstructed slice thickness.

What is an image matrix?

An image matrix is a two-dimensional grid of rows and columns composed of individual pixels that forms the digital image. A standard CT matrix size is $512 \times 512$, totaling $262{,}144\,\text{pixels}$.

What is the operational difference between raw data and image data?

Raw data consists of unprocessed attenuation measurements straight from the detector array and DAS that cannot be visually interpreted; it is stored temporarily for reconstruction. Image data consists of processed, reconstructed pixel values organized in a matrix that can be displayed, manipulated, and exported to PACS.

How is a CT Accuracy Quality Control test conducted?

The accuracy test involves scanning a standardized water-and-air phantom to confirm baseline calibration. Pure water must measure $0\,\text{HU}$ and air must measure $-1{,}000\,\text{HU}$. Deviation requires service recalibration or a routine air calibration run.

What does CT Linearity test?

CT Linearity evaluates whether the relationship between a material's linear attenuation coefficient and its assigned CT number ($\text{HU}$) remains directly proportional. It is verified by scanning a phantom with inserts of known densities and plotting the resulting CT numbers to ensure a straight line.