INTRODUCTION — Computed tomography (CT) is an imaging technique that has revolutionized medical imaging. It is widely available, fast, and provides a detailed view of the internal organs and structures [1-3].
The two major types of CT are helical CT and conventional, axial, step-and-shoot CT. Helical CT is most prevalent, but conventional step-and-shoot, axial technique is used for high-resolution CT scanning of the lungs, coronary artery calcium scoring, and prospective ECG-triggered coronary CT angiography.
The technical aspects of CT are reviewed here. The role of CT in various clinical situations is described separately. (See "High resolution computed tomography of the lungs" and "Cardiac imaging with computed tomography and magnetic resonance in the adult".)
GENERAL DESCRIPTION — The principal components of a CT scanner are the x-ray tube and a diametrically opposed array of detectors :
●The x-ray tube rotates around the patient and generates an x-ray beam, which can have a pencil-beam, fan beam, or cone beam configuration (figure 1). The gantry motion is helical during helical CT and circular during conventional, axial, step-and-shoot CT.
●The detectors concurrently record the radiation that traverses the body, while rejecting scattered radiation emanating from outside the x-ray tube focal spot or target of the x-ray tube, so called off-focus radiation, with collimators on the x-ray tube side and on the detector side.
During conventional, axial, step-and-shoot CT, the patient is moved into the desired position (ie, step) and then the x-ray tube rotates around the patient (ie, shoot). These two steps are repeated, with the table inching sequentially along the longitudinal or z-axis while the x-ray production is interrupted, until the scan is complete. In contrast, during helical CT, the rotation of the x-ray tube and the movement of the patient through the gantry (the portion of the CT scanner that contains the x-ray tube and detectors) are continuous. This uninterrupted production of radiation accounts for the much higher radiation exposure during helical acquisition.
Helical CT facilitates the uninterrupted acquisition of raw data. The transmitted intensity of the x-ray beam decreases exponentially as it traverses the patient, due to absorption in tissue. The data acquisition system converts the residual intensity profile in the detector to attenuation values per voxel in three-dimensional volume. These values are processed to projection data after being filtered with either soft tissue (standard) or bone (high frequency) algorithm. These raw data are then projected onto the image domain (filtered back projections) where they are digitized by analog to digital converters and then reconstructed into axial images using array processors and interpolation algorithms . These devices record, digitize, store, and reconstruct 800 to 1500 views per 360 degrees of rotation around the patient. Each view is acquired at a slightly different angle using mathematical algorithms.
The resulting tomographic images are composites of individual volume elements (voxels) displayed as two-dimensional picture elements (pixels). Each pixel has a numerical value that is called a CT number:
CT number = [(AM - AW)/AW] x 1000
where AM is the linear attenuation coefficient of a material (eg, soft tissue, bone, fat) and AW is the linear attenuation coefficient of water. The CT number is expressed in Hounsfield units (HU). The HU scale or attenuation scale is calibrated from -1000 HU for gas to 0 for water and to +1000 HU for bone. The scale can rarely be expanded to +4000 for higher attenuation metals but attenuation values that exceed the typical dynamic range of HU can lead to metal streak artifacts . The physical density of materials correlates closely with the HU and can be calculated with the following formula in mg/mL:
HU + 1000 = Density
A change of 1 HU corresponds to a change of 0.1 percent in physical density.
The window width and window level utilized when displaying images on the workstation determine the window range. For example for display of soft tissues, a window width of 400 and a window level of +50 imply a window range that encompasses +250 to -150.
For the optimal display of the lungs, a window width of +1500 and a window level of -600 yield a window range of +150 to -1350.
In order to determine the relevance of the window width in describing contrast, it is assumed that the eye can easily differentiate 16 shades of gray. Dividing the window width by 16 yields the number of HU encompassed per shade of gray or step. High contrast display with a window width of 400 implies 26 HU per step. Low contrast display with a window width of 1500 entails 94 HU per step. At the workstation, moving the computer mouse to the right increases the window width, thus decreasing the contrast; while moving the mouse to the left narrows the window width and increases the contrast. Moving the mouse down increases the window level towards the bone and high attenuation spectrum (black background) and decreases brightness, while moving the mouse up decreases the window level towards low attenuation and lung (white background) and increases brightness .
After processing or reconstructing the raw data into axial source images, post-processing or rendering to multiplanar reformation, maximal or minimal intensity projections or volume rendering can be performed. (See 'Two-dimensional and three-dimensional rendering' below.)
Configuration — The configuration of CT scanners has changed with the advent of slip rings. The stepwise approach to imaging used by conventional, step-and-shoot, axial CT was previously necessary because gantry cables needed to unwind through reverse rotation of the gantry. However, the invention of slip rings allowed electrical current to be supplied to the scanner and electronic data to be transferred from the gantry to the computer without gantry cables. It also allowed helical CT scanners to be developed, since continuous rotation of the x-ray tube became possible without gantry cables. Most imaging is now done by helical CT scanners, which are also called spiral or volumetric CT scanners (figure 1). (See 'Advantages' below.)
Many institutions are using helical CT scanners with multidetector-rows and multiple channels. Such scanners are also called multislice helical CT scanners. Multislice helical CT scanners use a cone-shaped x-ray beam with wider collimation, which strikes many of the 16 to 320 parallel detector rows during a single scanning cycle [8-12]. Sixty-four slice helical CT scanners are in wide clinical use, while 320 slice helical CT scanners and dual source helical CT scanners have been introduced . (See 'Multislice helical CT' below.)
Dual source CT scanners are capable of either of the following:
●Using two x-ray sources to scan simultaneously generates two data sets at a 95 degree angle offset. This improves the temporal resolution by decreasing the smallest time interval that can resolve a dynamic process with helical CT to 66 msec .
●Using dual energy of 80 kVp and 140 kVp simultaneously allows for spectral imaging:
Subtraction of unwanted calcium or other high attenuation materials, also called material decomposition, is possible. Measured attenuation of materials varies with different incident energy beams and different atomic numbers of scanned substances. This is achieved with different dual-energy ratios (ie, HU at 80 kVp divided by HU at 140 kVp). Iodinated contrast material has a ratio of 600 HU divided by 300 HU and a dual-energy ratio of 2, whereas calcified bone has a ratio of 800 HU divided by 533 HU and a dual-energy ratio of 1.5.
Iodine maps after intravenous injection of contrast material allow for determination of pulmonary perfusion . Dual-energy imaging facilitates the generation of synthetic monochromatic beams with spectral monochromatic imaging, which eliminates streak artifacts and blooming artifacts and allows for increased intravascular contrast when the keV of the monochromatic beam approaches the iodine k-edge (33 keV). Usually, a 50 keV monochromatic beam is used . This allows for a reduction in the injected iodine dose.
So-called water imaging, also called virtual nonenhanced imaging, is possible after electronic extraction of iodine from the vascular lumen . Dual-energy imaging can be achieved with single-source scanners that have the capability of dynamic rapid energy switching every 0.5 milliseconds or with dual-layer detectors, also dubbed detector-based spectral CT. The latter method does not increase the overall radiation dose.
ADVANTAGES — CT is superior to conventional radiography because of its exquisite contrast resolution. CT uses point attenuation, which can detect differences in contrast of less than 0.5 percent. In comparison, conventional radiography uses ray attenuation, which can detect differences in contrast of approximately 5 percent. This represents a tenfold increase in contrast resolution. Another advantage of CT over conventional radiography is that anatomic structures in different planes are displayed free of superimposition.
Helical CT has several advantages over prior generations of CT scanners. Its primary advantage is its speed, obtaining an image by scanning only 180 degrees (ie, half-scan imaging) with a gantry rotation time of 0.25 to 0.35 seconds [17-21]. Consequences of this speed are reduction of motion artifacts, implementation of cardiac imaging, and improved scan coverage during a single breath hold. The helical, volumetric data set eliminates respiratory misregistration, defined as gaps or overlap in images obtained during several respiratory cycles due to variation in the depth of inspirations.
Helical CT can reconstruct high quality, near-isotropic, overlapping axial source images that allow for multiplanar two-dimensional and three-dimensional reformatted images at arbitrary levels within the scanned volume [17-21]. It can also be used for CT angiography because it can rapidly acquire data during the bolus or arterial phase of intravenous contrast material injection.
SETTINGS — Several of the parameters used in helical CT are set by the operator. These include the collimation of the x-ray beam in single-slice CT scanning and the reconstructed section thickness in multislice CT scanning, pitch, field of view, duration of the helix, reconstruction interval, reformation algorithm or kernel, and reconstruction phase in cardiac CT.
Collimation — Collimation refers to the thickness of the x-ray beam. In standard single-slice helical CT, collimation determines the thickness of the section:
●Standard collimation for scanning the chest is 2.5 mm for the mediastinum and 1 to 1.25 mm for the lung parenchyma.
●Narrow collimation (1 to 1.25 mm) is used for CT pulmonary angiography, high-resolution CT scanning of the lung parenchyma, and imaging of small pulmonary nodules. A collimation of 0.5 to 0.6 mm is used for cardiac CT and for preprocedural imaging prior to robotic bronchoscopy.
●Wide collimation (2.5 mm) is used in heavy patients to decrease the quantum noise, although at the cost of decreased spatial resolution.
Collimation determines the thickness of a section in standard single-slice helical CT. However, it is the detector width that determines the minimal achievable thickness of a section in multislice helical CT. Subsequently, the data acquisition system can combine several detector row widths in order to achieve the desired reconstructed section thickness.
Pitch — The pitch represents the table speed per rotation of the x-ray beam (ie, the table feed or table increment) divided by the beam collimation .
●A pitch of 1 is generally used.
●A pitch >1 allows for greater coverage per unit time. The advantages of a pitch >1 are a reduction in the patient's radiation exposure and a shorter acquisition time of the data set. However, it also decreases the z-axis (ie, longitudinal) resolution, which blurs the image. As an example, helical CT with a pitch of 2 increases the perceived slice thickness by 20 to 30 percent compared to conventional CT with a pitch of 1 [5,23,24]. High pitch spiral imaging utilizes a pitch of 3.2 with a table speed of 737 mm/sec for cardiac CT and marked reduction in radiation dose .
●A pitch <1 increases the quality of reformatted images considerably. However, it also increases the duration of the scan and the patient's radiation exposure. In cardiac CT scanning, a pitch of around 0.25 is frequently used with retrospective ECG gating.
Duration of the helix — The duration of the helix is the time that is necessary to scan an entire volume of tissue. As examples, with a 64 slice helical CT, the cardiac structures can be scanned in 5 to 8 seconds, the entire chest in 10 to 12 seconds, and the entire body in 15 to 20 seconds. A 320 slice helical CT can scan the entire heart in less than one second.
Reconstruction interval — The reconstruction interval (also called the slice index) is the distance between reconstructed axial images. It determines the degree of overlap of reformatted sections [5,26,27]. A reconstruction interval of half the detector width is frequently used, according to the Nyquist sampling theorem. This achieves 50 percent overlap, which increases the visibility of small lesions and eliminates partial volume effects.
Gantry rotation time — The gantry rotation time (ie, the cycle or revolution time) is generally not set by the operator. It refers to the duration in which the x-ray beam rotates 360 degrees. It is usually 0.25 to 0.7 seconds. The limiting factor for decreasing the rotation time is the gravitational stress on the equipment, since G-forces in excess of 15 are generated.
BREATHING INSTRUCTIONS — Breathing instructions are important when the chest is being scanned. The goal is for the patient to suspend respiration at the end of a deep inspiration to total lung capacity. This prevents motion artifact and respiratory misregistration (gaps or overlap in breathing cycles due to variation in the depth of inspiration).
Patients should hyperventilate by taking three or four deep inspirations. Then, they should suspend respiration at the end of an inspiration for 10 to 32 seconds. This is called the single acquisition, single breath-hold technique .
For patients who cannot hold their breath long enough to complete a scan, multiple preprogrammed helical scans of 5 to 15 seconds can be performed with breathing intervals of 6 to 7 seconds between scans. This approach is called the split breath hold or variable mode technique (figure 2) . It is only rarely used now, since scan cycles of less than one second and multidetector row CT scanning have decreased the duration of the breath hold necessary for patients to complete the scan.
INTRAVENOUS CONTRAST — Following intravenous injection, contrast material distribution occurs in three pharmacokinetic phases:
●The early vascular or bolus phase corresponds to the duration of the contrast material injection. Arterial enhancement increases gradually and reaches a peak value at the completion of the injection. The rate of the increase is proportional to the rate of the injection, the iodine concentration, and the volume of contrast material. In contradistinction, the rate of increase is inversely proportional to the patient's cardiac output and body weight [26,28].
●The redistribution phase occurs within one to three minutes after the injection of contrast material. During this phase, the contrast agent diffuses from the intravascular to the extravascular compartment.
●During the equilibrium phase, the contrast material in the intravascular and extravascular compartments reaches a dynamic equilibrium. This phase is not desirable for imaging.
For a helical CT of the chest, the goal is to scan the chest during the bolus phase and the liver during the redistribution phase. However, scanning the lung parenchyma can extend into the redistribution and equilibrium phases without any decrement in quality if pleural, hilar, or mediastinal pathology is not expected. If only lung pathology is expected, then scanning should be performed without intravenous contrast material administration in order to avoid potential streak artifacts from vascular structures, so called hurricane artifacts .
Timing the scan with respect to the intravenous injection of contrast material requires that the injection rate, duration of the injection, volume of contrast material, delay time, and scan duration be considered. Low osmolar or iso-osmolar nonionic contrast agents are preferred. They are injected intravenously at a rate of 3 to 7 mL per second, with a delay of 30 to 45 seconds. Shorter delay times are used when performing CT pulmonary angiography or CT coronary arteriography. In such cases, the optimal delay is determined with a timing bolus or with bolus tracking technology . (See 'CT angiography' below.)
The diagnosis and management of adverse effects of intravenous contrast material including hypersensitivity reactions and nephropathy are discussed separately. (See "Diagnosis and treatment of an acute reaction to a radiologic contrast agent" and "Contrast-associated and contrast-induced acute kidney injury: Clinical features, diagnosis, and management" and "Prevention of contrast-associated acute kidney injury related to angiography".)
DATA PROCESSING — Each rotation of the x-ray tube generates data (800 to 1500 views) for an angled plane of section. From these data, interpolation algorithms generate virtual axial displays by interpolating data points below and above the plane of section . A wide algorithm interpolates points separated by a 360 degree rotation, while a slim algorithm interpolates points separated by a 180 degree rotation (figure 3).
IMAGES — The x- and y-axis resolutions (also called in-plane resolution) are determined primarily by x-ray tube focal spot but also by the interpolation algorithm, and voxel matrix configuration. A matrix size of 512 is most frequently used. The pixel size can be computed by the quotient of the field of view diameter in mm and the matrix size (eg, 450 mm/512 = 0.9 mm). New scanners can achieve a spatial resolution of 0.23 mm . The z-axis resolution (also called through-plane or longitudinal resolution) is determined primarily by the section thickness and section sensitivity profile which is a curve that shows the effect of increased section thickness along the patient’s longitudinal axis on spatial resolution in the z-axis. The interpolation algorithm, pitch, and reconstruction interval contribute to the z-axis resolution.
Ideally, the z-axis resolution should be the same as the x- and y-axis resolutions. This is called isotropic imaging and it is important because it allows for accurate multiplanar reformations, devoid of “stair step artifacts.” It facilitates the depiction of findings in any desired plane and in three dimensions.
In multidetector row CT scanners, the number of slices generated per rotation is determined by electronic channels that combine several detector rows to form one slice. Newer helical CT scanners with 64 or more channels, flying spot technology and a detector width of 0.5 to 0.625 mm increase the z-axis resolution, so that near-isotropic resolution is achieved.
TWO-DIMENSIONAL AND THREE-DIMENSIONAL RENDERING — Production of two-dimensional and three-dimensional images is a sophisticated application of helical CT. The images are more dramatic than routine axial images, but the process is labor intensive, requires a dedicated workstation, and includes interpolation, segmentation, and rendering [32-34]:
●Interpolation involves constructing new data points within the range of known data points
●Segmentation is the process of partitioning a digital image into multiple segments thereby allowing structures of little interest to be eliminated or subtracted
●Rendering, post-processing or reformatting is the generation of an image from a model utilizing the axial source images
Two-dimensional — Maximum intensity projection (MIP), minimum intensity projection (MINIP), multiplanar reformation (MPR), and curved MPR are two-dimensional rendering techniques.
●MIP is obtained by orienting algebraic rays along a preselected projection and then determining the maximum attenuation value encountered by the rays [34-36]. The pixels with the highest values are preferentially displayed. The two-dimensional quality of the images produced with MIP is excellent and approaches the quality of conventional angiograms. They are diagnostically more useful than volume rendering or shaded surface display. Their disadvantage is the loss of contrast resolution ensuing from the preferential display of the maximal attenuation in the data set which is brought forward in the image and may cover up other structures in the vessel lumen.
●MINIP is similar to MIP except that it preferentially displays the pixels with the minimum attenuation values encountered by the rays. It provides an excellent image of the trachea and bronchi.
●Curved MPR uses a single voxel plane that is perpendicular to a curvilinear line to longitudinally display a structure, also called center line technology. A volume is generated by stacking the axial source images sequentially. Curved multiplanar images track the center of a tubular structure and the entire length of the tubular structure is then displayed as a single image (image 1). The appearance of the tubular structure's lumen and wall can be excellent, although the quality of the image is dependent upon the skill of the operator or the quality of the software. The surrounding structures are frequently distorted . Newer algorithms automatically render two-dimensional images using both curved MPR and stretched planar reformatting (similar to curved MPR except that it uses a straight line instead of a curvilinear line). Curved MPR is useful for CT coronary arteriography. Centerline technology facilitates rendering and display, and relies heavily on precise placement of the center line.
Three-dimensional — Three-dimensional rendering requires precisely defining the volume that is to be imaged. The volume should be kept small, since the z-axis resolution is inversely proportional to the length of the imaging volume. Several advanced techniques are available to display the volumetric data set in three dimensions once the imaging volume has been defined:
●Shaded surface display (SSD) establishes a threshold value, above and below which pixels are excluded. The remaining pixels are manipulated by the computer to produce images with an excellent vascular contour, although a loss of contrast resolution remains a problem (ie, calcified arteriosclerotic plaques are not visible) [34,37]. SSD has been replaced by volume rendering and is not used anymore.
●Volume rendering (VR) assigns each pixel its own unique opacity and color. These opacity and color assignments are then used to create a three-dimensional image, which displays the surface and interior of a structure with tissue-like effects.
●Virtual reality (ie, perspective rendering or "fly through" imaging) is used for CT bronchoscopy and other endoluminal applications. It gives the viewer the perspective of being within the lumen. The attenuation values for the mucosa or endoluminal surface are highlighted by special algorithms .
CT ANGIOGRAPHY — CT angiography is performed with a helical CT scan that lasts 15 to 30 seconds. The collimation, gantry rotation time, and duration of helix are set as described above (see 'Settings' above). The detector width is generally 0.5 to 2 mm and the pitch depends upon the type of CT angiogram being performed [39,40]:
●During CT pulmonary angiography, the pitch is usually 1.5 to 2
●During CT coronary arteriography, the pitch is usually 0.23 to 0.25, unless high-pitch spiral technique is utilized with a pitch of 3.2
Optimal vascular enhancement with contrast material is essential for high quality CT angiography. In most cases, 100 to 160 mL of low or iso-osmolar contrast material (320 to 370 mg iodine per mL) are injected at a rate of 3 to 6 mL per second. A typical delay for CT pulmonary angiography is 15 to 20 seconds and for CT coronary arteriography is 25 to 30 seconds.
The exact volume and rate of the contrast material injection are based upon the delay time, which can be estimated using either a timing bolus of 20 mL of contrast material injected at 5 mL per second or bolus tracking technology. The optimal delay time can be calculated with a timing bolus and represents the sum of the contrast material travel time (“time to peak”) and the acquisition time needed to scan the volume or organ of interest. Patients who have a short contrast material travel time (ie, a high cardiac output or large central blood volume) are administered a larger volume of contrast material at a higher injection rate. This increases vascular enhancement, mitigating the tendency of a high cardiac output to decrease vascular enhancement. (See 'Intravenous contrast' above.)
The helical data set is reconstructed into virtual axial sections using a reconstruction interval that provides 50 to 66 percent overlap. This facilitates a smooth, three-dimensional display without partial volume effects.
Axial source images followed by two-dimensional multiplanar reformations have the best spatial and contrast resolution. They are also the least susceptible to artifacts. The three-dimensional rendering techniques described above have been used to display CT angiograms, while two-dimensional rendering with retrospective ECG gating or prospective ECG triggering may improve visualization of the coronary arteries, especially with volumetric 64 to 320 slice CT scanners .
MULTISLICE HELICAL CT — Multislice helical CT is a major technical development in helical CT scanning [8-10,42-46]. During multislice helical CT, a cone-shaped x-ray beam (wider collimation than the conventional fan-shaped x-ray beam) strikes many of the 16 to 320 detector rows that are arranged in parallel along the longitudinal axis (ie, the direction of the table motion). The combination of a cone-shaped x-ray beam and multiple rows of detectors allows a larger proportion of the x-ray beam to be used for imaging purposes (ie, the detective quantum efficiency is 11 to 22 percent instead of 5.5 percent), instead of being converted to heat. Multiple channels extract the data that are obtained simultaneously from different anatomic sections.
Benefits — Multislice helical CT scanners acquire data up to 64 times faster than single slice helical CT scanners. This is primarily a consequence of the simultaneous acquisition of 64 slices, since gantry rotation is only slightly faster (gantry rotation times are limited by the G-forces that are generated). Thus, multislice helical CT scanners can cover 64 times more patient length per unit time, the same patient length in a shorter time, or the same patient length in the same time with thinner slices, when compared to the original single slice helical CT scanners. Increasing the pitch can further increase the anatomic coverage or decrease the examination time. The entire chest can be scanned in less than ten seconds with a multislice helical CT scanner.
Multislice helical CT has additional advantages, both technical and clinical, compared to single slice helical CT:
●Its 360 degree reconstruction algorithm reduces signal noise by a factor of 1.4, compared to the 180 degree reconstruction algorithm that is used in single slice helical CT.
●Several thousand image slices can be generated in 60 seconds or less, facilitating the visualization of image stacks. The abundant images are best viewed and manipulated on a PACS workstation rather than conventional film. During stack mode viewing, radiologists simulate motion by scrolling through sequential images that form a three-dimensional data set and search for lesions that stand out from the background. The strategy for viewing the three-dimensional data set can be either scanning every individual section with foveal vision in the x and y planes or drilling while limiting the search to a quadrant and quickly scrolling through sections in the z-axis .
●Faster image acquisition results in fewer motion artifacts. Temporal resolution is not influenced by the number of detector rows, but is directly dependent upon the rotation speed of the gantry (ie, 250 msecs to 350 msecs).
●The in-plane spatial resolution (ie, the ability to distinguish objects that are separate but close together) and z-axis (through-axis) resolution are both improved, thereby generating nearly isotropic voxels.
●Anatomic coverage is increased.
●Contrast enhancement is optimal, with enhanced conspicuity of vascular structures. This is particularly true of CT pulmonary angiography, CT aortography, and CT coronary arteriography.
●Enhanced image processing includes multiplanar reformation, maximum and minimum intensity projection, volume rendering, and virtual reality imaging. (See 'Three-dimensional' above.)
●Retrospective reconstructions can be performed without the need for additional data acquisition. Virtually any section thickness can be reconstructed, as long as the initial data set was acquired with a 0.5 mm to 0.625 mm detector width.
●The diagnostic accuracy of CT angiography is improved, with respect to the detection of aortic dissections, penetrating ulcers, intramural hematomas, aneurysms, coronary imaging, and pulmonary emboli in subsegmental arteries [48,49].
●The detection and characterization of small pulmonary nodules is enhanced [50,51].
●The visualization of airways disease (eg, tracheal imaging, bronchiectasis) is markedly improved.
●Assessment of the lung parenchyma using 1 mm collimation is just as detailed with multislice helical CT as with high-resolution CT. However, high-resolution CT requires only 10 percent of the radiation dose that is required with multislice helical CT. (See "High resolution computed tomography of the lungs".)
●Display of the lungs with multiplanar reformats in the coronal and sagittal planes facilitates better demonstration of the distribution of diffuse lung disease.
Technical parameters — The following parameters are unique to multislice helical CT. Other technical parameters for helical CT were described above. (See 'Settings' above.)
●The number of data channels determines the number of image slices, rather than the number of detector rows (each data channel can correspond to one or several detector rows). With flying spot technology the number of displayed slices can be twice the number of detector rows.
●Detector row width is the minimum thickness of a detector row element eg, 0.5 mm to 0.625 mm.
●Effective detector row thickness is the sum of the width of the detector rows for each data channel. This determines the section thickness, eg, 0.625 mm x 2 = 1.25 mm or 0.625 mm x 4 = 2.5 mm. The summation of the individual detector rows is achieved by the data acquisition system.
●Detector configuration is determined by the number of channels per z-axis and the effective detector row thickness of each channel, eg, 64 x 0.625 mm, 32 x 1.25 mm or 16 x 2.5 mm.
●Collimation is the product of the number of data channels and the effective detector row thickness, eg, 64 x 0.625 mm = 40 mm.
●Table travel speed per second (ie, the table feed) is the product of the collimation, pitch, and rotations of the x-ray beam per second, eg, 40 mm x 0.25 x 3 = 30 mm per second in cardiac CT. With high-pitch spiral technology, table speeds of 737 mm per second can be achieved.
ELECTRON BEAM CT — Electron beam CT (also called ultrafast CT or fifth generation CT) uses an electron gun instead of an x-ray tube and detector array [4,52]. The anode is bombarded by electrons, producing an x-ray beam that sweeps the patient (figure 4). Images are obtained in 50 to 100 milliseconds, virtually freezing motion. This technique is available in a limited number of centers and is used primarily for pediatric and cardiac applications:
●Imaging the upper airways .
●Imaging noncooperative children without sedation.
●Imaging the heart .
●Imaging the coronary arteries, including detection of subtle coronary artery calcifications and calculation of a coronary calcium score. This evaluates the risk (risk stratification) for future acute coronary events . The absence of calcifications makes coronary artery disease less likely in asymptomatic persons. Multislice CT scanning can achieve similar or better results and has replaced electron-beam CT scanning for most applications.
Photon-counting CT is a form of x-ray CT, in which individual photons in the x-ray beam are detected by a photon-counting detector that registers the interaction of individual photons rather than energy-integrating detectors. Photon-counting detectors record an energy spectrum (ie, energy-resolved CT technique). The currently used energy-integrating detectors detect the total energy generated from a summation of multiple photons. They register only the photon intensity but not the spectral information .
Photon-counting detectors yield better signal-to-noise and contrast-to-noise ratios by removing electronic noise, reduce radiation dose, and have a high spatial resolution devoid of electronic noise. Photon-counting CT scanners are already in clinical use .
X-Ray dark-field CT is a new, experimental technique that utilizes the number of multiple refractions at air-tissue interfaces such as the pulmonary alveoli; it takes advantage of the wave component of x-rays rather than the particle characteristics of photons utilized in conventional CT scans; the signal is reduced with impairment of the alveolar integrity. Dark-field CT scanners are able to provide three-dimensional information about the alveolar structure. Compared with conventional attenuation-based CT, the radiation dose is reduced .
RADIATION DOSE — The effective radiation dose for CT of the chest varies between 1 and 10 mSv, which is approximately 10 to 100 times more than chest frontal and lateral radiography (0.1 mSv) . Among the various CT scanners, helical CT yields a radiation dose that is greater than that of conventional, axial high-resolution CT.
Specification of the amount of radiation generated by CT scanning is best described by a dose metric known as CT Dose Index (CTDI). It is measured in a cylindrical acrylic phantom with a diameter of either 16 or 32 cm, placed at the CT scanner isocenter. CTDI is measured with a long pencil-shaped ionization chamber. Radiologists use CTDI volume that specifies the radiation intensity of the beam used to perform a specific CT examination. For identical radiographic techniques with similar tube current and tube voltage, the CTDI volume can differ by a factor of two due to different x-ray tube design, tube filtration, and beam-shaping filters. CTDI volume is fixed and independent of patient size or scan length. It can be reduced by reducing the tube current (mAs) [58,59].
The total radiation on the patient is expressed as dose-length product (DLP) in mGray times centimeters dependent on the scan length. DLP is also easily accessible to the interpreting radiologist. The amount of radiation used is directly proportional to the patient's effective dose (ED) in millisievert. The ED to DLP ratio yields a conversion factor that allows for conversion of DLP into effective radiation dose, which implies the total body detriment from radiation to a limited part of the body. This conversion factor is 1.4 for thoracic CT scanning [58,59]. The effective radiation dose for an examination decreases with increasing patient age and patient weight.
To put this into context, all individuals are exposed to natural background radiation of approximately 3.6 mSv per year . The average population exposure in the United States had increased to 6 mSv per year by 2006 with CT scanning [61,62] accounting for half the dose increase [58,61,62]. Overall, diagnostic and interventional procedures accounted for an effective per capita dose of 2.9 mSv in 2006 but decreased to 2.3 mSv by 2016 due to a collective medical dose decrease of 20 percent between 2006 and 2016 [59,63]. Close to 90 million CT scans are performed every year in the United States.
Studies of atomic bomb survivors from Japan suggest that the risk of developing cancer significantly increases when the effective radiation dose exceeds 50 to 100 mSv. Whether the risk of cancer is increased with lower levels of radiation in the diagnostic radiology range of exposure is uncertain and controversial.
To minimize radiation exposure, CT of the chest can be performed using low radiation protocols with lower tube current and lower tube potential as well as iterative reconstruction , which provide effective radiation doses of only 0.5 to 1 mSv [65-67].
Technologists play an important role in reducing radiation dose during CT scanning. Patient positioning in the scanner is important to allow automatic tube current modulation to optimize radiation dose according to patient body attenuation characteristics. The scanner must estimate the patient's attenuation characteristics over the range provided by the anteroposterior (AP) and lateral localizer digital radiographs (topograms or scout views). The automatic tube current modulation adjusts the tube current (milliampere seconds) according to the patient size and attenuation in order to maintain optimal noise level and lowest-possible radiation dose.
Another piece of equipment, the bowtie filter, compensates for variation in patient attenuation at the level of the detector during scan rotation. The filter is composed of a thin and thick segment allowing for maximum beam intensity for anatomical structures with high attenuation (eg, mediastinum) and lower beam intensity for structures with lower attenuation (eg, lungs). In order to ascertain that the automatic tube current modulation and the bowtie filter work as intended, the patient must be properly centered in the scanner in relation to the CT gantry. The prerequisite is the accurate localizer radiograph. If the patient is positioned too close to the x-ray tube, the scanner will overestimate the patient's size and the automatic tube current modulation will transmit unnecessary higher tube current, thus increasing the patient's radiation dose. The correct functioning of the bowtie filter also relies on centering in order to avoid higher than necessary dose in thinner parts of the anatomy. Poor centering can increase the dose to the patient by up to 67 percent. In addition, statistically significant differences in the HU standard deviation can occur due to inadequate patient positioning in the scanner.
The following recommendations for improving patient positioning have been offered [68-72]:
●The x-ray tube should be positioned at the top of the gantry, rather than at the bottom, to avoid offsets below the isocenter.
●AP and lateral scout views should be used with the lowest possible tube current and voltage setting.
●They should be repeated if patient position changes.
●The AP scout view is more robust than the lateral scout view.
Overall, CT should be used judiciously and overutilization avoided to reduce the potential induction of radiogenic cancers [61,62].
SUMMARY AND RECOMMENDATIONS
●Computed tomography (CT) is an imaging technique that has revolutionized medical imaging. It is widely available, fast, and provides a detailed view of the internal organs and structures. Helical CT is most common, but conventional, axial, step-and-shoot CT is used for thin section high-resolution CT scanning of the lungs, coronary artery calcification scoring, and prospective ECG-gated coronary CT angiography. (See 'Introduction' above.)
●The principal components of a CT scanner are the x-ray tube and a diametrically opposed array of detectors. The x-ray tube generates an x-ray beam as it rotates around the patient and the detectors concurrently record the radiation that traverses the body. The acquired data are digitized and then reconstructed into axial images. (See 'General description' above.)
●CT is superior to conventional radiography because of its exquisite contrast resolution and its lack of superimposition of anatomic structures. (See 'Advantages' above.)
●Helical CT is superior to conventional, step-and-shoot, axial CT because it is faster and it minimizes respiratory misregistration and motion artifacts. In addition, helical CT can enable the performance of CT angiography with subsequent reformation of high quality multiplanar two-dimensional and three-dimensional images. (See 'Advantages' above and 'Two-dimensional and three-dimensional rendering' above and 'CT angiography' above.)
●Several of the parameters used in CT are set by the operator. These include the collimation of the beam and reconstructed section thickness, pitch, the field of view, the duration of the helix, the reconstruction interval, and the reconstruction algorithm or kernel. Operators must also be familiar with proper breathing instructions and protocols for the injection of intravenous contrast material. (See 'Breathing instructions' above and 'Intravenous contrast' above.)
●Multislice helical CT is a major technical advancement in helical CT scanning. It has many advantages over conventional single slice helical CT. However, its major drawback is its higher effective radiation dose. (See 'Multislice helical CT' above and 'Radiation dose' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟