Chapter 3 Computed Tomography Stewart C. Bushong CT Gantry Every CT imager has three distinguishing components – the operating console, the computer, and the gantry The operating console performs two major functions – imaging control with pre-selected technique conditions and image viewing and manipulation (window/level) CT Gantry There may be several operating consoles, each dedicated to a separate function, such as CT control or post-processing and image analysis (3D, diffusion/perfusion analysis, cardiac scoring, measurements, region of interest) The CT computer has no physically distinguishing features (it typically looks like any other computer) CT Gantry The CT computer has high capacity and is very fast due to the large number of computations required on an extensive data set – e.g. if there are 750 detectors and 1500 projections are acquired in 360 degrees of rotation that would equal 1,125,000 samples (750 x 1500) for EACH SLICE!!!! Each image at a 1024 x 1024 matrix requires approximately 2 MB of memory CT Gantry Some CT imagers have the computer built into the operating console Computers capable of multiprocessing are used in CT (multiprocessing means that each processing unit works on a different set of instructions to increase speed or computing power) CT Gantry Multiprocessing allows a computer to perform several functions at the same time, which reduces reconstruction time and increases capacity The gantry is special to CT. It houses the x-ray source, the detector array, the collimator assembly and a generator. Sometimes the generator is attached to the rotating framework along with the tube and detectors. Other times the generator is positioned on the floor of the gantry and does not rotate CT Gantry The patient aperture of a CT gantry has a diameter of approximately 70 cm. The CT gantry can be tilted in a cephalic or caudal angle plus or minus 30 degrees. The capability to tilt is especially useful for extremity imaging and facial imaging. E.g. by having a patient lie prone with their head extended, coronal images of the sinuses may be obtained Coronal Sinus CT The X-ray Source CT imaging places two demands on an x-ray tube – high x-ray intensity and rapid heat dissipation. High x-ray intensity is accomplished with a high mA generator and a generous focal spot size, up to 2mm Rapid heat dissipation is provided by large diameter, thick anode disks rotating at 10,000 rpm Thick Anode X-Ray Tube Components The X-Ray Source X-ray tubes developed for CT have very high heat capacity Anode heat capacity of 6 MHU (million heat units) are common. That compares to less than 1 MHU for general radiography. The anode-cathode axis is perpendicular to the patient axis to avoid the heel effect Anode Heel Effect http://learntech.uwe.ac.uk/radiography/RScience/diag_xray_tube/anode_heel.htm Anode Heel Effect http://learntech.uwe.ac.uk/radiography/RScience/diag_xray_tube/anode_heel.htm Referring to slide # 13 “Close examination of the x-rays emitted from the target shows that because they are produced below the surface they have to pass through some tungsten before they can escape from the tube.” “ X-Ray A has to pass through a much greater thickness of anode material before escaping from the x-ray tube” Anode Heel Effect http://learntech.uwe.ac.uk/radiography/RScience/diag_xray_tube/anode_heel.htm “X-ray B only has to pass through a small amount of tungsten” “As the angle of the anode is increased, the anode heel effect increases” The X-Ray Source Computed tomography x-ray tubes have high speed (10,000) rpm rotors X-ray tube failure is the principle cause of CT imager malfunction X-ray tube current of 200 to 800 mA are common. Too low mA can result in unacceptable image noise (caused by a lack of sufficient x-rays striking the detectors) The X-Ray Source X-ray tube potential is usually 120 kVp to 140 kVp three phase of high frequency Such high kVp is used for higher intensity and penetrability, and therefore, less x-ray tube loading and lower patient dose. Dual focus tubes are common, usually having .5 and 1.0 mm focal spots, with the smaller focal spot used for better spatial resolution Dual Focus Cathode http://learntech.uwe.ac.uk/radiography/RScience/diag_xray_tube/components_cathode.htm The X-Ray Source The improved spatial resolution does not result from projection geometry as in radiography, rather from better x-ray beam – radiation detector collimation Still, the principal effect os spatial resolution is matrix size and field of view (FOV) For third generation CT imagers, the x-ray source is pulsed. Each pulse creates an image projection from each detector The X-Ray Source When pulsed, up to 100 mA is used with pulse widths of 1 to 5 ms at pulse repetition rates of 60 Hz For fourth generation imagers the xray tube is energized continuously Each pass of a fourth generation fan beam over a detector produces an image projection The X-Ray Source Computed tomography x-ray beam are filtered to harden the beam and make it more unifrom at the detector array Filtration produces a higher energy, more homogeneous x-ray beam and reduces the beam hardening artifact A shaped x-ray beam filter is used in CT to produce a more uniform intensity at the detector array The X-Ray Source A “bow tie” filter is often used to even radiation intensity at the detector array High Voltage Generator High kVp is used to minimize photoelectric absorption and, therefore, patient dose High kVp is used to reduce bone attenuation relative to soft tissue allowing a wider dynamic range of the image High kVp is used to increase radiation intensity at the detector array High Voltage Generator High kVp is used to reduce x-ray tube loading, and thereby , extend tube life Three phase or high frequency voltage generation is used for CT imagers Three phase voltage is usually generated by a stand alone module near the gantry. Cables that will only wind 360 degree must be used, causing a reversal of gantry position High Voltage Generator High frequency generators are small enough that they can be mounted on the rotating gantry Heat units and joules are equivalent measure of energy Slip rings make possible continuous rotation of the x-ray source leading to spiral CT High Voltage Generator Slip rings incorporate circular electrical conductors, one type of which rotates and passes power to the high-voltage generator; the other passes signals from the data acquisition system to the computer: further explanation can be found at http://www.amershamhealth.com/medcyc lopaedia/medical/Volume%20I/SLIP%20RI NG%20TECHNOLOGY.ASP High Voltage Generator Essentially all CT imager now use high frequency generators • Three phase power was used until the mid 1980’s The high frequency generator can be positioned on the rotating gantry with the xray source The high frequency generator can be positioned on the fixed part of the gantry and connected to the x-ray source through slip rings High Voltage Generator The DAS is located between the detector array and the computer The DAS • Amplifies the detector signal • Converts the analog signal to digital • Transmits the digital signal to the computer High frequency generator voltage generation eliminated the need for massive high-voltage transformers Detector Array The evolution of the CT radiation detector has progressed with continuous improvements Detector efficiency is important because it determines maximum tube loading and controls patient dose Three important features of the detector array are efficiency, number of detectors, and detector concentration Detector Array Early CT imager used a scintillation crystal photomultiplier tube as a single element detector A grouping of detectors is called a detector array There are two types of detector arrays- gas filled and solid state Detector Array Gas filled detectors – high pressure xenon – have very fast response and no afterglow but only about 50% detection efficiency Gas filled detectors can be packed more tightly than solid state detectors with less interspace septa Most solid state detectors today use a scintillator, cadmium tungstate, optically coupled to a photodiode Detector Array Solid state detectors have nearly 100% detection efficiency but cannot be tightly packed The detector array consists of many individual detector fashioned as a module that are positioned on a receptor board for easy exchange and service A gas filled detector array uses small ion chamber filled with high-pressure xenon or other gas Detector Array Each ion chamber is about 1 mm wide with essentially no interspace The geometric efficiency – the percent area of the detector array that is detector, not interspace – is more than 90% The intrinsic detection efficiency for high pressure xenon is approximately 50% Total detector efficiency = geometric efficiency x intrinsic efficiency Detector Array Solid state detectors are made of a scintillation crystal, which when irradiated emits light that is converted to an analog signal by a photodiode Solid state detectors have approximately 90% intrinsic detection efficiency. Essentially, all incident x-rays are detected Detector Array Total detection efficiency depends on the number of detectors and how tightly they are packed When there is interspace between detectors, detection efficiency is reduced and patient dose increased Eighty percent total detection efficiency is common for solid state detector arrays Detector Array Solid state detectors are automatically recalibrated between scans Solid state detectors are more expensive than gas-filled detectors and their increased efficiency can result in less x-ray tube loading, reduced image noise and reduced patient dose Detector Array The DAS is positioned just after the detector array to amplify each signal, convert each signal to digital form, and properly sequence each signal to the computer Multiple detector array allow the collection of two or more image data sets simultaneously Multiple detector arrays can reduce the heat loading of the x-ray tube Detector Array Multiple detector arrays allow simultaneous imaging of two or more slices Collimator Assembly There are two collimator in CT – prepatient and post-patient The pre-patient collimator is positioned near the x-ray source The pre-patient collimator controls the patient dose and determines the dose profile As the pre-patient collimator is narrowed, patient dose increases and the dose profile becomes rounded Collimator Assembly pre-patient collimation controls slice thickness The dose profile is a plot of dose across the slice thickness The dose profile should be square but is rounded because of scatter radiation The post-patient collimator controls the slice thickness (sensitivity profile) Collimator Assembly When the post-patient collimators are narrowed, slice thickness is reduced Sensitivity profile is a plot of detector response versus distance (mm) The ideal sensitivity profile is square; in practice, it is rounded because of scatter radiation. Collimator Assembly pre-patient and post-patient collimators are controlled together to match dose profile and sensitivity profile If dose profile exceeds sensitivity profile, the patient dose is excessive If sensitivity profile exceeds dose profile, image quality is compromised Nominal slice thickness is controllable between 1 and 10mm (sub-millimeter scanning is available on newer multi-slice system) Collimator Assembly As the slice thickness is changed so is the voxel size Collimator Assembly Thinner slices are required for rapidly changing anatomy, for example, the inner ear Thinner slices result in improved spatial resolution Thinner slices result in higher patient dose because of increase overlap of slices When imaging with thin slices they are usually contiguous so that no tissue is missed Collimator Assembly High voltage slip rings are oil insulated and transfer power from an external high voltage generator to the gantry Low voltage slip rings are air insulated and transfer data from gantry to computer When a spiral (helical) CT is based on low voltage slip rings, the high voltage generator is high frequency type and mounted on the rotating gantry Collimator Assembly Please refer to page 31 of your textbook for a nice example of how all the CT Gantry parts fit together. Sources Computed Tomography: physical principles… – Seeram Helical Scanning – Blanck Introduction to Computed Tomography – Romans Computed Tomography – Bushong http://www.impactscan.org/slides/xrayct/i ndex.htm http://learntech.uwe.ac.uk/radiography/R Science/diag_xray_tube/d_xray_contents. htm