An Introduction to Computerized Tomography Systems and Applications

by

 Dr. Rajneesh Kumar Sharma MD (Homoeopathy)

Dr. Rajneesh Kumar Sharma MD (Homoeopathy)

Dr. (Km) Ruchi Rajput BHMS

Homoeo Cure Research Centre P. Ltd.

NH 74- Moradabad Road

Kashipur (UTTARANCHAL) - INDIA

Ph- 09897618594

 

O

ur masters told us to draw our attention towards the patient as a person, as a whole, as an individual and not to waste our energy in studying diseases itself. But the time has changed. To make a perfect diagnosis, ascertaing correct prognosis and therefore the true remedy with suitable dose with reliable posology i.e. absolute similimum, there are several shortest ways currently available. These are nothing but various means of investigations to reach the diagnosis and plan for treatment. One of these ways is the Computerized Tomography or CT scan.  The purpose of this article is to present a brief overview of CT systems. Various computer components and their features that are particularly relevant to CT will be described. It is assumed that the basic aspects of radiologic equipment and physics are generally known. Overall performance characteristics such as resolution and image artifacts will also be discussed.

HISTORICAL DEVELOPMENT

CT systems have been classified according to-

First-generation scanners-

With the earliest scanners, each line integral was collected using an x-ray tube and a single detector. During the scanning procedure, the x-ray tube and detector were translated across the scan field of view, and a series of transmitted intensity measurements were made. The tube and detector combination were rotated through a small angle (typically 1), and the translation repeated. This process of translation and rotation was repeated until 180 of projection data had been acquired. It took 41/2 minutes to complete a scan.

Second-generation scanners-

A significant reduction in scan time was obtained by using multiple detectors. These were placed opposite the tube so that the beams sensed by the detectors differed in angle by a small amount (e.g., 1). These scanners still required translation but each translation produced many views, one from each detector position. The rotation angle between translations was increased. With these systems, scan times were of the order of 20 to 80 seconds.

Third-generation scanners-

To decrease scan times further, rotate-only systems have been introduced. In these systems a fan beam x-ray source irradiates a large array of detectors. For third-generation systems, both the tube and detector array are rigidly coupled and rotate jointly about the patient. At any instant of time a complete fan or view of data is acquired.

Forth-generation scanners-

These are also called rotate-stationary systems in which only the tube rotates about the patient. A fixed detector array completely surrounds the patient. The fan beam projections are formed by grouping all the measurements made by a single detector. Scan times of 1 to 10 seconds are possible with these rotate-only systems.

COMPONENTS OF A CT SCANNER

The main components of a CT scanner are-

X-RAY SOURCE AND GENERATOR

To achieve good contrast sensitivity, a sufficiently large number of x-ray photons must be detected. However, only 1% of the energy incident on the x-ray tube anode is converted to x-rays; the rest merely produces heat. Furthermore, only a small fraction of the generated x-rays are used, because only a narrow fan beam exits the collimator. As a result, techniques as high as 200 mAs to 800 mAs at 120 kVp to 140 kVp are used for each scan.

The newer and faster rotate-only scanners use rotating anode tubes, which can operate at higher instantaneous power but only for reduced scan times.

"Off-focus radiation" is the term used to refer to x-rays that do not originate from the focal spot. Limiting or correcting for off-focus radiation is important, because off-focus radiation can reduce contrast and cause halo like artifacts around high-contrast image edges.

Because of the long exposure times, high voltage stability with little ripple in beam intensity is desired. X-ray systems can be operated in either a pulsed or continuous mode. Ideally, a pulsed system with a low duty cycle is preferred, because it provides an almost instantaneous measurement and reduces blurring caused by gantry motion during data acquisition. Furthermore, scan time reductions force the view acquisition rate to be increased while at the same time require higher average x-ray intensities (higher average mA) during the scan. These factors combine to make x-ray tube pulsing prohibitive. At the same time, the development of more stable detectors and data acquisition systems have reduced the need for electronic offset stabilization in the data acquisition system. As a result, many newer third-generation scanners use continuous x-ray sources.  

Prepatient collimators are used to limit the thickness of the scan section, and must be accurately adjustable for several thicknesses in the 1- to 10-mm range. Thus, the focal spot should be as small as possible and the focal spot-to-collimator distance as large as possible. The slice sensitivity profile, or width of the imaging slice, can be further limited by a postpatient collimator but at the expense of wasted radiation dose.

Typically, the rays measured near the center of the fan beam experience much more attenuation because of the patient than rays toward the edges of the fan beam. Some systems use beam-shaping filters to pre-attenuate the edges of the fan beam prior to irradiating the object.

All CT detectors are operated in the "current integration" mode, in which the signals produced by the x-ray photons incident on the detector over a brief period of time are summed. The high x-ray intensities used preclude counting individual x-ray photons, because detectors able to count at rapid enough rates are not readily available.

Scintillator materials that have been used include sodium iodide (NaI), bismuth germanate (BGO), and cesium iodide (CsI). Cadmium tungstate (CdWO4) and other tungstate crystals are now frequently used. Technique changes such as kV, mA, and collimator aperture can account for another factor of 100:1. Thus, CT detectors must be able to handle signals that cover a 10,000:1 range.

To maximize the information generated from patient dose, high quantum detection efficiency (QDE) is desirable. QDE is the product of absorption and geometric efficiencies. Scintillators can be designed to achieve almost 100% absorption.

The data acquisition system (DAS) performs the function of converting signals from the x-ray detector into measurements for use by the reconstruction hardware. Signals received from each detector channel must be integrated or smoothed to obtain a     measure of the number of x-ray photons received by the detector in a given period of time. These analog signals are amplified and converted into digital form by an analog-to-digital converter (ADC).

A CT DAS must be able to handle signals covering an amplitude range of 10,000:1. This can be compared, for example, to a high-quality TV camera used in digital radiography, which handles a signal range of no more than 1000:1.

GANTRY AND PATIENT TABLE

The x-ray tube, collimator, detector array, and DAS are mounted on the gantry. This contains the mechanisms for driving the source and detector precisely through the scanning movements. Because obtaining accurate positional data is required for artifact-free images, alignments between focal spot, collimator, and detector are critical. The gantry also contains a cable take-up mechanism to handle the high-voltage cables to the x-ray tube and, for all but fourth-generation systems, the signal cables to the DAS. Most scanners alternately rotate clockwise and counterclockwise to acquire data. In most CT scanners, the complete gantry can be tilted 20 to the vertical to allow the imaging of planes other than perpendicular to the long axis of the patient.

The table on which the patient is placed is motorized to provide for accurate positioning of the required body plane in the scan plane, and for automatic incrementation between scans. Manual positioning can usually be accomplished by light beam markers; longitudinal movement can be remotely controlled to assist in performing a series of scans at set intervals.

COMPUTATIONAL AND DISPLAY SYSTEMS

The computational subsystem usually consists of a general purpose computer and associated peripheral devices, such as disks and magnetic tape drives. In addition, it often includes special computational hardware such as array processors and back projectors. During scanning the computer functions as a digital controller for the gantry and table motions, x-ray generation, and data acquisition. Patient throughput, the ability of the system to perform data acquisition, image reconstruction, and display quickly, can be improved by careful attention to data paths and by including large memories to hold raw projection data and reconstructed images. The computational requirements for fast image reconstruction with no sacrifice in image quality have necessitated using array processors for the calibration and filtering operations of image reconstruction, and using special-purpose hardware to perform the back projection operation. Depending on the requirements, array processors able to perform 5 million to 50 million floating point operations per second are used.

Display subsystems consist of a display controller, image memory, digital-to-analog convertor (DAC), and monitor (CRT). As described elsewhere, CT image data consist of numbers among - 1000 to 3000. The human eye, however, cannot distinguish between more than 256 gray levels. The digital numbers are converted to analog signals using an eight-bit DAC, and a window level control is used to control the mapping of CT numbers into gray levels. Software for calculating the mean and standard deviation within a region and for image manipulations such as filtering and magnification is often included as part of the display program.

Related terms with CT Scanning

SCANNING GEOMETRIES

During the discussion of components, the effect of scanning geometry was mentioned.

PARAMETERS AFFECTING PERFORMANCE

The major components of a CT system have been briefly described above from the viewpoint of a design engineer, in which the requirements and advantages or disadvantages of the components were considered. From the viewpoint of the user, it is the overall system performance that is important.

SPATIAL RESOLUTION

Spatial resolution is characterized by the ability to visualize individual, small, high-contrast objects that are spaced closely together. Resolution is usually measured by using phantoms consisting of sets of air-filled holes or bars in Plexiglas.

In general, there will be a resolution loss toward the edges of the field of view. Loss of radial resolution results mainly from an enlargement of the apparent focal spot size toward the edges of the fan beam.

LOW CONTRAST DETECTABILITY

Low contrast detectability is defined as the smallest object visible at a given percent contrast level. It involves both image noise and spatial information.

SCAN TIME

Scan time is the time necessary to acquire projection data. Faster scan times can improve image quality by minimizing the effect of peristaltic motion and of moving organs in abdominal studies and in trauma. Faster scan times also increase the potential number of scans that can be performed during peak enhancement periods following contrast injection.

RADIATION DOSE

The radiation dose at any point in the region of the object refers to the energy absorbed per unit mass. The radiation dose can vary considerably in different parts of the object. Not only does it vary in a direction perpendicular to the section, but it also varies within the scan plane.

ARTIFACTS

Any deterministic structure, pattern, or CT number variation in the reconstructed image that does not correspond to a structure in the object scanned is an artifact. Artifacts always detract from image quality, and in some cases could result in a misinterpretation of a pathologic condition. Some artifacts, such as beam hardening and partial volume, are inherent in the physics of CT imaging.

REFERENCES

  1.       Newton TH, Potts DG (eds):Radiology of the Skull and Brain; Technical Aspects of Computed Tomography, Vol 5. St Louis, CV Mosby, 1981

 2.        Riederer SJ, Pelc NJ, Chesler DA:The noise power spectrum in computed tomography. Phys Med Biol 23:446, 1978

 3.        Yaffe M, Fenster A, Johns HE:Xenon ionization detectors for fan beam computed tomograph scanners. J Comput Assist Tomogr 1:419, 1977

 4.        Gordon BM:Linear electronic analog-digital conversion architectures, their origins, parameters, limitations, and applications. IEEE Trans Circ Syst CAS-25:391, 1978

 5.        Glover GH, Eisner RL:Theoretical resolution of computed tomography systems. J Comput Assist Tomogr 3:85, 1979

 6.        Hilal SK, Trokel SL:Computerized tomography of the orbit using thin sections. Semin Roentgenol 12:137, 1977

 7.        Cohen G, DiBianca FA:The use of contrast-detail-dose evaluation in a computed tomography scanner. J Comput Assist Tomogr 3:189, 1979

 8.        McCullough EC, Payne JT:Patient dose in computed tomography. Radiology 129:457, 1978

 9.        Shope TB, Morgan TJ, Showalter CK et al:Radiation dosimetry survey of computed tomography systems from 10 manufacturers. Br J Radiol 55:60, 1982

 10.      Brooks RA, DiChiro G:Beam hardening in x-ray reconstruction tomography. Phys Med Biol 21:390, 1976

 11.      Herman GT:Correction for beam hardening in computed tomography. Phys Med Biol 24:81, 1979

 12.      Duerinckx A J, Macovski A:Polychromatic streak artifacts in computed tomography images. J Comput Assist Tomogr 2:481, 1978

 13.      Joseph PM, Spiral RD:A method for correcting bone-induced artifacts in computed tomography scanners. J Comput Assist Tomogr 2:100, 1978

 14.      Glover G, Pelc N:Non-linear partial volume artifacts in x-ray computed tomography. Med Phys 7:238, 1980

 15.      Joseph PM, Spiral RD, Stockham CD:The effects of sampling on CT images. Comput Tomogr 4:189, 1980

 16.      Brooks RA, Glover GH, Talbert AJ et al:Aliasing:A source of streak artifacts in computed tomograms. J Comput Assist Tomogr 3:511, 1979