| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
IMAGING & THERAPEUTIC TECHNOLOGY |
1 From the Departments of Medicine (Cardiovascular Medicine Section) and Medical Physics, University of Wisconsin, H6/333 Clinical Science Center, 600 N Highland Ave, Madison, WI 53792. From the AAPM/RSNA Physics Tutorial at the 1999 RSNA scientific assembly. Received June 29, 2000; revision requested July 18 and received August 14; accepted August 16. Address correspondence to the author (e-mail: vanlysel@facstaff.wisc.edu).
 |
Abstract |
|---|
 Top Abstract IntroductionOptical CouplingVideo SystemConclusionsReferences |
|---|
Index Terms: Fluoroscopy • Physics • Video systems
|
Introduction |
|---|
|
TopAbstractIntroductionOptical CouplingVideo SystemConclusionsReferences |
|---|
The purpose of this article is to discuss two components of this imaging chain: the optical distributor and the video system. After reading this article, the reader will be able to (a) describe how the optical distributor forms an image for the cameras to record; (b) describe how the camera aperture sets the patient x-ray exposure level and the image noise level; (c) describe the operation of tube-based and charge-coupled device (CCD)–based video cameras; (d) calculate the resolution limitations imposed by the video camera; and (e) explain the video terms lag, gamma, progressive scanning, interlaced scanning, fields, frames, standard-line, high-line, and upscanning.
|
Optical Coupling |
|---|
|
TopAbstractIntroductionOptical CouplingVideo SystemConclusionsReferences |
|---|
The optical distributor is responsible for three functions: (a) transmission of a focused image from the II output phosphor to the focal planes of all cameras present on the system; (b) ensuring that the intensity (ie, brightness) of the images at the camera focal planes is correct for each camera; and (c) sampling the light level of the II output phosphor and transmitting this information to the automatic exposure control (AEC) circuitry of the x-ray generator.
Optical Distributor Components
Let us look at the components of an optical distributor that has both a video camera and a cine film camera (Fig 1). During cine operation, a semisilvered, beam-splitting mirror simultaneously provides images to both cameras. During coronary angiography, for example, this mirror allows real-time monitoring of the injection with the video camera while images are recorded on film. (It is typical to also videotape the video signal as backup against loss of the film during processing.) During fluoroscopy, the mirror is removed from the beam path so that all of the light passes to the video camera.
|
Also included in the beam path is a small mirror or prism for directing light to the AEC sensor. The sensor can be either a photodiode or a scintillating crystal coupled to a photomultiplier tube. A lens is used to form an image, then a field stop is used to prevent light from the periphery of that image from reaching the sensor. It is not uncommon for the edge of the image to consist of unattenuated radiation. If the AEC were to include this region in its determination of exposure level, the central portion of the image, which is presumably the area of interest, would be underexposed (ie, dark and noisy). If only the central 50% (for example) of the image is sent to the AEC sensor, this problem is avoided.
Optical Considerations
The optical elements in the distributor are always arranged in a particular manner. The distance between the II output phosphor and the collimating lens is equal to the focal length of the collimating lens (ie, the phosphor is in the focal plane of the collimating lens) (Fig 1). As a result, the light leaving the lens is collimated (ie, the light rays that were emitted from a single point on the phosphor are parallel after they leave the lens). The image is said to be "focused at infinity." The video camera target is placed in the focal plane of the video camera lens, and the film in the cine camera is placed in the focal plane of the cine camera lens (Fig 1). Because the light received by each camera lens is collimated, the image formed by the lens is at the focal plane. Cameras can go out of focus over time as vibrations cause the distance from the lens to the target or film to change. Periodic quality control measurements are necessary to detect this problem.
A property of collimated light is that every region within the beam path contains light from all points on the output phosphor (however, there are limits to this statement that, if not heeded by the optical designer, will lead to vignetting, or darkening of the image periphery). As a result, when objects such as an aperture or the small mirror for the AEC sensor are placed in the beam path, these objects do not show up in the image. The intensity of the image is reduced due to the reduction in light reaching the image, but this reduction is uniform across the entire image. This fact is why apertures are always placed between the two lenses. In contrast, the field stop in the beam path for the AEC sensor is placed after the second lens. At this location, after the image has been formed, the field stop serves to occlude light from the image periphery; as described earlier in this article, such occlusion of light is desirable for operation of this device.
The most important practical matter to understand regarding the optical distributor is the function of the camera apertures. To understand that, we need to discuss the concept of the speed of the optical system. The term speed is used here in the same manner that it is used for a screen in screen-film radiography. For a fixed dose rate, the faster the optical system, the less total x-ray exposure is required to properly expose the camera. The f-number of the optical system (often written f/#) designates the system's speed. Amateur photographers will recognize that it is the lens f/# that they vary to adjust for the brightness of the scene they are photographing. The f/# of a lens is defined as the ratio of its focal length to its diameter (technically, the diameter of the entrance pupil):
These two properties determine the brightness of the image formed by the lens (Fig 2). The size of the image is in direct proportion to the focal length; longer focal lengths form larger images. Because the lens collects a fixed amount of light, larger images must necessarily be dimmer, since the same number of light photons are spread over a larger area. Doubling the focal length quadruples the area of the image, which is then only one-fourth times as bright.
|
Although the focal lengths of the lenses are fixed and thus can be ignored during everyday operation of the system, the aperture can be adjusted to change the speed of the system. Introduction of an aperture that is smaller than the lens diameter reduces the brightness of the resulting image (Fig 2). The reduction in brightness is in direct proportion to the reduction in the effective area of the lens (thus it is also in proportion to the square of the diameter). Apertures allow the system to operate over a wide range of x-ray exposure levels, as described later in this article.
Two different aperture styles are used in fluoroscopic/fluorographic systems (Fig 3). Generally speaking, clinical procedures that make use of a film camera are performed at the same "detected" x-ray exposure level for all patients (eg, 16 µR [4.1 x 10-9 C/kg] per frame incident on the II for coronary angiography in the 17-cm mode); thus, a fixed aperture is used for these cameras. When the angiographic unit is installed, the service engineer determines the aperture size that will produce a properly exposed film at the desired x-ray exposure level. Periodically, this aperture may need to be replaced with one of a different size as the II gain falls with age, if the II is replaced, or if the film type or the desired x-ray exposure level changes.
|
A sample calculation illustrates the relationship between f/# and the x-ray exposure incident on the II (Fig 4): If the f/# of the video camera lens is f/2 during fluoroscopy performed at 2 µR per frame (Exp1), what is the proper f/# for the lens during cine angiography performed at 16 µR per frame (Exp2)? We answer this question by noting that an eightfold increase in entrance exposure results in an eightfold increase in II output brightness. Therefore, to keep the light intensity reaching the video camera constant, the area of the camera aperture must be decreased by eight times. Working through the calculation (Fig 4), we see that to decrease the light transmitted by the aperture by eight times, we need to increase the f/# (note the reciprocal relationship) by 
8, to f/5.6.
|
|
|
|
Video System |
|---|
|
TopAbstractIntroductionOptical CouplingVideo SystemConclusionsReferences |
|---|
The components of a closed-circuit video system used for fluoroscopy/fluorography are as follows (Fig 6): The camera head is attached to the optical distributor, allowing it to view the II output phosphor. The camera head contains an image sensor that converts the light image from the II into a voltage signal that can be transmitted to other components of the video system, where it can be modified, displayed, and recorded. The camera control unit, which is usually located in one of the generator's electronics racks, processes the video signal from the camera head so that it is suitable for display and digitization. The camera control unit also synchronizes the scanning of all of the components of the system by generating and transmitting horizontal (H sync) and vertical (V sync) synchronization signals.
|
|
|
Two operating characteristics of a video system that affect image quality and that users should understand are lag and linearity. Video camera lag is described as follows (Fig 9): When the light level falling on any region of the video sensor changes (either increasing or decreasing), it is desirable for the video signal out of the sensor to change instantaneously to its new value. The term lag refers to the condition in which the video signal output rises or falls to the new value more slowly than the changing light input. It is sometimes said that the camera is "sticky." The magnitude of lag varies greatly between different types of video cameras, depending on the details of target construction and materials. CCD cameras have no lag. At the other extreme, a photoconductive pickup tube called a vidicon has a very high degree of lag. The vidicon uses SbS3 as the target material. In the middle are several pickup tubes (eg, Plumbicon, Saticon, Chalnicon, Newvicon) that have different target configurations but all deliver about the same amount of lag. These "other" tubes are referred to as "low-lag" cameras but actually are low lag only when compared with the vidicon. In fact, the amount of lag for these cameras is appreciable. Lag is signal dependent; it is much higher in the dark parts of the image than in the bright parts. Lag increases as a pickup tube ages. Tube replacement is often prompted because the tube has become excessively "laggy."
|
|
where V is signal voltage and L is input light intensity, curves are usually characterized by their gamma value (2). The vidicon has a gamma of about 0.7, whereas all of the other pickup tubes and CCD cameras have a gamma of unity (g = 1 indicates a linear response). However, because a gamma of less than 1 has desirable properties, many systems with linear cameras impose a nonlinearity on the video signal. Imposition of nonlinearity can be done by either the camera control unit or the digital image processor. Small signals in the dark parts of an image are amplified by a transfer curve with a gamma of less than 1, allowing better visualization of these difficult-to-see signals (Fig 11).
|
|
The book-reading analogy suffers a bit when we consider vertical scanning in more detail. There are two ways in which vertical scanning can be performed. Progressive scanning (also known as sequential scanning) does conform to the analogy. In progressive scanning, lines are scanned in consecutive order. Line 2 is scanned after line 1. Line 3 is scanned after line 2. When we get to the bottom of the image, we will have scanned all of the lines. Interlaced scanning is a bit more complicated. After scanning line 1 we scan line 3, skipping line 2. After scanning line 3 we scan line 5, skipping line 4. We scan only the odd lines, reaching the bottom of the image in one-half the time required for progressive scanning. However, having read only half of the lines, we must go back to the top of the image and read the even lines in succession (lines 2, 4, 6, etc). The odd and even fields are interlaced to produce one frame. In this discussion, it is important to carefully note the difference between fields and frames.
Interlaced scanning was devised in the early days of broadcast television as a means to reduce the required bandwidth of video transmissions, allowing more channels to be broadcast within a market. Successful video transmission relies on the fact that the eye-brain system of the observer perceives the scanned image as continuous. In reality, the monitor phosphor begins to decay in brightness as soon as the electron beam moves on (see the discussion of monitor operation later in this article). After a few hundred microseconds, the phosphor is dark again. If your visual system were fast enough, you would see a bright bar, representing the image, sweeping down the face of a predominantly dark monitor. The monitor refresh rate is set so that your visual system blends the scanning bright bar into a continuous, bright image. The "flicker-fusion frequency" at which this blending occurs is a variable that depends on image brightness but generally requires a refresh rate of at least 50 images per second.
The standard scanning mode employed by conventional broadcast video is described as 30 frames per second, 525 lines per frame, 2:1 interlaced video. The 2:1 interlace process was described earlier; two fields are interlaced to produce one frame. The 525 lines per frame means that there are 525 horizontal lines making up one frame (one "image"). Thirty of these 525-line images are transmitted every second. We know that 30 frames per second is not fast enough to avoid a flickering image. However, when viewed from a sufficient distance, the individual fields, presented at 60 fields per second, blend together enough (because of the limited resolution of the human eye) to greatly attenuate the perception of flicker.
For many years, the video standard used in fluoroscopic/fluorographic systems was the same as that used in conventional broadcast television, 30 frames per second, 525 lines per frame, 2:1 interlaced. This system is still used today for inexpensive systems. Substantial cost savings are enjoyed by using off-the-shelf video components (eg, cameras, monitors, tape recorders). However, as the use of video has shifted from merely monitoring a procedure to being the primary diagnostic medium, and also the means to perform sophisticated interventions, the quality of standard video is no longer sufficient. One improvement is greater use of progressive scanning. Although interfield flicker is usually not apparent under casual observation of an interlaced display, close inspection of the image (as is likely to occur during medical procedures) does reveal an annoying flicker, especially at a horizontally oriented boundary between bright and dark regions. This flicker is due to the fact that the 30 frames per second display rate is, in fact, below the flicker-fusion frequency. The best systems now employ progressive scanning monitors with display rates of 60 frames per second or above (as opposed to a 60 fields per second, 30 frames per second, interlaced scanning monitor).
More important than display considerations, interlaced scanning by the video camera during acquisition can cause significant image artifacts. The nature of the artifact depends on the manner of x-ray production. The most common artifacts are distortions of moving objects, such as coronary arteries and guide wires. For both continuous fluoroscopy and pulsed fluoroscopy/fluorography at a rate of 60 pulses per second, the moving object is in two different locations for the even and the odd fields. The resulting image of the artery in the interlaced frame will have a serrated appearance (3). For pulsed operation, reduction of the x-ray pulse rate to 30 pulses per second eliminates the serration artifact, but a more severe artifact is produced. In this case, the intensity of the two fields is very different, resulting in severe flicker. Cardiac angiographers who were practicing before the advent of digital imaging are familiar with the flashing that occurred on the video monitor during 30 frames per second cine-film imaging. This flashing occurs with a pickup tube system because, while the scanning electron beam is reading the odd field, it inevitably reads a great deal of charge that properly belongs to the even field, depressing the intensity of the even field when it is read next if a new x-ray pulse does not replenish the lost charge.
A very common method used in modern fluoroscopic video systems to provide both the superior performance of progressive scanning of the video camera during acquisition and the economy of an interlaced display is scan conversion (Fig 13). Scan conversion requires a digital image processor to buffer the data received from the camera until it is time for display. In an analog system without a buffer, the signal read from the camera is transmitted immediately to the display monitor. Without a short-term storage method, both the camera and the monitor must scan in lockstep. Therefore, if the monitor is a conventional interlaced scanning model, the camera target must be read in interlaced fashion. However, with a digital buffer, acquisition and display are decoupled. The camera reads progressively, and the data are stored in a digital memory. Once one frame has been read, the camera begins to fill a second memory, freeing the first memory to be read out in interlaced fashion for display.
|
|
where u is the spatial frequency in cycles per millimeter. In the coronary angiography example, the MTF of the high-line video system is significantly better than the MTF of the II and 1-mm focal spot. The result is that the additional resolution loss imposed on the total imaging system by the video component is small. In contrast, in the peripheral DSA example, MTFvideo and MTFtotal are virtually identical, indicating that the resolution limitation of this system is dominated by the video system. In such a case (specifically, when a large-II format is used), a standard 525-line video system is insufficient to provide adequate resolution. The reader should understand this point after the following discussion.
The spatial resolution of the video system is determined by separate considerations in the vertical and horizontal directions (Fig 15). Vertical resolution is primarily determined by the number of horizontal lines the system uses. Horizontal resolution is primarily determined by the bandwidth of the video system. Unlike IIs and screen-film systems, which tend to have rotationally symmetric resolution properties, the resolution of the video system can potentially be very different in the vertical and horizontal directions. However, the video system designer usually chooses to make the resolution in the two directions approximately equal.
|
|
Actually, we should say that the maximum achievable vertical resolution is 245 line pairs per scan height. Several factors conspire to reduce the typical vertical resolution from that which we just calculated. For one, the electron-beam spot in the pickup tube is not pillbox-shaped (with sharp edges) but tends to be more Gaussian, resulting in overlap between the lines (especially during interlaced scanning). More significantly, how the line-pair gauge lines up with the raster scan (ie, its phase) can significantly affect the outcome of our observation (Fig 16). The calculation performed earlier assumes that the spaces and bars of the line-pair gauge line up exactly with the scan lines of the video system. If the line-pair gauge is shifted by half the width of a line (90° phase shift), contrast (modulation) in the resulting image disappears, since the signal read by each video line represents half a bar and half a space. In the interest of quoting a single specification for limiting resolution, it is common to reduce the maximum achievable vertical resolution by an amount referred to as the Kell factor, after R. D. Kell, who researched this issue in the 1930s. A typically quoted value for the Kell factor is 0.7, resulting in a practical vertical resolution for our 525-line system of 245(0.7) . 170 line pairs per scan height.
|
|
where SID is the distance from the source to the II face (source-to-image distance) and SOD is the distance from the source to the object plane (source-to-object distance).
As an example, calculate the object-plane resolution of a 525-line video system when the 23-cm II mode is used. Assume that the geometric magnification is 1.2 and the Kell factor is 0.7:
The Table presents video resolution as a function of II mode. Resolution is seen to decrease with increasing II size, since a fixed number of video lines are stretched across an increasing field of view. What is striking about this result is how poor the video resolution is. Compare a video resolution of approximately 1 lp/mm with the approximately 5 lp/mm limiting resolution of the II. Film resolution (either cine or photospot film) is so high that the imaging chain resolution for these modalities is limited to that provided by the II.
|
Horizontal Resolution.—Resolution in the horizontal direction is limited by the bandwidth of the video system. The concept of video bandwidth can be explained by drawing an analogy with the stereo systems that many people have in their homes. When designing the electronics for an audio system, the designer makes use of the fact that the human ear is capable of detecting audio frequencies of up to 20,000 cycles per second (20 kHz). The designer therefore makes sure that the electronics are capable of transmitting voltage waveforms of up to 20 kHz (ie, that the bandwidth is 20 kHz). If the designer does not do so, the listener will not hear the high-frequency tones in the music because they were not passed from the source to the speakers by the electronic components (such as the amplifier). Conversely, any money or effort spent to pass frequencies above 20 kHz is wasted because the listener cannot hear them.
This analogy can be extended to video systems by imagining a low-pass filter at the end of the amplifier chain that transmits the signal from the camera to the monitor (Fig 18). A low-pass filter passes all frequencies up to a cutoff frequency, attenuating frequencies above the cutoff frequency. We can equate the video system bandwidth with this cutoff frequency. It is the task of the system designer to make sure that this filter has the appropriate frequency-transfer characteristics. The consequences of improper bandwidth selection are significant (Fig 18). Say we desire to transmit the image of a line-pair gauge. If the bandwidth is set too low, a rounded-off, blurry representation of the gauge will be passed to the video monitor. We will have degraded the horizontal resolution of the video system. However, we also want to make sure that we do not set the bandwidth too high because there is a great deal of electronic noise, generated by the video system, at high frequencies. Passing these frequencies through to the monitor will result in an excessively noisy image.
|
Step 2 is to determine how much time is spent scanning one line. In our example, we have a 525 lines per frame system that scans 30 frames per second. Therefore, the period of a single line is as follows:
However, we learned earlier that 15% of this period is spent in horizontal retrace, during which no information is transmitted by the system. Therefore, we have only 63.5(1 - 0.15) = 54 msec to transmit our 245 cycles.
Step 3 in our bandwidth calculation is to divide the required number of cycles per scan line (245 cycles) by the time we have to transmit them (54 µsec):
The only other video scanning mode typically encountered in a fluoroscopy system is the high-line mode discussed earlier. The required bandwidth for a 1,049 lines per frame, 30 frames per second scanning mode is calculated as follows:
In this calculation, it is assumed that there are 980 active scan lines (69 are lost to vertical retrace) and that the active horizontal period is reduced by 15% due to horizontal retrace. The reader will note that the high-line bandwidth is four times that of the standard-line mode. This increase is a result of two facts: (a) We require twice as many cycles per scan line to increase the horizontal resolution by the same amount as the improved vertical resolution (the equal vertical and horizontal resolution criterion) and (b) the horizontal period is half as long (31.75 msec vs 63.5 msec) because we need to scan twice as many lines in the same 1/30th of a second frame period.
What if the bandwidth of a particular system is less than that required? We are now in a position to calculate the resulting resolution. Let us assume that the video system in question has a bandwidth of only 3 MHz. If we are imaging in the 23-cm II mode, the limiting resolution at the face of the II is calculated as follows:
If the geometric magnification was 1.2, then the object plane resolution would be 0.7 x 1.2 = 0.85 lp/mm.
Frame Rate.—Inspection of the relationships in the previous section indicates a direct proportionality between frame rate and bandwidth. Doubling the frame rate doubles the required bandwidth because it halves the horizontal line period. However, a point that is sometimes misunderstood is that lower x-ray pulse rates do not typically reduce the required video bandwidth. It is becoming increasingly common to perform both fluoroscopy and fluorography at reduced pulse rates (eg, 15 or 7.5 pulses per second) to reduce x-ray dose (as discussed later in this article). It is common to refer to these rates as 15 or 7.5 "frames" per second. However, the implication that the video system is operating at a reduced frame rate is usually incorrect. For example, both coronary angiography and pulsed fluoroscopy are now often performed at "15 frames per second." The typical manner in which this rate is achieved is to produce x-ray pulses at 15 pulses per second but continue to scan the video camera at 30 frames per second. The first camera scan after the x-ray pulse contains the video information from that pulse (Fig 19 [middle waveform]). This frame is both sent to the video monitor for display and stored in a digital frame buffer. The second time the camera target is scanned, there is no new information because there was not an x-ray pulse. Therefore, the first frame, stored in the frame buffer, is displayed again. This process retains the necessary 30 frames per second display rate (to avoid display flicker). At the beginning of the third frame, a new x-ray pulse provides new video information.
|
When the x-ray pulse width becomes greater than about 10 msec, other methods must be used to scan the camera. This situation occurs in low frame rate (eg, 1 frame per second) DSA and digital photospot imaging. In this case, the detected x-ray exposure can run from 100 µR to 1,000 µR per frame (at the face of the II) to reduce the x-ray noise amplitude in the resulting image. To achieve this exposure level, x-ray pulse widths on the order of 0.1 second or greater must be used. The x-ray pulse then extends over several video frame periods. Two different methods can be employed to scan this signal. One method is to operate the video camera continuously in the normal 30 frames per second scanning mode. Video frames that span the x-ray pulse period are integrated (summed) by the digital image processor to produce one image. This image is stored in the image processor and displayed repeatedly until the next x-ray pulse provides a new image. The other method is as follows (Fig 19): For a pickup tube–based camera, the scanning electron beam is blanked, just as it is during retrace, during the x-ray pulse. After termination of the x-ray pulse, the beam is turned back on (synchronous with V sync) to read the image. CCD camera operation is similar; the shift operations are suspended during the x-ray pulse.
As mentioned earlier, it is becoming increasingly common to encounter systems that offer "frame rate reduction" as a means to decrease patient and staff x-ray exposure. Such systems require that the x-ray source be pulsed. Cardiac angiography (ie, cine) has always been pulsed, but in the past most fluoroscopy was performed with "continuous" x rays. Frame rate reduction also requires that scanning of the video camera be progressive rather than interlaced. However, it is not true, as is often claimed, that pulsed-progressive fluoroscopy, in and of itself, results in exposure reduction. Rather, pulsed-progressive fluoroscopy enables frame rate reduction, which results in exposure reduction.
Exposure reduction is not as straightforward as it might at first appear. If the frame rate is dropped from 30 to 15 frames per second, the x-ray exposure drops by a factor of two only if the milliampere-seconds per pulse remains constant. However, the vendor might raise the milliampere-seconds per pulse to "compensate" for the reduction in frame rate, reducing or negating the exposure reduction. If it is assumed that the milliampere-seconds per pulse remains fixed, it is important to understand that, although the exposure does drop by two times, image quality also drops. This effect occurs because, when viewing a rapid succession of images, the eye integrates the noise content of all images presented within a period of approximately 0.2 seconds. If fewer images are presented within this "eye integration period," the observer perceives an increase in noise. Research shows that if image quality is held constant by increasing the milliampere-seconds per pulse as the frame rate is reduced, frame rate reduction results in only modest exposure reduction (4). However, as a practical matter, frame rate reduction can be a powerful tool for exposure reduction. In the past, changing the fluoroscopic exposure rate was an involved process requiring a service engineer. However, with pulsed-progressive fluoroscopy, the physician can be given table-side control of the fluoroscopic frame rate. If it is assumed that the vendor maintains a constant milliampere-seconds per pulse, the physician now has the ability to decrease or increase the exposure level in response to the demands of the procedure.
Image Display
The image that is read from the video camera target is displayed on a video monitor (Fig 20). The similarities in operation between the monitor and the camera pickup tube are conspicuous.In a monitor, an electron beam is produced by thermionic emission from a hot cathode. This beam is accelerated toward the face of the monitor by a high-potential anode. When the electron beam strikes the fluorescent screen on the monitor face, light is emitted. Deflection coils cause the electron beam to be scanned across the monitor screen in the required manner (usually in a 2:1 interlaced fashion). The magnitude of the electron beam current is determined by the voltage applied to the control grid, which in turn is determined by the video signal originating from the camera. If the point at which the electron beam is currently striking the monitor face is to be dark, the control grid reduces the electron beam current. If the image is to be bright, more electrons are allowed to pass through the control grid to reach the phosphor. During horizontal and vertical retrace, the control grid blanks the electron beam.
|
|
Conclusions |
|---|
|
TopAbstractIntroductionOptical CouplingVideo SystemConclusionsReferences |
|---|
Proper performance of the video system is critically important to good image quality. Ensuring optimum performance starts with the equipment purchase; here again, a sound understanding of system operation will allow intelligent choices. Once the system is in operation, parameters such as edge enhancement, lag, and gamma, which can often be adjusted through the digital system, can have significant effects on the perception of image detail. Finally, modern systems often afford the operator the option of reducing "frame rate" to reduce radiation exposure. It is beneficial to understand how such reduction is accomplished.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionOptical CouplingVideo SystemConclusionsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. A. Geise Fluoroscopy: Recording of Fluoroscopic Images and Automatic Exposure Control RadioGraphics, January 1, 2001; 21(1): 227 - 236. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
