The Image Intensifier (II)
The image intensifier is comprised of a large cylindrical, tapered tube with several internal structures in which an incident x-ray distribution is converted into a corresponding light image of non-limiting brightness. A picture of an image intensifier television (II-TV) system is shown below. X-ray to light amplification is achieved in several sequential steps. First, x-rays incident on and absorbed by a cesium iodide (CsI) structured phosphor produce a large number of light photons resulting from the energy difference of x-rays (30-50 keV average) to light photons (1 -3 eV average). Absorption and conversion efficiency is on the order of 60% and 10%, respectively. A fraction of the light photons interact with an adjacent photocathode layered on the backside of the input phosphor, releasing a proportional number of electrons (typically on the order of 5 light photons / electron). Being negatively charged, the electrons are accelerated through a potential difference of approximately 25,000 volts towards the positive anode positioned on the tapered side of the evacuated tube. Electro-magnetic focusing grids maintain focus and at the same time minify the electron distribution as it interacts at the output phosphor structure, producing a large increase in the light intensity compared to the amount of light originally produced at the input phosphor. Overall brightness gain of the II is achieved through the acceleration and kinetic energy increase of the electrons impacting on the output phosphor (known as electronic or flux gain) as well as the geometric area reduction of the electron density from the large area input phosphor to the small area output phosphor (known as minification gain, equal to the ratio of the input to output phosphor areas, or ratio of the square of the diameters). The combination of electronic and minification gain results on the order of 5000X increase in brightness. Variable brightness gain occurs with a change in the input phosphor active area; as the field of view (FOV) is reduced, the minification gain is reduced, decreasing the overall brightness gain (and vice-versa). Optical coupling of the output phosphor to a TV camera or photospot, cine, or other light detector allows the detection of the image and subsequent display.
Control of the II "speed" is achieved with the inclusion of a light-limiting aperture in the conjugate lens system. For situations requiring high dose (i.e., digital subtraction angiography), the aperture diameter is reduced (large f number), while for low-dose fluoroscopic applications (i.e. upper GI fluoro), the aperture diameter is increased (small f number).
Figure E illustrate a typical overtable II/TV system, housing, carriage (allows vertical and horizontal positioning, and table (the x-ray tube is mounted under the table with a fixed geometry relative to the II detector). Figure F shows an internal cross-section of the II, and important structures including the image intensifier envelope, the input phosphor comprised of structured cesium iodide (CsI) scintillator material, the photocathode comprised of a light-sensitive, electron emitting material (Sb2Cs3), the electron focusing electrodes, the anode structure, the output phosphor comprised of zinc cadmium sulfide (ZnCdS:Ag), the tandem conjugate lenses with light limiting apertures and partially silvered mirror (to partially reflect light), and light photon detectors (film, TV camera, CCD camera) to capture the output image and convert into a useful image for viewing.
|Figure E. Image intensifier-TV system components are shown for a conventional fluoroscopic room.|
|Figure F. Cross section of the II-TV system illustrates the various components that are used to create the highly amplified output light image. A four stage process (x-rays to light; light to electrons; electrons to light; light to electronic signal) is shown.|
Image Intensifier Size
Image intensifiers come in a range of input field of view (FOV) diameters for diagnostic imaging applications, from 6 inches (15 cm FOV) to 16 inches (40 cm FOV), and many dimensions inbetween, depending on the type of imaging procedure. II's have a spherical input phosphor structure, with a curvature designed to withstand the large force on the II enclosure, resulting from the internal vacuum required for operation. The output phosphor dimension is typically about 1 inch (2.54 cm) diameter. The difference in size between the input and the output results in minification of the output image, whereby the electrons emerging from the photocathode of the input are focused and minified during acceleration through the evacuated tube. Brightness gain achieved by minification is equal to the area of the input phosphor to the area of the output phosphor, resulting from the increased electron density and corresponding increased light intensity at the output phosphor. In Figure G, an illustration of two FOV examples demonstrate the electronic "magnification factor" that is available on most image intensifier systems.
|Figure G. An image intensifier can interactively change the input FOV from a large to a smaller area. Effects on the output image are described in the figure inset information. In addition, an aperture collimator within the x-ray tube collimator assembly must limit the x-ray beam to the active area of the II.|
Image intensifiers can electronically vary the size of the input radiation field of view whilst keeping the output field fixed, equal to 2.54 cm (1 inch), corresponding to the size of most television cameras. If the input field of view is halved, then the size of the patient being viewed is also halved which results in a two fold magnification of the image. This type of magnification, known as electronic zoom, doubles the spatial resolution performance. For example, the limiting spatial resolution for a 25 cm field of view and a conventional television camera (525 lines) is ~ 0.7 line pairs per mm; reducing the input field of view to half this value (i.e., 12.5 cm) by the use an "electronic zoom" would improve the limiting spatial resolution performance to ~1.4 line pairs per mm.
Figure H. Image intensifer TV system with 4 FOV diameters: 37 cm, 30 cm, 22 cm, 17 cm, and corresponding intrinsic resolution capabilities (bar pattern is taped next to input phosphor of the image intensifier). Top row shows full FOV image, and bottom row shows bar pattern magnified view.
If the input field of view is halved, then only one quarter of the input phosphor is being irradiated since the area is proportional to the square of the field of view. Halving the input field of view, while keeping all the other parameters constant, would reduce the image brightness to a quarter of the original brightness at the full field of view. To compensate for this effect, the amount of radiation that is incident at the input of the image intensifier needs to be quadrupled to compensate for reduction in exposed area. Automatic brightness control feedback circuits in the image intensifier / x-ray generator system accomplishes this with feedback signals to adjust the kVp (kV modulated with mA fixed), mA (mA modulated with kV fixed), or both (kV and mA are both modulated) to maintain the brightness at the output phosphor. The consequence to the patient is an increase in the dose when the "magnification mode" is utilized, but also spatial resolution enhancement.
Pelvis Phantom (Fluoroscopy)
|Figure I||Figure J||Figure K|
Figure I shows a single frame from a fluoroscopy run using an image intensifier diameter of 38 cm; Figure J shows an magnified image (electronic zoom) with an image intensifier diameter of 25 cm and Figure K shows further magnification achieved by reducing the image intensifier input diameter to 15 cm. Relative to the image in Figure I, the magnification in Figure J is x 1.5 (i.e. 38/25), and the magnification in Figure K is x 2.5 (i.e., 38/15).
The choice of technique in Figure I was 75 kV and 2.4 mA, which resulted in an entrance skin air kerma of 35 mGy/minute. When the image was magnified by a factor of 1.5 (Figure J), the system increased the x-ray tube voltage to 85 kV and used a tube current of 2.7 mA, which increased the entrance air kerma rate to 50 mGy/minute. When the image was magnified by a factor of 2.5 (Figure K), the system further increased the x-ray tube voltage to 94 kV and used a tube current of 2.8 mA, which increased the input air kerma to 61 mGy/minute.
Use of magnification modes in fluoroscopy is usually associated with an increase in the choice of x-ray tube voltage for two reasons:
(a) higher voltages will reduce the entrance skin air kerma which needs to be kept below 90 mGy/minute (10 R/min) for regulatory purposes. Adjusting the x-ray tube voltage with increasing magnification resulted in only relatively modest increases in the entrance air kerma rate (35 mGy/minute -> 50 mGy/minute -> 61 mGy/minute).
(b) the tube current needs to be kept below ~5 mA to minimize the power input into the x-ray tube anode permit continuous fluoroscopy operation without overheating the x-ray tube. The increased in power input to the anode (power is kV x mA watt) was also relatively modest (190 watt -> 230 W -> 260 W).
By contrast, maintaining a constant x-ray tube voltage with an increase in magnification of x 2.5 would have required an increase in entrance skin air kerma (and power loading to the x-ray tube anode) of 625% (i.e., 2.5^2) because of the six fold reduction in exposed area of the input phosphor.
Pelvis Phantom (Digital Photospot )
|Figure L||Figure M||Figure N|
Figure L shows the digital photospot image that corresponds to the fluoroscopy frame depicted in Figure I (i.e., 38 cm diameter field of view). The choice of techniques used to acquire this image was 65 kV/9 mAs, and the corresponding entrance air kerma rate was 1.4 mGy. Increasing the magnification by a factor of 1.5 by reducing the field of view to 25 cm (Figure M) resulted in a digital photospot technique of 65 kV/18 mAs and an entrance air kerma of 2.8 mGy. A further increase in image magnification to x 2.5 by use of the 15 cm field of view (Figure N) resulted in a digital photospot technique of 65 kV/33 mAs, and an entrance air kerma of 4.8 mGy.
In digital photospot imaging, reducing the field of view does not normally require an increase in the x-ray tube voltage for the following reasons:
(a) Increased x-ray tube voltage will reduce the amount of image contrast and the corresponding contrast to noise ratio. Maintaining the contrast to noise ratio is desirable as this improves lesion detectability.
(b) Patient dose consideration is of much less concern than in fluoroscopy for several reasons. Diagnostic image quality is normally the paramount concern, and one does not wish to compromise diagnostic performance by using too little radiation. Furthermore, there are no dose limits in radiography, which is used for diagnosis, whereas there are dose limits in fluoroscopy (entrance air kerma must normally be < 90 mGy/minute). Even though the patient dose per frame is high compared to fluoroscopy, the total number of photospot images acquired is very low.
(c) There are no x-ray tube heating problems in digital photospot imaging. The total energy deposited into the anode is a product of the power (kV x mA) and the total exposure time (s) (i.e., Energy (J) = kV x mAs). The total energy deposited in the three examples shown above ranges from 0.6 kJ (65 kV and 9 mAs) to 2.2 kJ (65 kV and 33 mAs). The anode tube capacities of x-ray tubes in typical fluoroscopy/radiography imaging systems is typically hundreds of kJ and x-ray tube heating is generally not an important issue.
In contrast to the variation of x-ray tube voltage in magnification fluoroscopy, the x-ray tube voltage is normally kept approximately constant in magnification modes for digital photo spot imaging. As a result, the entrance skin air kerma and energy deposition into the x-ray tube anode are approximately inversely proportional to the exposed area of the input phosphor of the image intensifier. Halving the field of view, which doubles the magnification and corresponding spatial resolution, would be expected to (approximately) quadruple the entrance air kerma in digital photospot imaging.
Skull Phantom (Fluoroscopy)
|Figure O||Figure P|
Figure O shows one frame of a fluoroscopy run of a head phantom obtained using a field of view of 25 cm. The radiographic techniques were 74 kV/2.2 mA, and the entrance air kerma rate was 26 mGy/minute. Reducing the field of view (Figure P) from 25 cm to 16 cm (i.e., increasing magnification by a factor of 25/16 or 1.6) resulted in the use of 92 kV/2.8 mA and an entrance air kerma rate of 49 mGy/minute.
In this example, note that the Figure O (25 cm field of view) includes a small region around the skull that consists of the image intensifier being directly irradiated by the x-ray beam. This is generally undesirable (see collimation section below), and will result the selection of technique factors that take into account the relatively large detected signal from these directly irradiated regions. Predicting changes in the selected radiographic techniques (kV/mA), patient dose and x-ray tube loading under such conditions is particularly tricky.