JP4229859B2 - X-ray detector for low dose scanning beam type digital X-ray imaging system - Google Patents

Description

  The present invention relates to a diagnostic X-ray imaging apparatus. More particularly, the present invention relates to a real-time scanning beam digital X-ray imaging system that improves resolution and reduces X-ray radiation by incorporating a large number of apertured collimation grids and a segmented X-ray detector array. .

  As therapeutic techniques develop, real-time X-ray imaging in medical procedures is increasingly desired. For example, many electrophysiological procedures in cardiology, peripheral vascular procedures, urological procedures, and orthopedic procedures rely on real-time radiography.

  Unfortunately, current clinical real-time X-ray equipment irradiates both patients and participating staff with high levels of X-rays. The US Food and Drug Administration (F.D.A.) reports anecdotal evidence of acute radiation sickness in patients and the problems of occupationally overexposed physicians (Radiologic Health Bulletin, Vol. 26, No. 8, August 1992).

  Conventionally, many real-time X-ray imaging systems are known. These are equipped with a system based on an X-ray fluoroscope, where X-rays are irradiated on the object to be examined, and shadows caused by X-ray opaque objects in the object are opposite to the X-ray source on the object. Displayed on the X-ray fluoroscope located on the side. Scanning x-ray tubes have been known in connection with fluoroscopic techniques since at least the early 1950s. See Moon's "Fluorescence Perspective Amplification and Enhancement with Scanning X-Ray Tubes," Science, October 6, 1950, pages 389-395.

  Scanning beam digital X-ray imaging systems are also well known in the art. In this system, an X-ray tube is used to generate X-rays. By generating an electron beam inside the X-ray tube and focusing it on a small spot on a relatively large anode (transmission target) of the tube, X-rays are generated from this spot. The electron beam is deflected (electromagnetically or electrostatically) into a raster scan pattern across the anode. A small X-ray detector is disposed at a predetermined distance from the anode of the X-ray tube. The detector converts the X-rays impinging on it into an electrical signal proportional to the detected X-ray flux. When the object is placed between the X-ray tube and the detector, the X-rays are attenuated and dispersed by the object according to the X-ray density of the object. If the x-ray tube is in scan mode, the signal from the detector is modulated in proportion to the x-ray density of the object.

  Examples of conventional scanning beam digital X-ray systems include Albert US Pat. No. 3,949,229, Albert US Pat. No. 4,032,787, Albert US Pat. No. 4,057,745, Albert US Pat. No. 4,144,457, Albert US Patent No. 4,149,076, Albert US Patent No. 4,196,351, Albert US Patent No. 4,259,582, Albert US Patent No. 4,259,583, Albert US Patent No. No. 4,323,779, Albert US Pat. No. 4,465,540, Albert US Pat. No. 4,519,092, and Albert US Pat. No. 4,730,350.

  In a typical conventional example of a scanning beam digital X-ray system, the output signal from the detector is applied to the Z-axis (luminance) input of the video monitor. This signal adjusts the brightness of the screen. The x and y inputs to the video monitor are derived from the same signal that deflects the x-ray signal of the x-ray tube. Therefore, the brightness of a point on the screen is inversely proportional to the absorption of X-rays that reach the detector from the X-ray source through the object.

  The medical x-ray system is operated at the lowest level of x-ray dose that meets the resolution requirements suitable for the procedure. Thus, radiation dose and resolution are limited by the signal to noise ratio.

  As used herein, the term “low radiation dose” means an X-ray exposure level of about 2.0 R / min or less that reaches the patient during operation.

  The time and region distributions of X-ray photons follow a Poisson distribution and have unavoidable randomness. Randomness is expressed as the standard deviation of the average radiation flux and is equal to its square root. The signal-to-noise ratio of the X-ray image under these conditions is therefore equal to the average radiation flux divided by the square root of the average radiation flux, and for an average radiation flux of 100 photons, the noise is ± 10 photons, The noise to noise ratio is 10.

  Thus, the spatial resolution and signal-to-noise ratio of the X-ray image formed by the scanning X-ray imaging system is highly dependent on the size of the detector sensitivity region. As the aperture area of the detector is increased, more divergent light is detected, the effective sensitivity is increased, and the signal to noise ratio is improved. At the same time, however, as the “pixel” size (measured in the plane of the object being imaged) increases, the larger detector aperture decreases the adjustable spatial resolution. This is necessary because most objects that are imaged in the medical field (eg, the internal structure of the human body) are some distance away from the x-ray source. Therefore, in the prior art, the detector aperture size must be selected to balance resolution and sensitivity, and both resolution and sensitivity could not be maximized simultaneously.

  In medical imaging technology, the patient dose, frame rate (number of times per second to scan the object and update the image), and the resolution of the object image are the key parameters. It is. High x-ray flux easily provides high resolution and high frame rate, but will result in unacceptably high x-rays for patients and participating staff. Similarly, even if the image is not visible and the refresh rate is insufficient, the dose can be reduced if this is allowed. A preferred medical imaging system must simultaneously achieve all of the low dose, high resolution, and sufficient update rate of at least about 15 images per second. Therefore, systems such as the scanning beam digital radiography system described above have relatively long exposure times and the same can always be said for real patients, but keep the X-ray dose received by patients at a minimum. Not suitable for diagnostic medical procedures that must be.

  SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a scanning beam type digital X-ray imaging system that can be used in a medical diagnostic procedure that can be safely received by a patient.

  Another object of the present invention is to provide a scanning beam digital X-ray imaging system that provides a high-resolution image at a sufficient frame rate and minimizes the amount of X-ray irradiation to the inspection object. .

  Still another object of the present invention is to provide a scanning beam type digital X-ray imaging system with improved resolution at a site away from the surface of the X-ray source while reducing the level of X-ray flux.

  The above and many other objects and advantages of the present invention will be apparent to those of ordinary skill in the art in view of the drawings and the following description of the invention.

  A scanning beam type digital X-ray imaging system (SBDX) according to the present invention includes an X-ray tube having an electron beam source and a target anode. A circuit is provided for focusing or scanning the beam in a predetermined pattern through the target anode. For example, the predetermined pattern may be a raster scan pattern, a curved or S-shaped pattern, a spiral pattern, a random pattern, a Gaussian distribution pattern centered at a predetermined point on the target anode, or a short work. Other patterns that are effective for the above may be used.

  A collimating element (preferably formed in a grid) may be interposed between the X-ray source and the object to be irradiated with X-rays. The collimating element comprises, for example, a circular metal plate having a diameter of about 25.4 centimeters (10 inches) and a 500 × 500 aperture array in the center of the collimating element. The collimating element is preferably arranged immediately before the light emitting surface of the X-ray tube. Other shaped collimating elements can also be used. In one preferred embodiment of the present invention, each of the collimating element apertures is configured to be directed toward (or toward) a detection point on a surface located at an arbitrary distance from the collimating element. ing. The distance is selected so that the object to be irradiated with X-rays is placed between the collimating element and the detection point. The function of the collimating element is to form thin X-ray beams (arranged like pixels) that are all directed from a point on the anode of the X-ray tube towards the detector.

The center of the segmented detector array having an array of detector elements (preferably an array of regions such as DET _X × DET _Y rectangles or squares, more preferably a relatively round array) is located at the detection point. Yes. The detector array preferably comprises a plurality of closely packed X-ray detectors. Such an array is designed, positioned and used in accordance with the present invention to provide high sensitivity without compromising resolution, so it has a resolution comparable to or better than conventional X-ray equipment. In addition, an X-ray system with an irradiation amount that is at least an order of magnitude less than that of the conventional X-ray system can be realized. This feature of the invention has important implications in medicine and other fields. The dose to patients or participating medical staff in the current procedure is reduced. Unexpected measures are possible due to the danger of exposure.

  The output of the detector array is an intensity value for each element of the detector array when the x-ray beam is emitted through an aperture in the grid. Since each aperture is located at a spatially different point with respect to the inspection object and the detector array, a different output is obtained from the detector array for each aperture through which the X-ray beam passes. The output of the detector array can be converted to an image in various ways. One method is to simply synthesize the array output, ie, sum and further normalize the array element intensity values corresponding to each scanned aperture. The output array can then be used to drive a video or other display. More preferred are the multiple image composition method and the multiple output composition method described later, which provide improved visual output.

  The SBDX imaging system can also be used for stereoscopic imaging in which a collimation grid having two groups of openings is used. In this case, the first group of openings is directed to the first detection point where the divided first detector is located, while the second group of openings is the first group where the divided second detector is located. Directed to two detection points. A stereo image is created by forming two images with two segmented detectors and using conventional stereo display methods.

  The SBDX imaging system can also produce enhanced images of materials that exhibit different x-ray transmissions at different x-ray photon energies. Thus, for example, microcalcifications that are precursors of breast cancer can be imaged. By configuring the grid and / or anode to emit two or more x-ray beams each having a different x-ray energy spectrum and directing each group to a detector array (multiple detector arrays can also be used) Since it is possible to image that the transmittance of the inspection object is different at various X-ray photon energies, it is possible to emphasize only a portion showing different X-ray transmittance inside the inspection object. For example, if optimized for calcium detection, the imaging system is a powerful tool that can detect breast cancer and other abnormalities early.

  Maximize sensitivity without sacrificing resolution obtained using a low surface area detector by utilizing a split array that blocks all collimated x-ray beams and processing the array output Can be provided. An undivided detector of the same size as the divided array has the same sensitivity but lower resolution.

  In addition, subsampling methods may be used to process data from the array detector, reducing system complexity, required processing speed, and energy consumption while providing substantially the same image quality.

  The system described here is described in US patent application Ser. No. 08 / 008,455 (CAM-003) filed Jan. 25, 1993, “Catheter with X-ray sensitive photosensor detection device. ", The contents of which are incorporated herein by reference. This US patent application Ser. No. 08 / 008,455 is owned by the assignee of the present application.

Embodiment of the present invention

  It will be appreciated by persons skilled in the art that the following description of the invention is illustrative and not limiting. Other embodiments of the invention will readily occur to those skilled in the art.

Summary of Apparatus FIG. 1 shows a scanning beam digital X-ray imaging apparatus according to a preferred embodiment of the present invention. The scanning X-ray tube 10 is used as an X-ray source. As is well known in the art, a power supply of approximately 100 KV to 120 KV is used to operate the X-ray tube 10. A 100 KV power supply provides an X-ray spectrum over 100 KeV. Here, the 100 KVX line refers to this spectrum. The x-ray tube 10 includes a deflection coil 20 that is controlled by a scan generator 30 as is well known in the art. The electron beam 40 generated in the X-ray tube 10 scans from one of the grounded anodes 50 to the other in a predetermined pattern in the X-ray tube 10. For example, the predetermined pattern may be a raster scan pattern, a meandering or S-shaped pattern, a spiral pattern, a random pattern, a Gaussian distribution pattern centered on a predetermined point of the target anode, or a near future Other patterns that are useful for tasks. Presently preferred are serpentine or sigmoidal patterns that eliminate the need for "return" in raster scans.

  The electron beam 40 impinges on the anode 50 at a point 60, a cascade of X-rays 70 is emitted, and travels out of the X-ray tube 10 toward an object 80 to be examined with X-rays. In order to take full advantage of the operation of the device, conical X-ray photons are generated that radiate out just to cover the detector array 110. This is preferably accomplished by placing a collimation grid between the anode of the scanning x-ray tube and the detector. Therefore, the collimation grid 90 is disposed between the object 80 and the X-ray tube 10. The collimation grid 90 is designed so that only X-rays 100 directed toward the detector 110 pass through it. The collimation grid 90 does not move relative to the detector array 110 while the device is operating. Thus, whenever the electron beam 40 scans the anode 50, there is only a single x-ray beam 100 that passes from the anode 50 to the detector array 110.

  FIG. 2 shows the X-ray distribution when there is no collimation grid. The output of the detector array 110 is processed and displayed on the monitor 120 as luminance values at x and y positions on the monitor 120 corresponding to the x and y positions of the anode 50. This is accomplished by using the same scan generator that drives the x and y positions of the electron beam 40 and the electron beam in the video monitor 120. Also, image processing techniques can be used to produce computer manipulated images on a suitable display or photographic media.

  The inventive devices disclosed herein are generally measured at the part incident on the patient and from about 0.15 R / min at an update rate of 15 frames / sec to about 0.005 at an update rate of 30 frames / sec. It is a low radiation dose device that gives a patient a dose of up to 33 R / min. With this device, an update rate of 30 frames / second would be about 0.50 R / min throughout the body. The actual range of radiation dose at the patient surface for operation of the invention disclosed herein is from 0.15 R / min to 2.00 R / min.

X-ray tube FIG. 3 shows an enlarged structure of the grid and the anode. The anode 50 is preferably formed on a beryllium anode support 130 and assembled with a target layer of a material having good vacuum characteristics and the ability to resist high temperature and electron impact. In addition, aluminum or other materials that are relatively transparent to X-rays can be used for the anode support 130. The presently preferred target layers are, in order of preference: (1) a second layer of tantalum sputter deposited approximately 5 microns thick on a first layer of niobium sputter deposited approximately 1 micron thick on the anode support (this The structure is such that niobium has a thermal expansion coefficient intermediate between that of beryllium (anode support 130) and tantalum, thus suppressing microcracks due to thermal cycling of the target anode between the on and off states of the tube or (2) a sputter deposited layer of tantalum approximately 5 microns thick, (3) a sputter deposited layer of tungsten-rhenium approximately 5 microns thick, and (4) approximately 5-7 microns thick This is a sputter weld layer of tungsten. Tantalum, tungsten, and tungsten-rhenium are preferred for the anode 50 because they have relatively high atomic numbers and densities and readily emit X-rays when irradiated with an electron beam. The high melting point of 3370 ° C. and good vacuum characteristics of tungsten are suitable for the high temperature and ultrahigh vacuum conditions in the X-ray tube. Tantalum and tungsten-rhenium have similar properties as known to those skilled in the art. The thickness of the anode layer is selected to be approximately equal to the distance required for the anode layer to effectively convert 100 KV electrons to x-rays. Beryllium is preferred for the anode support 130 because it is strong and does not significantly reduce or dissipate X-rays emitted from the anode 50. The thickness of the beryllium anode support 130 is preferably about 0.5 cm. The anode support 130 is subject to the physical constraints that it must be strong enough to resist pressure gradients with an external 1 atmosphere, but it should be as thin as possible.

  The collimation grid 90 comprises, according to a preferred form of the present invention, an array of apertures 140 each oriented or oriented in the direction of the detector array 110. That is, the holes in the collimation grid 90 are not parallel to each other. For use such as chest X-ray applications, the grid 90 is positioned from 0 ° in the center of the collimation grid 90 with respect to the front surface 260 of the collimation grid 90. An angle between about 20 ° at the end of the can be made. In use of the present invention in breast examination applications, the grid 90 can be assembled such that the opening makes an angle with the front surface at a range of 45 ° at the end of the grid. The number of openings 140 in the collimation grid 90 can correspond to, for example, the number of pixels from 500 × 500 to 1024 × 1024 at the center of the round collimation grid 90, and the resolution of the device can be determined. Also, an aperture smaller than the number of pixels can be used in connection with the sub-sampling technique described later. The thickness of the grid 90 and the dimensions of the openings 140 are the distance from the X-ray tube 10 to the detector array 110 (here, preferably 91.4 cm (36 inches)) and all undirected to the detector. The need to reduce x-rays is determined by the dimensions of the detector elements 160 (not shown in this figure) of the detector array 110. Although not critical to the present invention, the openings 140 as viewed from the front surface 260 are preferably designed in a pattern of oblong columns and rows having a circular boundary of 25.4 cm (10 inches) in diameter. The aperture array can be a convenient layout relative to one another using the detection and convolution techniques outlined below to analyze the image of the object 80. This aperture array is called a “circular active area”. In the middle of the circular active area, the total number of openings is preferably 500 × 500 according to a preferred embodiment of the present invention. The non-opening portion 150 of the collimation grid 90 is designed to absorb shifted X-rays so that the shifted X-rays do not irradiate the object 80. This means that the X-rays that collide with the non-opening portion 150 are “1/2 value” (the amount of material required to reduce the X-rays that collide with the energy of this device, here 100 KeV, to 1/2). This is done by assembling the grid to show at least 10 times. The shifted X-rays irradiate the object and the staff with X-rays, but do not give useful information to the image. As shown in FIGS. 3A and 3B, the collimation grid 90 is formed from a number of sheets 143, 144 of X-ray absorbing material having apertures 140 therethrough to pass the X-ray beam 100 to the detector. be able to. The collimation grid 90 is preferably assembled by stacking 50 thin sheets of 0.0254 cm (0.010 inch) thick molybdenum and holding them together. Molybdenum is preferred because X-rays that are generated in the X-ray tube 10 and are not directed to the detector 110 are stopped before the object 80, including a human patient, of course, is useless and harmful. Lead or similar X-ray dense materials can also be used.

  The openings 140 in the collimation grid 90 are preferably square in cross section in order to obtain maximum packing density and to match the preferred square shape of the detector array element 160. Other shapes, particularly hexagons, can also be used. The square opening 140 has a size of 0.0381 cm (0.015 inch) × 0.0381 cm, resulting in a cross-sectional area of about 1/100 that of a conventional collimator used in fluoroscopy. preferable. This closer collimation achieves a smaller beam width for the x-ray beam 100. This means that the cross-sectional area of the detector surface can correspondingly be made smaller than in conventional devices. As a result, the X-rays scattered by the object are not detected by the detector, and the image is clouded as in the conventional apparatus using a detector having a relatively large surface area. Absent.

  A preferred method of assembling the collimation grid 90 is by photochemical milling or etching. Photochemical milling is preferred because it is cost effective and accurate. According to this method, a set of 50 photomasks is made to etch holes and gaps in 50 sheets of 0.0254 inch (0.010 inch) thin sheet material. The etched sheets are stacked, arranged and held together to form a grid assembly having a number of stepped openings, each having a predetermined angle with respect to each sheet. FIG. 3A shows a variation of the original collimation grid 90. This variation includes a number of X-ray absorbing sheets 143 having individual openings of a constant cross section (but not requiring the cross section to be constant). The resulting aperture 14 is stepped as shown, but passes the x-ray beam 100 to the detector array 110. The modification shown in FIG. 3B is quite similar to that shown in FIG. 3A except that the individual openings formed in the X-ray absorbing sheet 144 are themselves stepped. These stepped openings can be made by milling or chemical etching from each side of the sheet 144 with slight misalignment, as will be apparent to those skilled in the art, in the form shown. The shape of FIG. 3B results in less X-ray energy absorption within the stepped openings 140 of the collimation grid 90, resulting in the same X-ray flux at the end of the X-ray beam 100 as in the variation shown in FIG. 3A. It is more preferable because it cannot be reduced to a certain extent.

  One preferred method of holding the etched sheet (s) forming the grid 90 is shown in FIG. The etched sheet 91 (preferably 50) is provided with an alignment hole or an alignment opening 94, respectively. Alignment pegs 95 are placed in the respective alignment openings 94 to align the etched sheet 91. The assembly of the seat 91 and the peg 95 is disposed in the aluminum ring 359. The aluminum ring 359 is provided with a vacuum port 370 that can be sealed with a pinch-off 375. The 0.1 cm thick aluminum sheet 365 is bonded and sealed to the upper surface 380 of the ring 359 with a vacuum adhesive. An aluminum sheet 360 is similarly bonded to the lower surface 385 of the ring 359. Partially evacuated through port 370 and port 370 is sealed at pinch-off 375 as is well known in the art. In this manner, the relatively X-ray transparent aluminum sheets 360, 365 provide a clamping action that helps to hold and orient the etched sheets 91 as the grid assembly 90 together.

  The opening 140 furthest from the center of the grid 90 has a stepped surface and is preferably square in cross section. X-rays are generally unaffected by the unevenness of the passage due to the stepped surface, and even if the x-rays are scattered, they will not affect the combined beam so much that they can be measured. As described above, the material used for the collimation grid 90 is currently preferably molybdenum, brass, lead, or copper with added molybdenum. The preferred error for the hole location is ± 0.00127 cm (0.0005 inch) from center to center, excluding cumulative error, and the hole size error is ± 0.00254 cm (0.001 inch). It is.

  Other methods of making a usable collimation grid 90 include electron beam machining, drilling, mini-machining, laser drilling. Drilling and laser drilling have the disadvantage of creating rounder holes than square holes. A circular opening works well but is currently not preferred.

  More details of the preferred scanning x-ray tube 10 are shown in FIGS. The electron gun 161 is located on the opposite side of the front surface of the X-ray tube 10 and operates at a potential of about −100 KV to −120 KV. The grounded anode 50 is located in front of the tube, and the electron beam 40 moves between the electron gun 161 and the anode 50. The grounded electron aperture plate 162 is located near the electron gun 161 and has an aperture 163 at the center through which the electron beam 40 passes. The magnetic field focusing lens 164 and the deflection coil 20 position the beam spot on the anode 50 using automatic focusing well known in the art. The tube is assembled with a circular active area with a diameter of 25.4 cm (10 inches) where the electron beam 40 intersects the anode 50 with the electron beam deflected to about 30 ° at the end of the circular active area. It is done. When the beam is not “fired” through an aperture, it is preferable to stop generating the beam, resulting in power savings of up to about 25%.

  FIG. 5 shows a cross-sectional view of the front portion of a suitable X-ray tube 10. The inside of the X-ray tube 340 maintained in a vacuum is the rear side of the anode 50. The anode 50 is a coating of anode material as described above. The front of the anode 50 is a 0.5 cm thick beryllium anode support 130. The front of the beryllium anode support 130 is a cooling jacket 350 that is preferably 0.4 cm thick and can be adapted to flow water or forced air. Aluminum grid supports 360, 365 are etched to a thickness of 0.1 cm and help support a collimation grid 90 which is preferably 1.27 cm (0.5 inch) thick.

  When X-ray tube 340 is used, only one opening 140 in collimation grid 90 will pass a significant amount of X-rays at any given moment. According to a preferred embodiment, the electron beam 40 can be turned off when the electron beam 40 is not located directly in front of the opening 140. Thus, the X-ray tube can operate effectively in the scan pulse mode to reduce power consumption and reduce wear and tear on the target anode 50.

Stereoscopic X-ray Imaging In FIG. 6, according to another preferred embodiment of the present invention, a grid having a plurality of focal points can be provided so that stereoscopic X-ray imaging can be obtained. For example, if all other rows of apertures in grid 90 are directed to focal point F1 (92) and the remaining apertures are directed to focal point F2 (93), the first sensor array is placed at F1 (92). By placing the second sensor array at F2 (93), it is possible to scan the aperture in a raster or serpentine pattern, whereby a “line” of data about the first sensor array and the second sensor A “line” of data about the array can be created. By repeating this, two completed images can be assembled as seen from two different points F1 and F2 in space, thereby providing a conventional stereoscopic imaging display device that provides a stereoscopic X-ray image. You can display them. FIG. 3C shows how a three-dimensional collimation grid is constructed from within the layer 144 of X-ray absorbing material. In this case, the apertures 140A, 140B provide a separate path along the “leg” of the “V” for the X-ray beams 100A, 100B, as shown, and are actually shaped like “V”. It can be. However, it is not a requirement that the openings 140A and 140B be connected as shown in the figure, but the advantage of the “V” shaped opening where the X-rays are incident at the apex of the “V” is the detection of both. The instrument is illuminated at the same time, and "V" is to act as a signal separator for X-rays going to F1 and X-rays going to F2. This halves the power required for the beam and deflection current. Although the price is lower, scattering and the resulting cloudiness of the image increase.

In order to achieve a resolution of several lines / mm at the array detector object plane, the spatial resolution limit is mainly determined by the detector size, as required in some medical applications. . This can generate very high power levels that would be required in today's X-ray tube technology to get very well directed X-ray radiation, and combined X-ray directions. This is because it is impossible to develop a means of attaching.

  If the detector is made smaller than the area intersecting the emitted X-ray cone, a large percentage of the X-rays emitted by the source 50 are detected by the detector 250, as shown in FIG. 7A. Not. In practice, industrial beam scanning digital X-ray inspection equipment is designed in this way, where the radiation dose is normal and not a problem. As a result, the irradiation dose is increased to maintain the desired resolution.

  Thus, the resolution is improved with the use of a smaller detector, but when the area of the detector is equal to or exceeds the area defined by the cone of emitted X-rays intersecting the detector plane 270, the X illumination The dose is minimized.

  The resolution of the scanning X-ray imaging apparatus is that the cross-sectional area of the detector elements projected onto the object plane 280 (the plane perpendicular to the line between the center of the anode 50 and the center of the detector 110 where the object 80 is located). Determined by. Thus, as shown in the front view of the detector array in FIG. 8, if the large area detector is divided into smaller array elements, the large capture area of the detector assembly is maintained, while at the same time , Keeping the image resolution proportional to the size of the individual small detector elements 160.

  The resolution defined by the individual detector elements 160 is maintained by distributing and summing readings from the individual elements to the memory buffer corresponding to a particular location in the object plane 280, each address or pixel. Is done. Since the X-ray beam 100 moves individually across the collimation grid 90 located in front of the X-rays radiating to the anode 50, the address at which the output of a given detector element is added changes. The imaging geometry is shown in FIGS. 7B and 7C. FIG. 7B shows how a single beam position is divided into five pixels. FIG. 7C shows how consecutive beam positions are added to each other within a single pixel.

  In other words, the signal for each detector element is stored in the image buffer at a very small specific area on the object plane 280, ie a memory address corresponding to a single pixel. Thus, the memory storage address for each detector element changes to the position of the scanned X-ray beam in a form in which each pixel of the memory is arranged to include the sum of the radiation that has passed through a particular spot on the object plane 280. In this method, the resolution of the device is determined by the size of a single detector element, while the sensitivity of the device is highest since substantially all the X-rays reaching the detector plane 270 are recorded.

  A further advantage of the imaging geometry of this array detector is that the object plane 280 is narrowly defined. Structures that exist in front of or behind it will become hazy (out of focus). X-rays from the first opening 141 and the second opening 142 pass through the object surface 280 at a distance SO from the openings 141 and 142, and pass through the surface 281 twice the distance SO from the openings 141 and 142. The line is shown in FIG. 7D. As can be easily seen, the resolution obtained at twice the SO falls to about ½ of the resolution obtained at SO. This feature provides improved placement and visualization of detailed structures on the surface of interest 280, while providing a sufficient depth of field that can be changed by the geometry of the device.

  The array of the presently preferred embodiment is a 96 element pseudo-circle array of 0.135 cm square detector elements arranged in a circle having a diameter of about 1.93 cm (0.72 inches). There is no need for such a large size, and it is also possible to have three or more detectors arranged so that they are not all in one line within a circle with a radius equal to 0.135 cm, here the side length of one detector. .

X-ray detectors Conventional image intensifier tube technology is basically constrained by system sensitivity limits. The thickness of this usable scintillator material is limited by its light transmission characteristics. Typically it is made thick enough to capture about 50 percent of the incident x-ray photons. Only half of the emitted photons reach the photocathode. In the photocathode, only about 10 percent of the incident photons produce photoelectrons. Thus, only 2.5 percent (0.5 × 0.5 × 0.1) of the incident X-ray photon energy is maintained in the image intensifier system. In addition to this limited conversion factor, photons are scattered longitudinally by the scintillator material, causing blurring that reduces the resolution of the system at a given dose level.

  One of the fundamental objectives of the present invention is the SBDX imaging system that ensures that the inspection object is exposed to the lowest possible x-ray level while achieving the appropriate image quality required for the procedure being performed. Is to provide. This means that the system used to detect X-ray photons emerging from the object must have the highest photon-electric signal conversion efficiency. To achieve this, the material used for the detector must be long enough to ensure that the length in which the photons fly is sufficiently long that no photons from the end far away from the incident x-rays enter. Don't be. For example, in order to maximize the detector output, photon energy must be properly wasted in the material. There are many types of detectors that can be used in the SBDX system described herein. Preference is given to scintillators in which X-ray photon energy is converted into visible energy and light intensity is converted into electrical signals with a photomultiplier tube, photodiode, CCD or such device means. Since each pixel of the SBDX image must occur in a very short time of about 140 nanoseconds, the scintillator material must have a high response and a minimum afterglow time. Afterglow refers to a phenomenon in which the scintillator continues to emit light after the excitation incident X-ray emission ends. Plastic scintillators such as polystyrene mixed with organic matter are suitable in that they have the required high responsivity, but they have a relatively small X-ray photon interaction cross section and a small linear X-ray absorption coefficient. Value. As a result, considerable thickness is required to stop the X-ray photons. For 100 kV x-rays, a typical plastic scintillator requires a 28 cm (11 inch) thickness to capture 99% of the input x-rays. More preferably (in a preferred choice): (1) cerium-doped YSO (available from Yttrium Oxy-Orthosilicate, Airtron (Litton), Charlotte, NY), (2) cerium-doped LSO ( Lithium oxy-orthosilicate, available from Schlumberger), (3) BGO (available from Bismuth Germanate, Lexington Component, Beachwood, Ohio). YSO and LSO are excellent in that they can be used at room temperature. The BGO must be heated to about 100 ° C. to achieve a suitable light output decay time on the order of 50 nanoseconds. These scintillator materials do not require the same length as plastic scintillators and are effective at a length of 0.10 cm.

  In accordance with the presently preferred embodiment of the present invention, the SBDX array detector 110 is 96 closely bundled, located at a distance of 91.4 cm (36 inches) from the x-ray source 50. It consists of an array of 12 × 12 pseudo circles of individual X-ray detectors 160. (For 5 × 5 and 3 × 3 arrays, it can be thought of as a non-square array with a square detector (filling a circle around the X-ray target), see eg the bottom of Table 1) Collimation grid aspect ratio The shape, the distance from the X-ray source 50 and the size of the X-ray source 50 are about 0.96 inches in diameter at the entrance of the detector array 110 and have a total depression angle of 1.46 °. Give a quadrangular pyramid. Each scintillator 170 must therefore be about 0.06 inches from center to center in the plane of the detector. If the scintillator 170 has parallel sides, X-rays incident near the edge will not be able to travel the required distance without hitting the scintillator wall. Therefore, these X-rays pass to neighboring scintillators, and if they are not shielded, an apparent output from the wrong spatial location results, resulting in a degradation of the image quality of the object. As shown in FIG. 9A, to avoid this effect, each scintillator according to a preferred embodiment of the present invention is preferably tapered, and the interface 173 is the most distal of the incident X-ray 100 ′. Has a depression angle equal to the angle. This is particularly useful with long plastic scintillators. In the preferred example quoted above, each scintillator 170 is a preferred quadrangular frustum with a length of 28 cm having an entrance surface 172 with a diameter of 0.285 cm and a photodetector end surface 174 with a diameter of 0.37 cm. Each of the 81 detector bundles therefore has a polyhedral end, and each face touches the surface of a sphere placed in the center of the x-ray source 50.

  A further improvement in scintillator detection efficiency is achieved by the scintillator taper having an angle greater than the angle of the incident x-ray beam 100. Near the edge of the scintillator, photoelectrons and scattered X-rays generated by the interaction between incident x-rays and scintillator atoms can be lost to the shielding material that preferably separates adjacent scintillators. Such lost photoelectrons do not generate any light and do not behave as light emitted from the scintillator. Therefore, such disappearance reduces the efficiency of the scintillator. The maximum distance that the photoelectron travels depends on its energy and the material that moves. In the interaction between electrons and 100 kV X-rays in a plastic scintillator, no photoelectrons can travel a distance longer than about 0.01 cm. If the quadrangular frustum of the scintillator is made with a depression angle larger than the X-ray beam 100, the size is a short distance compared to the detector length (28 cm), 2 × 0.01 cm. The reduction in efficiency due to the lost photoelectrons is minimized because it is larger than the enclosed beam. In this case, the center of the sphere in contact with the surface is no longer coincident with the X-ray source 50 and is closer to the detector array 110.

  Scattered photons travel a longer distance than photoelectrons. To prevent these from escaping to adjacent scintillators, the taper value of the scintillator pyramid is set larger than that required for complete photoelectron capture to maximize the capture of scattered photons.

  Referring now to FIG. 9, in accordance with a preferred embodiment of the present invention, a light pipe that contacts each scintillator element 170 and optically couples each scintillator element 170 to a corresponding photomultiplier tube 190 or solid state detector. Alternatively, a fiber optical cable 180 is provided. An alternative scintillator 170 may be placed in physical proximity to a particular photodetector.

  FIG. 10 shows a preferred shape of the detector element 160. A sheet 200 that does not pass X-rays and has an opening 210 corresponding to each detector 160 is placed in front of the detector array 110. Each detector element 160 is surrounded by an enclosure 220 that does not leak light and does not pass X-rays. A light blocking window 230, preferably formed of an aluminum plate, is disposed in front of the enclosure 220 where light does not leak. The light blocking window 230 transmits X-rays. In the enclosure 220 where light does not leak, the scintillator element 170 is located close to the photomultiplier tube 190 electrically connected to the preamplifier 240. Preferably, the analog signal from preamplifier 240 is converted into a digital signal in a conventional manner by further processing.

  Alternatively, the scintillator can be placed directly or close to an array of photodiodes, a phototransistor, or a solid-state imaging device (CCD) to obtain a more crude and compact detector. In particular, when a solid state device such as a CCD is used, cooling using a Peltier cooler can be used to increase the signal to noise ratio of the device.

  Alternatively, the scintillator array can be placed directly or close to the photomultiplier tube that detects one or more positions that provide an output signal that identifies a position comparable to the light source as well as its size. .

  In other preferred embodiments, the sensor array may consist of a collection of aligned pencil detectors 285, for example as shown in FIG. In FIG. 11, a tapered scintillator 290 is aligned with the path of the X-ray beam 100, and the scintillator corresponding to a particular cross-sectional area of the beam 100 absorbs all X-rays within that cross-sectional area. The optical amplifying tube 300 is provided in physical proximity to the scintillator 290, and an electrical signal is generated corresponding to the absorption of X-rays by the scintillator 290. A solid element can also be used in place of the optical amplifying tube 300.

  According to this preferred embodiment of the present invention, the scintillator is prevented from escaping (or entering) light along the longitudinal direction and at the entrance surface, such as silicon dioxide to assist internal reflection within the scintillator. Covered with a material that reflects light.

  According to another preferred embodiment of the invention, each scintillator element 179 is separated from adjacent scintillator elements by a thin film 171 of a high radiopaque material such as gold or lead. The membrane 171 is preferably about 0.004 inches to 0.005 inches thick. The position of the membrane 171 between the scintillators 170 is shown in FIG.

  As shown here, the circular active area of the collimation grid 90 is wider than the 110 area of the detector array. In this way, all the X-ray pencil beams emitted from the respective openings 140 of the collimation grid 90 are collected in the detector array 110. On the other hand, each independent X-ray beam 100 branches and spreads like a flash beam.

Image Processing An important aspect of the present invention relates to the application of an image processing system to further reduce the required exposure dose. In practice, the signal from the detector is usually not used directly on the “Z” or luminance input of the video monitor. Instead, the digitized intensity of each pixel is stored in a separate address of the “frame store buffer”. Pixel addresses in the buffer can be accessed randomly, and numerical intensity values are handled mathematically. This function has many applications of image enhancement algorithms that can be applied, allowing the allocation of pixels of data from independent parts of the detector array.

  According to a preferred embodiment of the present invention, the SBDX image consists of about 250,000 or more pixels (corresponding to 500 columns and 500 rows in the center of the collimation grid 90) consisting of 500 columns and 500 rows. For the example described below, it is assumed that the scanning x-ray source is at the center P on a pixel at 100 columns and 100 rows of the collimation grid 90 at a certain moment. Further, for this embodiment, detector array 110 consists of a 3 × 3 array containing 9 segments 179 (FIG. 12), with each segment 179 blocking all X-ray radiation combined with a single pixel. Suppose the size is large. Obviously, other array shapes may be used as detailed herein.

  Numeric values digitized from individual segments of the detector array 110 are assigned the following pixel addresses:

Segment 1-99 columns, 99 rows
Segment 2-99 columns, 100 rows
Segment 3-99 columns, 101 rows
Segment 4-100 columns, 99 rows
Segment P-100 columns, 100 rows
Segment 6-100 columns, 101 rows
Segment 7-101 column, 99 rows
Segment 8-101 columns, 100 rows
The same data allocation pattern is repeated as the segment 9-101 column, 101 row scanning X-ray beam passes through all pixels.

  In the displayed image, the value of each pixel is equal to the sum of “n” portions. Here, “n” is the number of segments 179 in the array 110 (in this example, n = 9). When configured as shown here, the detector array 110 has a fixed operating distance for optimal focus and was not available in a conventional non-divided (undivided) detector array SBDX image system. It has the effect of providing an optimal focal plane.

  The following parameters must be taken into account in detector design.

    1. The size and shape of the parallel beam from the X-ray source (anode target) 50.

2. Distance between x-ray source 50 and detector array 110; "SD"
3. The distance between the X-ray source 50 and the center of the object 80 of interest; "SO"
4). 4. Desired resolution or pixel size of the object 80 of interest In medical applications, the entire area of the array must be large enough to block x-rays from the collimation grid 90.

  In the SBDX system for the preferred embodiment of the present invention, the distance between the x-ray source 50 and the outlet 260 of the collimation grid 90 is about 0.894 inches (see FIGS. 3 and 5). The opening 140 is a 0.0381 cm (0.015 inch) by 0.0381 cm square. The spot size of the electron beam 40 on the anode 50 is about 0.010 inches in diameter. The detector array 110 is 36 inches from the anode 50. Thus, the beam width of the X-ray beam 100 is 2 * ARCHTAN ((spot diameter / 2) / ((aperture width / 2) + (spot diameter / 2)) * 2.271 cm (0.894 inch), In other words, at a distance of 91.4 cm (36 inches) from the anode 50, the irradiated X-ray beam diameter is 91.4 * TAN (1.6 °) cm. The instrument array 110 should be approximately 1 inch in this aspect of the preferred embodiment, for example, if the object to be photographed is located 9 inches from the anode and desired The pixel size of the object is 0.0508 cm (0.020 inches), the distance from the X-ray source to the detector is 91.4 cm (36 inches), and the size of the optical detector array is 2.54 cm. If it is 1 inch square, the size of the illuminated pixel at the detector plane 270 is simply (SD / SO) * pixel size at the object, or 0.080 inch. Dividing 54 cm (1 inch) by 0.2032 cm (0.080 inch) gives the desired resolution using a squared detector with 12 to 13 segments on the surface. In addition, many other shapes will be used depending on the circumstances in which the SBDX system is used.

  Outside the surface SO of optimal resolution (280 in FIG. 7D), the resolution deteriorates by half at 0.5 × SO and 2 × SO (281 in FIG. 7D). This allows a reasonable depth of focus for many applications. In some applications, such as a human heart image, degraded focus outside this depth range is seen as an advantage. Blur of details outside the region of interest increases recognition of details within the region of interest.

  Many methods can be used to obtain a usable image from the data obtained as described above. As mentioned above, a simple convolution can be used, but in this case the resolution is not optimized at all. Two additional methods are preferred to obtain maximum resolution and sensitivity from the data obtained. These are called multi-image convolution and multi-output convolution. In both cases, the following is assumed:

The collimation grid 90 has an aperture AP _X column and an aperture AP _Y row. The intersection of each row and column is a “pixel”. Those pixels outside the circular active area of the collimation grid 90 are treated as if they were “dark”, for example, as if there was no measured intensity of the image. During scanning, pixels that are not illuminated by the X-ray beam are similarly treated as if they were “dark” as if there was no measured intensity of the image.

Referring to FIG. 15, the pseudo-circular sensor array 110 has a maximum value in the DET _X column of the sensor elements 160 in the detector array and a maximum value in the DET _Y row of the sensor elements 160 in the detector array 110.

  ZRATIO is a real number between 0 and 1. If ZRATIO = 1, the focus is set on the sensor surface. If ZRATIO = 0.5, the focal point is set at the midpoint between the X-ray source and the sensor surface. PIXELRATIO is the number of pixels for the physical distance between neighboring sensors in a column or row. For example, if the distance between the pixel centers on the object plane 280 is 0.01 cm, the distance between the sensors on the detector plane 270 is 1.0 cm, and PIXELRATIO = 10 and FOCUS = ZRATIO * PIXELRATIO.

IMAGE is a DET _X × DET _Y dimensional data string containing intensity information for special scanning and corresponding to special pixels. PIXEL is a four dimensional array of dimensions AP _X × AP _Y × DET _X × DET _Y containing DET _X × DET _Y IMAGE data sequences obtained by scanning all (or part) of the aperture. PIXEL is refreshed after each scan according to one preferred embodiment of the invention. As the beam is scanned across the anode surface, it is effectively placed in front of the center of the selected aperture 140, “fired”, and then reset. Thus, an IMAGE column of data is obtained for each irradiation. While these images are assembled into displayable images that can be used in some direct way, high resolution and high intensity are obtained by combining them. A preferred method for combining the first images is called a multi-image convolution method. In this multi-image convolution method, an OUTIMAGE column of AP _X × AP _Y- dimensional intensity that can be displayed on a CRT or similar display means is formed by assigning OUTIMAGE (y, x).

ＡＰ_X×ＡＰ_YＩＭＡＧＥデータ列を有用な画像に結合させる第２の現在好ましい方法は、マルチアウトプット回旋法と呼ばれる。この場合、ＤＥＴ_X×ＤＥＴ_Yセンサのセンサアレイとともに、ＤＥＴ_X×ＤＥＴ_Yのデジタイザ（または同等なものや、多重のもの）および同数の画素合算回路が有る。各センサからデジタル化された値は、ＳＥＮＳＯＲ（ｊ，ｉ）と呼ばれる。最後のＯＵＴＩＭＡＧＥ列は、以下のように計算される。出力イメージ列ＯＵＴＩＭＡＧＥ（ｙ，ｘ）［ｙ＝１〜ＡＰ_Y、ｘ＝１〜ＡＰ_X］の各画素について、ＤＥＴ_X×ＤＥＴ_YソースイメージＳＥＮＳＯＲ（ｊ，ｉ）の各々からの１つの画素は、合計されて［ｊ＝１〜ＤＥＴ_Y、ｉ＝１〜ＤＥＴ_X］、目的のイメージ画素ＯＵＴＩＭＡＧＥ（ｙ−ｊ＊ＦＯＣＵＳ，ｘ−ｉ＊ＦＯＣＵＳ）となる。それから、各素子がＤＥＴ_X＊ＤＥＴ_Yで割られることにより、ＯＵＴＩＭＡＧＥ列の規格化が行なわれる。

A second currently preferred method of combining the AP _X × AP _Y IMAGE data sequence with a useful image is called the multi-output convolution method. In this case, there is a DET _X × DET _Y sensor array as well as a DET _X × DET _Y digitizer (or equivalent or multiple) and the same number of pixel summing circuits. The digitized value from each sensor is called SENSOR (j, i). The last OUTIMAGE column is calculated as follows: For each pixel in the output image sequence OUTIMAGE (y, x) [y = 1 to AP _Y , x = 1 to AP _X ], one pixel from each of the DET _X × DET _Y source images SENSOR (j, i) is And [j = 1 to DET _Y , i = 1 to DET _X ] to be the target image pixel OUTIMAGE (y−j * FOCUS, x−i * FOCUS). Then, each element is divided by DET _X * DET _Y to normalize the OUTIMAGE column.

  Further improvements in these techniques can be obtained by performing linear interpolation based on the fractional part of the FOCUS element.

  The advantage of the multi-image convolution method over the multi-output convolution method is that the former can select the optimal focus plane in software after taking the data, whereas the latter cannot. However, the latter allows faster operation when time is at a limit.

Reconstructing 3D Images from SDBX Data The SBDX system described herein can be used to create a set of continuous planar images that are used to form a tomographic 3D image of object 80. A set of images can be analyzed to create an image in three dimensions consisting of a series of images of varying depths by reanalyzing the data set with different FOCUS values. The natural FOCUS value used is n / DET _X or n / DET _Y , where n is an integer from 0 to DET _X or DET _Y , respectively. Typically, the focus value is analyzed to match the plane of interest in the object 80. For example, in the SBDX system shown in Table 1 below, the focal plane is close to a normal focal plane of 22.86 cm (9 inches) (optimal focal plane) and spaced approximately 2.54 cm (1 inch). Set in.

  The following formula shows that a series of planar images are located in terms of distance from the anode 50.

ここに、Ｆ_t（ＦＯＣＵＳ）＝ アノードから関心のある特定の焦点
面までの距離
Ｆ_d＝ 検出器から焦点面までの距離（アノ
ードから検出器機までの距離、Ｆ_t
より小さい）
λ_t＝ コリメーショングリッドの隣接した
開口の中心と中心の距離
λ_d＝ 検出器アレイ１１０内の隣接した検
出器１６０の中心の間の距離
である。

Where F _t (FOCUS) = the specific focus of interest from the anode
Distance to face
F _d = distance from detector to focal plane (ano
Distance from sensor to detector, F _t
Less than)
λ _t = adjacent to the collimation grid
Center of opening and center distance
λ _d = adjacent detection in detector array 110
The distance between the centers of the ejectors 160.

  When using a sub-sampling technique, the calculation remains the same and data from the “not skipped” collimation grid aperture is handled. However, even if there are no collimation grid holes in between, λt is the same.

Negative Feedback X-Ray Flux Control The BDX imaging system shown in FIG. 13 uses a negative feedback path 305 that controls the x-ray flux of the x-ray beam 100. Preferably, negative feedback from the sensor array controls the x-ray flux so that the sensor array always sees approximately the same x-ray flux level. In this way, when soft tissue (which is relatively transparent to x-rays) is scanned, the x-ray flux falls, reducing the overall dose to the patient (object). By using negative feedback flux control, contrast and dynamic range are improved. According to this embodiment, the differential amplifier 310 has an adjustable reference level 320 that can be set by the user. The negative feedback loop 305 performs negative feedback of the x-ray flux with respect to the x-ray tube.

Time Domain Scan Mode A time domain imaging system can also be implemented using the principles described herein. In such a system, the time to reach pre-measured x-ray flux from various pixels can be calculated and mapped. Next, negative feedback control can be used to remove or reduce x-ray flux from apertures corresponding to pixels that have reached a predetermined flux level for the scan period handled. In this case, the collected information is the time to the flux level, and the mapped and photographed information corresponds to time rather than intensity. The ability of such a system provides a much larger signal-to-noise ratio, improves contrast, makes it possible to dramatically reduce the x-ray dose to the object under examination and improve the dynamic range.

Multiple Energy X-Ray Imaging Mode According to one embodiment of the present invention, two or more groups of x-ray beams 100 travel toward one or more detector arrays. The first group of x-ray beams has a first characteristic x-ray energy spectrum. The second group of x-ray beams has a different second characteristic x-ray energy spectrum. By comparing the measured transmittance of the x-ray beams of the first group and the second group, the presence of an object in the object under inspection can be detected. The basic concept of differential x-ray photography is known in the art and is disclosed in US Pat. No. 5,185,773 (“Method and Apparatus for Nondestructive Selective Determination of Metals”). Is incorporated herein by reference.

  The two groups of x-rays can be generated in a number of ways. In one method, a special anode comprising a first material or first thickness of material near the first group of openings and a second material or second thickness of material near the second group of openings. Manufacture. Thus, the aperture associated with the first group emits x-rays having a first characteristic energy spectrum, and the aperture associated with the second group emits x-rays having a second characteristic energy spectrum. Alternatively, K-filtering (K-edge filtering) can be used by placing a filter material (such as molybdenum) in the portion of the opening 140 that produces a similar effect. In this case, the first group of openings comprises a first filter inserted therein, and the second group of openings comprises a second filter inserted therein. The second filter may not be a filter. As in the previous case, two groups of x-rays with different characteristic energy spectra are associated with two groups of apertures.

  When at least two groups of apertures are associated with different characteristic energy spectra, it is possible to detect microcalcifications (breast cancer) and other abnormalities that are not normally visible with broadband x-rays. For example, by scanning a first group of apertures forming a first image, scanning a second group of apertures forming a second image, and then enhancing their ratio by image division, Microcalcifications and other abnormalities can be detected in real time using a low dose scan x-ray imaging system. Similarly, a multi-detector array device can be used with a first group of apertures towards the first detector array and a second group of apertures towards the second detector array.

  Another embodiment of multi-energy imaging will now be described. Since the magnitude of the detected electrical pulse from the x-ray photon is proportional to the photon energy (kV), it is possible to separately count pulses coming from photons in two or more energy bands. The pulses are separated by intensity, then counted and processed separately to produce two or more separate images. These images can be displayed as a ratio.

  It is also possible to change the selected energy level on the fly in order to distinguish different density regions in the object. The effect of this embodiment is that it is more flexible than the above-described embodiment and does not require a special collimation grid, anode material or dual detector.

While a number of preferred embodiments have been described above for various configurations of the present invention, the following description describes a preferred SBDX imaging system according to the present invention.

Table 1
A. Grid shape Circle Diameter 25.4cm
Opening pitch 0.0508cm
Number of openings in diameter 500 (166 for subsampling type)
Grid area 506.45cm2
Number of openings about 196,350
(About 21,630 for sub-sampling type)
Cross-sectional shape of the opening Circle Space between the opening 0.081cm
Distance between anode surface and collimation grid output surface
2.5cm

B. Distance between source and detector 91.4cm
Optimum focal plane position 22.86cm from grid

C. Adjustable to scan frequency 30Hz

D. X-ray tube operating voltage 70-100kV

E. Overall shape of detector array Almost circular (see Fig. 15)
The shape of the input surface of the detector element is square. The size of the input surface of the detector element is 0.133 cm × 0.135 cm.
Number of detector elements 96 in a nearly circular array of 12 diameters
Array diameter 1.83cm
The total included angle from the center point of the detector to the outer diameter of the collimation grid 15.8 °
Field of view at optimum focal plane 19.05cm
Pixel size at optimum focal plane 0.038cm
Pixel size on detector plane 0.152cm (center-to-center detector
interval)
Resolution 13 line pairs / cm

Therefore, this SBDX imaging system using a segmented detector array has high resolution and high sensitivity, and at the same time, the irradiation dose to the object under examination is low. The system can also set an optimal focus at any point between the source 50 and the detector array 110 and can provide an effective working depth of field.

Beam Subscan Technique The following description relates to a particularly preferred embodiment of the present invention, which uses a beam subsampling technique to reduce the processing overhead of the computer.

  A standard video image uses 640 × 480 pixels and is updated at 30 Hz. This requires a pixel sample rate of about 12 MHz. Positioning the x-ray tube high voltage electron beam at exactly this rate behind 250,000 consecutive different apertures requires high accuracy and relatively high power consumption. Digitizing the signal from a large array of x-ray detectors at a rate of 12 MHz is likewise effective and consumes power. Thus, without a significant reduction in the spatial or temporal resolution of SBDX, a reduction in pixel sample rate below 12 MHz can lead to initial unit costs, operating costs due to power consumption, and wasted heat due to x-ray tubes. Useful for reducing cooling demands.

  Thus, mechanisms have been developed that provide substantially the same spatial and temporal resolution while reducing the pixel sample rate. This mechanism is referred to as sub-sampling and is obviously applicable to other configurations of SBDX, but is best performed in the SBDX embodiments described herein. The advantages of this embodiment are reduced power consumption, a simpler circuit for deflecting the electron beam in the x-ray tube, lower costs for creating the collimating grit 90, and calculations required to resolve the image of the object 80. And other effects that will be apparent to those skilled in the art.

In this embodiment, the collimation grid 90 is created with a number of openings less than 500, preferably AP _X = AP _Y = 166 (although other numbers can obviously be used). The effect of this reduction from a computational point of view will become apparent below. However, the effect from a manufacturing point of view is a much simpler structure, about 1/9 of the number of openings required for manufacturing. Since the number of apertures is reduced, manufacturing a grid with a higher deflection angle (ie, the angle that the aperture makes with respect to the front surface 260 of the collimation grid) creates the problem of having apertures that intersect adjacent apertures. It is easier. This is because when a 3D grid is manufactured, adjacent apertures in the 3D grid are directed toward different detector arrays, and therefore require a wider physical separation than non-stereo grids to prevent aperture crossing. Is particularly useful.

The opening (s) of the collimation grid are arranged in a circle of AP _X rows and AP _Y columns in a circle of maximum size. For calculation purposes, this is outside the circle that does not contribute to the information, ie a rectangle with dimensions of AP _X rows and AP _Y columns with elements that are always “dark” or not illuminated by x-rays. Can be handled as

The sensors 160 of the x-ray detector array 110 are arranged in a circular array of DET _X rows and DET _Y columns with maximum dimensions, as shown in FIG. The pixel sampling rate can be reduced by illuminating less than all of the collimation grid, ie, subsampling. Preferably, a collimation grid without openings that are not illuminated is used. To make an image using the detector array, it is necessary to bore the only the DET _Y th collimator in hand all DET _X th collimator hole and each row is irradiated in each row, therefore, the image is It can be assembled from image tiles of pixels, each of dimensions DET _X and DET _Y. This corresponds to a subsampling ratio of DET _X × DET _Y , but subsampling does not correspond to a sampling rate of 1 × 1. Therefore, the sub-sampling ratio can be adjusted from 1 to DET _{X in} the x direction (row) and from 1 to DET _{Y in} the y direction (column). According to this preferred embodiment, DET _X = DET _Y = 12 as shown in FIG.

  When a 12 × 12 detector is used and the subsampling ratio is 12, the image is much like David Hockney's light mosaic, and is made up of multiple non-overlapping images that are pasted together substantially. Created. The actual scintillator and detector are not completely perfect and do not respond exactly the same, so the x-ray pencil beam is not perfectly uniform and the collimation grid aperture is ideal The exact area is not exactly the same. And since a circular detector is used rather than a square detector, a certain degree of overlap is highly desirable in order to be able to average the detector non-linearity and noise.

  If the subsampling ratio is less than the size of the detector in the pixel (ie, less than 12 in this preferred embodiment), the image must be assembled from overlapping “tiles” that must be summed or averaged. Don't be. If the subsampling ratio (s) is not multiples of the detector dimensions (in pixels), or if the detector array is not a cuboid, a different number of samples are added to each pixel and a different division Is required to average each pixel. Techniques for handling numbers less than the ideal environment are well known to those skilled in the art and are not described here to avoid overcomplicating the description.

  In the following calculations, SSX represents the sub-sampling dimension in the X direction (row) and SSY represents the sub-sampling dimension in the Y direction (column). For example, if SSX = SSY = 1, there is no subsampling and the processing is exactly the same as in the other embodiments of the invention described above. Similarly, if SSX = SSY = 1 as in the present embodiment, the process returns to “light mosaic” that does not use pixel averaging. If SSX and SSY are 3, and the active area of the circle is 500 × 500, then a 166 × 166 aperture is scanned, 1/3 in the x direction and 1/3 in the y direction, and the resulting data Decreases by a factor of 9. If a 1/9 aperture is used for the entire time, there is no need for these calculations and a collimation grid need not be included.

  Therefore, to create an image, only 1 / (SSX * SSY) apertures (500 × 500 apertures) in the original collimation grid need to be used, ie by an electron beam for x-ray irradiation. It needs to be irradiated. If the frame rate is kept constant, i.e. 30 Hz, the number of motions of the electron beam is reduced by SSX * SSY, like the frequency response of the circuit driving the electron beam. The overall distance that the electron beam travels (and the number of scan lines) is reduced by 1 / SSY, so that the average beam velocity through the target anode is reduced by 1 / SSY. The pixel speed at which the image is reconstructed is the same as the collimation aperture speed (the speed at which the aperture is scanned or illuminated) and is similarly reduced by 1 / (SSX * SSY).

  According to this scheme, the number of samples averaged into each display pixel is (DDTx / SSX) * (DDTY / SSY). With the maximum sampling rate (SSX = DDTx and SSY = DDTy), only one digitizer sample is averaged for each display pixel (“light mosaic” mode). Sample averaging is important to smooth out non-uniformities in the beam, scintillator, detector and amplifier. The sub-sampling magnitudes (SSX and SSY) must be set to an appropriate level for the conditions presented to ensure acceptable image quality. This can be set by the user for the fly depending on the user's preference for the quality of the image and the conditions submitted by a particular set of environments.

  The detector array 110 shown in FIG. 15 is preferably an array of 96 individual detector elements 160 arranged approximately in a circular area having a diameter of about 1 inch. At the center of the array, there are 12 detectors (DDTx) in the vertical column, and at the center of the array, there are 12 detectors (DDTy) in the horizontal row column. The scintillator crystals are preferably supported by a “egg crate” structure consisting of a 0.005 inch thick stainless steel strip cut into a square horizontal cross section. The circle 400 of FIG. 15 with the entire scintillator crystal (hatched) positioned therein is preferably about 0.800 inch in diameter.

  The length of the scintillation crystal in detector array 110 is preferably 0.10 cm, and the input surface in front of it is preferably 0.135 cm × 0.135 cm. The scintillation crystal is preferably YSO, LSO or BGO, but other materials can be used as described above. The BGO must be heated to about 100 ° C. for a suitably reduced decay time (50 nS) for its light output in this application. Therefore, the resistance heating element may be provided.

  FIG. 17 illustrates a detector assembly 402 according to a preferred embodiment of the present invention. X-rays enter the lead shield 406 from above through the x-ray window 404. To allow the x-rays to exit the aperture of the collimation grid 90 and impinge on the detector array 110 while attenuating the scattered x-rays, the x-ray window 404 is preferably circular and has a diameter of 1.91 cm. is there. A light shield 408 is provided to shield the detector from stray light. The light shield 408 can be made from a thin sheet of aluminum or beryllium selected to attenuate light without substantially attenuating x-rays.

  The detector array 110 is positioned near a suitable heating element 410 for use with a BGO scintillator. The heating element 410 may be a resistive heating element designed to maintain the detector array 110 at an operating temperature of about 100 degrees Celsius. Photons emerging from the bottom 414 of the detector array 110 with a fiber optic imaging taper 412 are directed to a 96 channel photomultiplier tube (PMT) 416. The detector assembly 402 is enclosed in an outer housing 418 that does not leak light to prevent stray light from generating noise. Three shoulder screws 420 and three centering screws 422 are provided for planar and linear arrangements, as is well known to those skilled in the art. Rotational alignment is achieved by rotating the outer housing 418 relative to the PMT mount 426. The fiber optic taper 412 is commercially available from Collimated Halls, Inc., Campbell, Calif., And has a circular input aperture with a diameter of 2.03 cm and a circular output aperture with a diameter of 3.38 cm. The taper 412 fits the pitch dimension (0.06 inch) of each scintillator crystal against the PMT 416 dimension (0.10 inch). That is, the taper has a magnification of 1.667 times. A highly viscous optical coupling liquid (type 200), commercially available from Dow Corning, has a refractive index approximately equal to the refractive index of the glass, from scintillator crystal 160 to taper 412 and from taper 412 to PMT input surface 424. In order to maximize the light transfer efficiency, it is used on two sides of the taper as an optical coupling medium.

  The photomultiplier tube 416 is a 96 channel tube (one channel corresponds to each scintillation crystal) commercially available as model XP1724A from Philips. The photomultiplier tube 416 includes a fiber optic faceplate so that the spatial arrangement of the scintillation array is accurately performed to the PMT photocathode located at the PMT on the other face of the optic faceplate. The x-ray photons that hit one scintillation 160 produce a light pulse that is coupled to the PMT photocathode. This produces a corresponding electron pulse on the photocathode, which is amplified up to 1,000,000 times in one channel of the PMT diode structure.

  This PMT output pulse is coupled to the input of an amplifier in the 30 MHz band, and the output of the amplifier is in the range of 0.5 to 5.0 volts and is about 30 nsec long so that the pulse is differentiated. This eliminates the need for a DC restorer circuit that maintains the baseline reference voltage as the pulse rate changes.

  The output of the amplifier is input to a comparator, and the comparator outputs an output pulse having a constant magnitude regardless of the magnitude of the input. The reference voltage for the comparator is set to a value slightly greater than the noise output level so that the amplifier is not triggered at the noise output level. The amplifier chain is repeated 96 times, once for each scintillation crystal in the detector array. The output pulse of the comparator becomes new data for the data acquisition and image reconstruction system. As tests have shown, the prototype system can randomly count x-ray photons up to a rate of about 10 MHz. While embodiments and applications of the present invention have been described above, it will be apparent to those skilled in the art that many more variations than those described above are possible without departing from the inventive concepts. Accordingly, the invention is not limited except within the spirit of the appended claims.

It is a figure which shows the basic element of a low dose scanning beam type digital X-ray imaging system. It is a figure which shows distribution of the X-ray from the SBDX system without a collimation grid. 2 is an enlarged view of a grid and an anode of an X-ray tube for a low dose scanning beam type digital X-ray imaging system. FIG. It is a fragmentary sectional view of a collimation grid effective for the device of the present invention. It is a fragmentary sectional view of a collimation grid effective for the device of the present invention. It is a fragmentary sectional view of a collimation grid effective for the device of the present invention. It is a figure of the X-ray tube for low dose scanning beam type digital X-ray imaging systems. It is sectional drawing which shows the structure of the X-ray tube for low dose scanning beam type digital X-ray imaging systems. It is a figure of a stereoscopic scanning beam type digital X-ray imaging system. It is a figure of the X-ray source with an opening which cooperates with an undivided simple detector. FIG. 3 is a diagram of X-rays from one aperture of an apertured X-ray source cooperating with a split detector array. FIG. 3 is a diagram of X-rays from multiple apertures of an apertured X-ray source cooperating with a split detector array. FIG. 3 is a diagram of X-rays from two openings of an X-ray collimation grid that cooperates with an inspection object and a split detector array. FIG. 3 is an exposed surface view of a 5 × 5 detector array for a low dose scanning beam digital X-ray imaging system. FIG. 5 is a diagram of a 5 × 5 detector array for a low dose scanning beam digital X-ray imaging system. It is a figure which shows the scintillator element which concerns on one preferable embodiment of this invention. FIG. 2 is a diagram of detector elements for a low dose scanning beam digital X-ray imaging system. FIG. 5 shows an array of pencil detector elements for a non-planar detector array. FIG. 2 is a diagram of a 3 × 3 detector array for a low dose scanning beam digital X-ray imaging system. It is a figure which shows the basic element of the low irradiation amount scanning beam type digital X-ray imaging system using negative feedback in order to control X-ray flux. It is a perspective view which shows a grid seal apparatus. FIG. 4 is a diagram of a 96 element detector array arrangement according to a preferred embodiment of the present invention. It is a figure which shows the interaction of a collimation grid and a detector array. FIG. 2 shows a preferred embodiment of a detector device.

Claims (1)

A scanning x-ray tube having an electron beam generator, an anode, and an output surface adjacent to the anode;
Including a circuit for positioning the electron beam generated by the electron beam generator at a first plurality of selected points of the anode, wherein the electron beam is predetermined at the selected plurality of points. A scan generator that stays in a predetermined pattern for a limited time,
A collimation grid formed of a material that absorbs X-rays provided substantially parallel to and adjacent to the output surface, and having a substantially flat input surface and a substantially flat output surface;
An x-ray detector array having at least three separate x-ray detectors;
A processor for generating an image from a plurality of measured values measured by the X-ray detector;
A second number of independent pixels greater than the first number, and a display for displaying the image,
The collimation grid has a plurality of openings formed so as to penetrate the collimation grid, and the plurality of openings penetrates the output surface in the first region on the output surface of the collimation grid; Provided so as to be non-parallel to each other and to direct X-rays from the anode to the X-ray detector array, respectively.
Each of the X-ray detectors includes one detection input surface, and is arranged so that all of the detection input surfaces are not aligned.
The processor is configured to cause the X-ray detector array to generate a plurality of pixel signals representing a density of a portion of an object by scanning and send the plurality of pixel signals to a display. A digital x-ray imaging system of a sub-sampled scanning beam using the output from the x-ray detector.

Images