GB2376162A

GB2376162A - Low cost digital X-ray imaging system utilising a document scanning apparatus

Info

Publication number: GB2376162A
Application number: GB0106729A
Authority: GB
Inventors: Peter Coxon
Original assignee: Individual
Current assignee: Individual
Priority date: 2001-03-13
Filing date: 2001-03-13
Publication date: 2002-12-04
Also published as: GB0106729D0

Abstract

An X-ray imaging instrument typically comprises a source of X-rays a specimen mounted on a holder or manipulator, and detection means to enable visualisation of the X-ray image. Traditionally the detection means has been photographic, but there is an increasing trend toward digital detection and display. The present invention relates to a digital X-ray imaging apparatus and method to enable the conversion of readily available document scanning technology into a 2-dimensional digital x-ray imaging detector, and thereby provide for the convenient acquisition of digital X-ray images directly into PC and other computer desktop applications.

Description

Low Cost Digital X-ray Imaging System Description The present invention relates to an X-ray imaging apparatus, and more particularly to an apparatus employing digital image acquisition techniques.

An X-ray imaging apparatus used in radiology is conceptually a very simple device, basically a point source shadow graph but using X-rays instead of visible light. This the advantage of being able to see through/inside objects which are otherwise opaque to human vision.

Traditionally X-ray imaging apparatuses (See Figure 1) as used in radiology have comprised an X-ray source (1.1), the specimen to be examined (1.2) possibly mounted on a manipulator or positioning device, and a means of detecting X rays (1.3). X-rays from the source (1.1) which is typically a point source some microns in diameter pass through the specimen (1.2) to be examined and are absorbed by a varying amount depending upon the density and the mass attenuation coefficient of the material in the specimen. The X-ray flux intensity, therefore arriving at the detector (1.4) varies as a function of the material in the specimen, and if measured over area (or mapped) provides a 2 dimensional view through/inside the specimen.

Conventionally the most common means of X-ray detection has been a photographic film (1.4) often sandwiched between two sheets of phosphor material (1.5), which scintillate when irradiated with X-rays and thereby increase the exposure of the photographic film, augmenting the detection sensitivity of the detector system.

With the advent of modern digital imaging technology there has been an ever increasing move away from the use of photographic media toward electronic acquisition of imaging data.

This change in technology is being driven by several different considerations :- 'Convenience is a prime benefit, as images are acquired directly into a computer where they can be easily manipulated and archived. Clearly this avoids the messy and time-consuming process of film development, plus the additional inconvenience of storing large format photographic media.

There is also a considerable cost saving to be made in terms of running costs, as photographic media are expensive, and the computerized images need only be printed when the user is satisfied with the quality.

For some large area digital imaging detectors there is a claimed advantage in terms of picture quality and resolving capability.

'For some digital imaging detectors particularly those equipped with imaging intensifying devices in front of them, there can be a profound improvement in detecting sensitivity. An alternative way of stating this advantage is to say that for the same image quality the X-ray dose to the specimen is dramatically reduced. Clearly this is a very important advantage in the field of medical radiography where one strives to minimize the patient exposure to harmful ionising radiation.

Electronic X-ray image detectors have been developed along several distinct lines.

In its simples form (see Figure 2) such a aevice might comprise a TV camera (2.1) (nowadays a solid state CCD/CMOS camera) with an X-ray conversion screen placed in front of it. The X-ray conversion screen is simply a glass disk (2.2) coated with an X-ray phosphor (2.3) material, which scintillates under X-rays thereby converting X-ray photons (2.4) into visible photons (2.5) that the CCD can'see'.

Although the price of solid state CCD TV cameras has fallen quite dramatically in last 10 years, such that monochrome cameras may be purchased for as little as E100, the cost of such cameras converted for X-ray viewing is typically dramatically higher than this. An additional stage of sophistication may involve using an optical image-intensifying device between the phosphor screen and the TV, to gain additional sensitivity. The cost of such a device is currently many thousands of pounds.

A disadvantage of the type of detector described above, is that the image acquisition area for a low cost off-the-shelf CCD TV will be quite small typically-8 by-6 mm, which limits the range of usefulness for such a detector. For the analysis of many real-world specimens it will be necessary to acquire fields of view considerably larger than this (a human chest X-ray for example). The field of view for such a detector is often increased by employing a fibre-optic taper device between the X-ray screen/image intensifier, and the CCD TV. In this way the field of view may be increased by a factor of three to four. An example of a device employing all these elements, image intensifier and fibre-optic taper as described above is depicted in Figure 3.3. 1 represents the CCD TV, 3.2 a fibre optic taper device converting a large field of view to a smaller field of view matched to the CDD, 3.4 an image intensifying device typically employing an X-ray sensitive photocathode (3.41), which converts X-ray photons (3.5) to electrons (3.42), and then multiplying them in number many thousands of times by channel electron multiplier plates before being converted back to visible photons (3.3) again by a suitable phosphor material (3.43). There are variations on this theme, but the same general principle applies. However, fibre-optic tapers are also very expensive and add considerably to the overall cost of the device. More recently large area'Mega Pixel' cameras covering several square centimetres, with many millions of pixels, are being used in such X-ray applications, and although these all also currently very expensive, their price is likely to fall with the increasing popularity of digital photography.

For applications requiring much larger fields of view [e. g. 300x300m) such as medical radiography, where whole parts of a human body for example need to be viewed, or industrial radiography where engine components, turbine blades etc., need to be viewed, alternative technologies for the manufacture of larger area detectors are used. One such technology uses larger area amorphous silicon arrays (Figure 4), where charge is detected in individual pixels using charge sensitive amplifiers, and where the charge in each pixel is generated directly by the X-rays themselves producing electron-hole pairs in each pixel diode. Such detectors may be as large as-300 mm square, and each pixel may be only 100 microns square, providing good resolution over large areas. However, the manufacture of such a large area semiconductor devices is not at all straight forward, and maintaining yield and uniformity over such large areas, is difficult and costly. Such larger area X-ray detectors are now sold commercially, as replacements for traditional X-ray film modules. They offer the advantages listed above in terms of convenience, low dose, and running cost, however of the initial investment cost is very substantial indeed, typically -E100000.

There are variations on this theme using extrinsic methods (as opposed to the above which may be considered intrinsic), where an additional coating is used, and the X-rays produce changes in the conductivity of the coating, or indeed produce

scintillations of light. The precise methodology of these various alternatives is not relevant to the present discussion, suffice it to say that they all have in common the need to produce large areas of millions of pixels with a very good uniformity of response, and this is a costly process.

It is a purpose of the present invention to make available some of the advantages of X-ray digital imagery, but at a fraction of the current typical costs as described above. Of the three advantages listed above, the convenience in terms of imaging acquisition, immediacy, image processing, archiva, and hard copy has the most general appeal, and applicability. The issue of X-ray dose, although extremely important when studying beam sensitive materials, such as live biological specimens, and particularly human beings where exposure to X-rays is deleterious to human health, is much less of an issue in a general industrial environment where the specimen might be a piece of inorganic material or machinery. The advantage these detectors provide in terms of low dose derives from two aspects of the design.

Firstly, their sensitivity provided by the quantum efficiency of the individual pixels in terms of converting X-ray photons to electrical signals combined with the gain provided by an image intensifier, and secondly the fact that all the image pixels are accumulated in parallel. As discussed previously it is this aspect that adds enormously to the cost, i. e. providing large areas of the very many uniform pixels, and providing image intensification over equally large areas.

A low-cost alternative avoiding this cost penalty might therefore be to provide a single row of pixels spanning the width of the specimen, which can be mechanically scanned across the field of view underneath the specimen, thus generating a 2dimensional image line by line, much as the raster scan of a TV set for example.

A similar principle is currently employed routinely in many low-cost document scanners in offices all around the world everyday. The basic principle of such a device is described in figure 5, which depicts a scanner both isometrically (above), and in plan view (below). The document or image to be scanned (5.1) is placed face down upon the transparent glass face-plate (5.2) of the instrument. Beneath the glass face plate of the instrument mounted on a motor driven carriage (5.7) is a strip light (5.3) which illuminates a strip of the document to be scanned, as the carriage is driven from left to right (depicted as'Y SCAN'). Mounted on the same carriage is an optical detection system comprising typically an aperture slot, a mirror, and an optical detecting array comprising the single line of multiple light sensitive pixels which convert light into electrical energy. Thus light from a line on the object (5.1), illuminated by the strip light (5.3), is collected through an aperture (5.4) and focussed by a suitably curved mirror (5.5) onto the detector (5.6) producing an electrical signal corresponding to one line's worth of information on the document to be scanned. The carriage carrying 5. 3, 5.4, 5.5, and 5.6 is transported laterally by 5.7 a motor driven mechanical traverse, and lines of information are collected sequentially to generate a 2-dimensional image. There are variations on this theme used by different manufacturers and in Xerox machines. For example an alternate focussing means using a flat mirror in combination with a lens (as depicted in figures 7 and 8), is sometimes employed. Film transparencies, or negatives are more typically scanned using a strip light illumination above the object such that the light collected passes through the transparency. The above examples are intended to be illustrative rather than restrictive and the invention to be described subsequently, which is an adaptation to an optical scanner may also be considered as applying to alternative configurations of optical scanner.

Such devices scan printed text, photographs, and transparency images, both monochrome and colour, and are used in homes and offices throughout the world on a daily basis. Consequently the sheer volume of demand has driven down the cost of these instruments to the point where, a good quality document scanner with

1200 dpl resolution over an A4 ISO sheet of paper (i. e. 210x 297mm) may be purchased for as little as E100.

This document will now describe how such low-cost high-quality imaging devices maybe modified and adapted to provide large area X-ray detecting devices at a fraction of the cost of the currently available technology.

Figure 6 depicts the simples such modification. (6.1) is a micro-focus X-ray source as used in Figure 1 typically but not exclusively providing X-rays (6.2) at energies in the range 10-200 keV, and with an intensity corresponding to tube powers 1-200w typically. The object (6.3) to be imaged/analysed is mounted from a manipulating device such that different areas of the sample may be placed beneath the flux to be analysed. Also by positioning the sample closer to or further away from the source relative to the detector the magnification may be advantageously changed from 1 : 1 when at the detector (direct shadow), to in principle infinity when positioned at the source. (6) represents a document scanner as described above comprising aperture slot (6.5), focussing means (6.6), detecting means (6.7), and Y scan traverse (6.8), however the illumination system has been removed completely, and replaced with a layer (which may be a separate sheet or coating of the glass face plate) of phosphor material. In this way X-rays (6.2) from the source (6.1) which produce a shadow image of the object (6.3) at any plane below the object fall upon the plane containing the phosphor screen and produce scintillations of visible photons in the screen at any intensity proportional to the X-ray flux. As the X-ray flux is attenuated by the specimen's mass attenuation and density, this represents a 2-D image of the sample density. Visible photons generated in the phosphor are collected through the aperture (6.5), and focussed by mirror means (6.5) (which may also incorporate a lens) onto the optical linear array, which as before is mounted together with the aperture, and mirror on a mechanical scan traverse system (6.8) such that successive lines of X-ray image data may be collected to form a 2-dimensional Xray image of the specimen.

A further advantage of this scheme versus the'in line'systems described earlier derives from the fact that there is not a direct line of sight for X-rays into the optical detecting element. X-rays which impinge upon the optical detecting element produce uncorrelated ionisation ! electron-hole pairs which add to the noise and degrade the image. Therefore fibre optic elements are often employed in line of sight systems simply to reduce this effect. In the current scheme there is no direct line of sight and the detecting element and associated electronics can be easily shielded. The aperture plate for example could be advantageously extended to perform this function.

Additionally, as in the previously discussed schemes, optical image intensification devices could be advantageously interposed in the optical path between the phosphor screen and the detecting means to provide for greater sensitivity/lower dose to the specimen.

Figure 7 depicts an alternative modification to a commercially available document image scanner, which will also provide for X-ray image acquisition. As before, (7.1) is a micro-focus X-ray source providing X-rays (7.2) again typically but not exclusively at energies in the range 10-200 keV, and with an intensity corresponding to tube powers 1-200w typically. The object (7.3) to be imaged/analysed is mounted from a manipulating device such that different areas of the sample may be placed beneath the flux and also by positioning the sample closer to or further away from the source relative to the detector providing variable magnification. (7) represents a document scanner as described above comprising aperture slot (7. 5), detecting means (7.7), and Y scan traverse (7.8), but this time however the illumination system, the glass face plate and the mirror have been removed completely, and replaced with a phosphor strip running the width of the image field of view. In this way X-rays (7.2) from the source (7.1) which produce a shadow image of the object (7.3) at any plane below the object fall through the aperture

plate, which is now used to define the line width, and impinge on the phosphor strip screen and produce scintillations of visible photons in the strip, which focussed upon the optical linear array (7.7) producing one line of X-ray image data. As before the strip, lens and detector are mounted together on a mechanical scan traverse system (7.8) such that successive lines of X-ray image data may be collected to form a 2dimensional X-ray image of the specimen.

At the expense of a slight increase in complication versus the preceding scheme an improvement in sensitivity of this scheme could be engineered. The phosphor strip screen may be advantageously angled such that the X-ray photons impinge at glancing incidence, and therefore the visible photons thus generated which have a much shorter mean free path (escape length) than the X-rays will stand a greater probability of escaping and being detected from scintillations generated by X-rays that have travelled further into the phosphor. Also detectibly could be enhanced by coating the phosphor on a surface which is optically reflective, such that that part of the light emitted in an opposite direction to the detector will be reflected toward the detector, thereby increasing the sensitivity.

Additionally, as in the previously discussed schemes, optical image intensification devices could be advantageously interposed in the optical path between the phosphor screen and the detecting means to provide for greater sensitivity/lower dose to the specimen. In the present case this would be less costly as the intensifying device need only be a long thin strip, and the cost of intensifying devices is normally a steep function of area.

As discussed earlier it is anticipated that although this technology will provide the advantages of digital image acquisition, in terms of convenience, manipulation, hard copy etc., speed of data acquisition versus a large area 2-dimensional detector will be slow as this invention provides for what is essentially a serial acquisition of data line by line, whereas the large 2-D array presumably collects all image data in parallel. In many cases speed of data capture is not an issue, and it is simply necessary to wait longer. However in the case of radiation sensitive materials it may simply be unacceptable to use a technology in which the sample is being irradiated all the time even though the image data is only being acquired from a single line at once. There is, therefore a third embodiment of the current concept, which although still does not acquire all image data at once, does minimize the specimen's exposure to the X-ray flux.

This is depicted in Figure 8, which is in most respects identical to Figure 7 with the exception of the addition of a'Lever Arm' (8. 9), and a second'Aperture Slot' (8.10).

The lever arm is suspended about a'Pivot' (8.11) point located above or below the plane of the diagram, but otherwise at the source position, and able to swing in the Y direction. The opposite end of the lever arm is suitably connected at (8.12) to the Y-scan traverse mechanism at a position approximate to the point where the X-rays impinge upon the phosphor screen. Mounted from the lever arm at a position close to the source is a second aperture slot, which therefore defines a fan shaped X-ray beam, extending over a narrow distance in the'Y'direction but sufficient to illuminate aperture slot 8.4, and extending over the full width of the X-line direction, and which is therefore scanned concomitantly with X-ray detector mechanism, thereby providing an image identical to that produced by the scheme defined in Figure 7, but with the additional benefit of minimising unnecessary specimen exposure to the X-ray flux.

In a forth embodiment of the current scheme as depicted in Figure 9 provision is made to disable the electrical signal that causes the detector carriage to be scanned in the Y direction, and use it instead to scan or otherwise control the scan of the specimen on a motor driven positioning device in the Y direction. Such an implementation could be used in an environment where specimens are travelling in the Y direction continuously such as for example on a conveyer belt production/ inspection system, or other continuous manufacturing process.

Larger area commercial document scanners (ISO A3 and bigger) are also readily available, and could equally easily be modified for X-ray detection using the principles outlined above, and thereby providing unparalleled detector size at unparalleled cost.

Claims

Claims 1. A radiographic imaging instrument in which specimens to be examined are positioned between an x-ray source and an imaging radiation detector, said radiation detector comprising a modified commercial document scanner connected to a computer and visualisation means and consisting of, a one dimensional array of multiple optical light sensitive detectors which collects one line of optical information (in say the X direction), mounted upon a motor dnven carriage which can be stepped or scanned in an orthogonal direction (Y) such that multiple single lines of optical information can be collected sequentially to produce a two dimensional Image and, wherein an x-ray to optical wavelength light conversion means is Interposed in the path of the x-ray flux at a point where the light thus generated by the x-ray photons can be advantageously collected by the subsequent optical ! electronic imaging means, as if deriving from a scanned document and to therefore produce a two dimensional x-ray image of the specimen on a computer display.
2. A radiographic imaging instrument as in claim 1, wherein the x-ray to optical wavelength conversion means simply comprises a sheet of x-ray phosphor material placed upon, or suitably coated upon the document scanner face plate where a document to be scanned would normally be placed, the illumination means having been disabled (or removed), and the light thus generated by x-ray scintillations alone collected by the means provided and used to generate a 2 dimensional X-ray image of the specimen.
3. A radiographic imaging instrument as in claim 1, wherein the scanners face plate has been removed, and the x-ray to optical wavelength means comprises a strip of x-ray phosphor material which may or may not be beneath a defining slit and mounted upon the scanner carriage along with the electronic detecting means such that 1-dimensional lines of x-ray induced light are collected sequentially to produce a 2-dimensional x-ray image of the specimen
4. A radiographic imaging instrument as in claim 3, wherein the x-ray to optical wavelength means is advantageously tilted to optimise the number of x-ray induced photons that can be detected.
5. A radiographic imaging instrument as in claims 3 and 4, wherein the x-ray to optical wavelength means is deposited upon a suitably curved highly reflective surface such that x-ray induced photons travelling away from the detection means are advantageously reflected and focussed toward the detection means
6 A radiographic imaging instrument as in claims 3,4, and 5, wherein a slit plate also comprises a shield preventing x-rays from directly illuminating the electronic detection means, and thereby causing uncorrelated signal manifesting itself as unwanted image noise.
7. A radiographic imaging instrument as in claims 3,4, 5, and 6, wherein an image intensification device is also employed between the x-ray to optical wavelength conversion means and the electronic detection means to improve the overall sensitivity of the apparatus.
8. A radiographic imaging instrument as in claims 3,4, 5, 6, and 7, wherein an additional defining slit is interposed between the source and the specimen in order to define a fan shaped X-ray beam, the slit being scanned synchronously with the scanner such X-ray images of the specimen are generated as above, but where specimen exposure to the x-ray flux is minimise to that part of the specimen being imaged.