WO1995012133A1

WO1995012133A1 - System and sensor for scanning and digitising x-ray images

Info

Publication number: WO1995012133A1
Application number: PCT/FR1994/001259
Authority: WO
Inventors: Catalin Stoichita
Original assignee: Catalin Stoichita
Priority date: 1993-10-29
Filing date: 1994-10-28
Publication date: 1995-05-04
Also published as: FR2711799A1; FR2711799B1

Abstract

A system for scanning and digitising X-ray images, comprising an X-ray image sensor (IS) based on a multiline charge-coupled device (CCD) and an assembly of optical fibres and a scintillation screen (PH) arranged opposite the CCD to enable a desired X-ray image of the object being X-rayed and displayed on the screen to be converted thereby into a visible light image and integrated by the CCD; electronic circuits for routing and processing the signal from the sensor (IS); and an electromechanical subsystem (MEC) enabling the sensor (IS) to scan the object (O) exposed to X-ray radiation. The X-ray image sensor (IS) includes a number of structures each consisting of a preferably rectangular elementary CCD and a scintillation screen segment, said structures (PH, FB, CCD) being arranged along the length of the sensor preferably in the same plane and, for at least a part of each structure, in two parallel zigzagging rows.

Description

SYSTEM AND SENSOR FOR SCANNING SCAN

X-RAY IMAGES

The invention relates to a system and an X-ray image sensor for its implementation, system and sensor which can be easily adapted on any film radiography apparatus known today, for directly obtaining in digital form the image. objects to be x-rayed by scanning, especially when it comes to large image or tomography formats. It is intended to replace the films usually used in radiology by making the image available in digital form on a computer. It will be called by abbreviation: SCRX.

STATE OF THE ART

Radiography of objects opaque to visible light has been possible for a very long time using silver films. There are a wide variety of devices that correspond to the diversity of applications: medical diagnosis, baggage control, defect control, parameter control, etc.

Most cameras allow you to obtain images on film films. They have significant drawbacks: insufficient sensitivity (see dose of X-ray used), a time to obtain the image long of several minutes due to the chemical development process, the need to use a dark room for handling films, the use of often toxic chemicals, etc.

There is another family of radiography devices which uses a television camera (digital or analog) directed towards a fluorescent screen sensitive to X-rays, screen on which the X-ray image to be acquired is converted into a visible image (see request FR 76 35547). To have the necessary sensitivity, an image intensifier is interposed between the camera and the screen. The main drawbacks of this range of devices are a reduced resolution or image field and a reliability and a lifetime limited by the image intensifier (see considerations in US 5,095,202 patent). There are also X-ray image sensors based on visible light image sensors of the charge transfer image sensor device type (CCD = Charge Coupled Device imagers). These sensors have many advantages but the unfavorable aspect common to all these sensors is the limitation of the image field to sizes of the order of a few cm ² (see for example patent application EP-A-0 129 451 and WO 92/13492) due to the impossibility of manufacturing very large CCDs at reasonable costs, in order to cope with the very small size of the CCD-s available, reductive optical fibers are used which, in addition to the difficulty of their manufacture, generate geometric distortions as well as significant non-uniformities of sensitivity.

Adding an optical lens in front can significantly widen the image field, but this automatically leads to a proportional drop in resolution (see for example US 5,142 ^ 57 patent).

Other attempts are made in the direction of enlarging the effective surface of image sensor devices by juxtaposing in various ways several units of CCD- s to make a single unit with a larger equivalent surface (see for example patent application FR 9210292, WO 91/03745, EP-A-0 208 255 and WO 93/04384). This kind of structure quickly becomes inadequate when it comes to sizes of the order of hundreds or even thousands of cm ² , because of the complexity and the price.

Fixed CCD-s based sensors provide the most advantageous solutions for X-ray imaging when it comes to small images while for large formats there are few direct scanning solutions known today. hui. Among these is that described in US Patent 4,878,234. According to a main characteristic, this patent proposes two arrangements of image sensor a first arrangement using a single CCD and a second arrangement using several CCD-s.

In the case of a single CCD, the X-ray image sensor, according to the cited patent, comprises "... an image transmission or coupling element (10), such as an optical fiber, which reduces the format of the opening of the secondary diaphragm (8) in the format of the image zone of the CCD sensor (11) "(see English text column 3 lines 7 to 11 of the cited document). Following the rest of the description, it follows that this block of optical fiber (10) must have a very large reduction factor, of the order of 15 times. However, to date, the best manufacturers in the world who are able to attempt the manufacture of such optical fibers have many technical problems with the manufacture of blocks of fibers with a reduction factor of 2 to 3. Consequently, in pending significant progress in this area, the proposed solution remains impracticable.

For the second arrangement, the same patent proposes a succession of CCD-s filled with blocks of optical fibers (14) cut in a very inclined manner (sin (a) ^* = 0.15) to obtain the image reduction effect. The authors realize the importance of the resulting loss of light and suggest remedial solutions such as the use of image intensifiers, a kind of phosphor support grid, etc. (see column 8 line 3 to 29).

Another characteristic of the same document cited above is that it proposes to use CCD-s style television cameras, that is to say CCD-s i frame transfer (Frame

Transfer CCD) or CCD-s with Line Transfer CCD. The image rendered by the CCDs is digitized using an A / D converter and then processed by a "processing unit pre-processing unit 17 "before being written to a large image memory. Another digital processing is applied to the information contained in this memory in order to obtain the image of a desired section of the object to be radiographed (see column 3, line 30, column 4, line 30) The authors actually suggest several variations on this topic.

The CCD-s considered by the authors, by their very principle of operation, can only use half of the surface of the component because half is occupied by the light-sensitive region and the other by the shaded region called "zone from memory "(see claim 1, line 66, and claim 2 or the whole document). In addition, this kind of CCD-s is limited in speed by the need to transfer the image from the integration region to the memory region. Other disadvantages are the complexity (cost) of post-digitization processing as well as the errors involved.

BRIEF GENERAL DESCRIPTION

The object of the invention is: a) to provide a system and a sensor for direct digitization, as announced in the introduction, to replace the films in conventional radiology installations, in particular for tomography; b) to remove the image size constraints; c) to reduce the complexity inherent in other known solutions for direct digital X-ray imaging; d) to allow manufacturing with components currently available at present To eliminate the drawbacks of known solutions, the scanning system for scanning X-ray images, proposed by the present invention, uses an X-ray image sensor based on one or more several full-frame charge transfer image sensors (Full Frame CCD) and specific control and digitization circuits. A mechanical subsystem allows the image sensor to scan by translational the transparency image of an object exposed to X-rays. During scanning, its movement is translated into electrical signals by various means, signals which, via a block generator of the servo signals, drive the scrolling of the image in photo-electrons on the CCD (s) of the sensor.

The servo cited above is done in such a way that, between all the images of the different regions of the object, projected simultaneously on its sensitive surface, only the image of interest is followed with precision and left sharp by l integration while the others are blurred. It is also explained, the relationship between the servo equations, the diagram kinematics of any system (to choose from many variants) and the positions of the object or its parts to be radiographed.

The system also uses a computer which is responsible for acquiring the image scanned by the sensor and for configuring the parameters of the sensor or the generator of the control signals.

The invention will be better understood, and other purposes, characteristics, details and advantages thereof will appear more clearly during the detailed description which refers to the attached schematic drawings given solely by way of example.

BRIEF PRESENTATION OF THE FIGURES

FIG. 1 represents the surface of a full frame type charge transfer image sensor device on which a moving image is projected.

Figure 2 shows a block diagram of the SCRX system. FIG. 3 represents a block electrical diagram of the generator of the control signals

FIG. 4 shows an exploded view of example 1 of an image sensor arrangement which comprises only a CCD.

FIG. 5 shows an exploded view of example 2 of an image sensor arrangement which comprises several CCD-s mounted on a common support.

FIG. 6 shows an exploded view of example 3 of an image sensor arrangement which comprises several CCD-s mounted in individual housings.

FIG. 7 shows a detailed view in section of the examples of arrangements of the image sensor. FIG. 8 represents a block electrical diagram of the image sensor.

FIG. 9 represents a kinematic diagram where the generator is fixed relative to the object, for the use of the system proposed by the present invention

FIG. 10 represents a kinematic diagram where the generator is movable relative to the object to be radiographed, for the use of the system proposed by the present invention

DETAILED DESCRIPTION

CCD used with multiple moving images

The sensor for implementing the system in accordance with the present invention comprises one or more charge transfer devices of the full frame type (Full Frame CCD). This choice has the advantage, compared to the other known solutions, of using practically the entire surface of the device for integrating photoelectrons, something which becomes essential when it comes to large sensors. Another advantage of this choice is the fact that, in tomography applications, the image of the desired section is formed directly at the level of the image sensor device and not by processing subsequent to scanning, which makes it possible to use a simpler electronic scheme (cheaper, more reliable) and which allows better fidelity.

In Figure 1, there is shown in a simplified manner, such an image sensor device. Neither the displacement electrodes nor the other electrodes are drawn. Generally, the operation of such a CCD comprises two stages: a so-called integration period, during which the visible light image projected on the CCD is converted into a photo-electron image, followed by a so-called reading period which takes place with the device in the shade. Reading consists of a sequence of vertical displacement (see arrow V on the CCD) of all the image pixels (in the form of electron packets). The movement is done line by line. Between two consecutive vertical displacements, the first line of the CCD, which passes through the output register, is displaced pixel by pixel (see arrow H on the CCD). These pixels (packets of electrons) are converted into voltage proportional to their charge, a voltage which finally constitutes the output video signal from the CCD.

In the case of the use of CCD-s in accordance with the present invention, a visible light image in movement with respect to the CCD (in reality, there are several but for a better understanding one of them is considered momentarily ), is projected onto this CCD, the movement in question taking place at a certain speed, in the vertical direction (V), the image is represented in FIG. 1, in two successive positions (IM) and (IM '). The photo-electrons integrated by the sensor are grouped in pixels which are organized in rows and columns but, to simplify the drawing, only a group of three has been represented (N, N + l, N + 2 or N ', N' + l, N '+ 2) adjacent lines, which covers the extent of the image (IM, IM'). There is no separate integration period; integration and reading are simultaneous.

It is essential to control the CCD as well as the speed of vertical displacement of the integrated image (the photoelectrons) so that it is equal to the speed of the image in visible light compared to the surface of the CCD . In other words, while the visible light image moves from its current position (IM) to the next position (IM '), the photo-electron image follows the lines (N, N + 1 and N + 2) move to positions (N ', N' + 1 and N '+ 2). The integration of light is therefore done while the image is passing through the CCD. It will be observed that at no time does the image projected on the CCD device correspond to the photo-electron image integrated by it. If the image projected at a certain time is described by a function f of type: f = f (xy) or

X, y are coordinates measured on the surface of the CCD (X in the direction of the lines and y in the direction of the columns of the CCD) then, the integrated image will be expressed by a function Sde type:

S = S (x, y) = k * x * f (x.y) where k is a constant depending on the sensitivity of the CCD

A similar but different technique is applied in the acquisition of aerial views (single image placed at infinity) as part of the LOREORS program (Long-Range

Electro-Optical Reconnaissance System - U.S. Air Force). It is called TDI (Time-Delay and Integration) - see Long Range Electro-Optical Reconnaissance System and Flight Test

Results by F. Palazzo in Proc SPIE 561, pag. 13-17 Λug. 1985.

According to a characteristic of the present invention and unlike the kind of applications mentioned above, there are several simultaneous images of the object to be examined, projected on the surface of the CCD, in movement relative to it, at different speeds. The CCD is controlled so that it only follows one of them which is called the image of interest. In the CCD output video signal, all the images are recovered but only the image of interest is clear, the others are blurred. This CCD control rule, to select an image from among several projected simultaneously on it, leads to the possibility of performing tomography.

Another difference compared to the applications cited above is that the CCD-s used by the present invention do not include means of adaptation to the level of lighting specific to the TDI-type CCD. This is an advantage because the control electronics become simpler

System block diagram (SCRX)

In Figure 2 is shown an example of a block diagram of the SCRX. The system includes a computer (C) in a configuration suitable for graphic applications: high resolution screen (D) and graphic printer, mouse, joystick, adequate discs and speed, etc. not shown in the figure. The computer (C) connects to the rest of the SCRX by a specific interface (I), using the cable (CI) and the cable (CV). The cable (CI) connects with the image sensor (IS) and the cable (CV) connects with the servo signal generator (SVPG).

The generator (SVPG) is furthermore connected with means (L1, L2) for detecting the limits of the sweep stroke as well as with means (MT) for translating the movement. of the sensor (IS) providing pulses at frequency proportional to the speed of the sensor. To facilitate the explanation, we will consider the means (MT) as an incremental position translator and the means (L1, L2), as switches. The incremental position translator (MT) and the limit switch switches (L1, L2) are part of the mechanical subsystem (MEC). The generator (SVPG) is also connected to the image sensor (IS) by a cable (CS).

The servo signal generator (SVPG), the image sensor (IS) and the interface (I) are elements with structures and functions particular to the present invention while the others are known standard elements ( examples: cables, computer, etc.). Various possible kinematic schemes for the subsystem (MEC) are not claimed by the present invention; there are two diagrams drawn only as an example. It will be observed that, insofar as the electromechanical subsystem (MEC) has for the needs of internal control means for determining with precision the speed of the sensor on its trajectory, it can be redundant to use for the needs of the present invention, an incremental position translator. This is the case, for example when the mechanical elements are set in motion by stepping motors controlled by microprocessors. We can already have the necessary signal, as part of the control circuits of the engine (s) without the help of the dedicated incremental translator

Computer interface (I)

According to a characteristic of this invention, the interface (I) has the electronic circuits necessary to fulfill, among other functions depending on the application of the SCRX, the following functions: a) the sending of the electric current supplying the image sensor (IS) by cable

(THIS); b) the reception of the image sent by the sensor (IS), via this same cable (CI), in the form of electrical signals (preferably serial digital transmission by modulation / demodulation in Manchester code); c) sending the configuration signals to the image sensor (IS) via the same cable

(THIS); d) sending the generator supply electric current (SVPG) via the cable (CV); e) sending programming signals to the generator (SVPG) via the same cable (CV); f) rapid transfer of the data stream representing the image received from the image sensor (IS) to the computer, preferably in direct memory access mode; One can imagine variants of this invention according to which there is no cable (CV) and consequently the interface (I) does not take care of the power supply or the programming of the generator (SVPG). This can have a fixed program and be fed differently (the meaning of "programming" of the block (SVGP) will be understood below, in the paragraphs which concern it). Likewise, one can forgo, for simplicity, the remote configuration of the image sensor (IS) by replacing the function “c)” mentioned above, by local configuration switches or simply use a image in a fixed configuration, suitable for a certain application. These variants are not the most efficient from the technical point of view but they can give justified savings.

The servo signal generator (SVPG)

Another important feature of this invention is that the servo signal generator (SVPG) has the following properties: a) it has a memory that can be downloaded by the computer (C) with an array of profile-specific data the object to be x-rayed (or a section thereof); the table contains what can be called a form factor which is a function of the position of the image sensor (IS) on its trajectory; b) during the exploration of the object (O) to be radiographed, it supplies the vertical displacement pulses of the image integrated by the image sensor (IS) so that it acquires a clear image of this object (or the desired section); the instantaneous frequency of these pulses is the result of the multiplication of the instantaneous frequency provided by the incremental position translator (MT) multiplied by a fixed scale factor during the shooting and divided by the form factor recorded in its memory ; c) it receives the position information supplied by limit switch switches (L1, L2) and takes it into account to determine the period during which it is necessary to send said vertical displacement pulses to the sensor (IS); d) it transmits on request to the computer (C,) the position of the image sensor (IS) on its trajectory; e) it is supplied by the same cable (CV) which is used for communication with the computer (C);

To better understand these functions, refer to Figure 3 which represents a block diagram of the generator (SVPG). The block diagram includes a high frequency oscillator (OSC) (for example 32 MHz), stabilized with quartz, which serves as precise time base. Input signal (IPT) (pulses from the incremental position translator

(MT), seen in Figure 2) is synchronized with the internal clock (OSC), using a synchronization block (SYNC) based on bistables to eliminate hazards. It is then issued under the name (SIPT) and passes to the block (SCALER).

The block (SCALER) has the task of supplying at its output the signal (SCIPT) consisting of pulses with a frequency proportional to the frequency of the input signal (SIPT) (or IPT). The proportionality factor is imposed by the block (SCTL) through a control word noted in the figure with (SF). It is recommended to carry it out using reversible and programmable counting registers or phase-locked loops

The diagram contains the block (DIV) which has the task of dividing the pulse frequency (SCIPT) by a number (RMD) which is supplied by the block (SCTL). For this, it is preferable to use a reversible and programmable counter type. Thus, this divider block in turn delivers the output signal of the diagram (VSH) which is used to synchronize the image sensor (IS) (cited in FIG. 2) for the vertical displacement of the integrated image. We will see below the importance of this enslavement

The block (SCTL) plays the role of local controller of the entire scheme. The connection denoted (SI), represents the remote connection with the interface (I) of the computer (C) by the cable (CV)

(as already presented in Figure 2). The block (SCTL) can in particular receive, by serial transmission, the proportionality constant (SF), mentioned above and the table which is passed to the memory (MEM). Information traffic with memory (MEM) is noted in Figure 3 with (SMD). The traffic is bi-directional and it is preferable to use serial memories EEPROM for this purpose, for reasons of size, reliability and price, because the capacity and the data rate allow it The data read are presented as a factor division (RMD) to the block (DIV).

The signal (SIPT) also enters the block (SCTL) by telling it to trigger the memory read cycles (MEM). To manage the reading or writing of the memory (MEM), the block (SCTL) is provided with an address counter which is incremented by the signal (SIPT).

Another feature of this same block is to communicate, on request, to the computer (C), via the serial channel (SI), the content of the address counter which signifies the position of the image sensor on its trajectory. In an alternative embodiment, this address counter does not exist in the block (SCTL), because it exists in the memory itself (quality specific to many available serial EEPROM memories). Of course, in the case where the block (SCTL) does not contain the address counter, the generator (SVPG) cannot give the computer information on the current position of the image sensor on its trajectory, something which can be bypassed by other standard means to be added to the SCRX or simply ignored if the application allows it. The block (SCTL) receives dts signals (IL1, IL2) arriving respectively from the limit of course detection means (Ll, L2) (seen in Figure 2). The same block must transmit the state of these lines (IL1, IL2) remotely to the computer (C), via the serial link (SI). Other details of the operation will become understandable when presenting the examples of use of the SCRX.

As shown in Figure 2, the SCRX contains the image sensor (IS). For this invention, we then present three examples of arrangements for this sensor, then a single electrical block diagram of the control and digitization circuit valid for all three.

First example of sensor arrangement

In a first variant embodiment, in accordance with FIG. 4, the image sensor (IS) is composed of a CCD image sensor device on the sensitive surface of which a block (FB) in the form of a right parallelepiped is stuck, made of glass optical fibers. On the other side of the block (FB), there is a scintillator screen (PH) (the screen can be a phosphor directly deposited on the fiber or separately, on a plastic film applied to the fiber). The CCD device (CCD) is mounted on a support means (SUPP) with which the device is connected by connections (CX). All of the elements mentioned above are mounted in a box (B) covered by a front cover (CF). In the same case, there is still a plate (Pb) of X-ray absorbent material, parallel to the plane of the CCD, covering the entire opening of the box except a rectangular opening corresponding to the size and position of the screen. (PH). On one wall of the housing are mounted electrical connectors (BNC1, BNC2) allowing electrical connections with the cables (CI, CV) external. At the rear of the housing (B) is a printed circuit board (PCB) on which is mounted the specific electronic diagram of the sensor. The electronic components are not visible in the drawing, as are the connections between the printed circuit (PCB) and the support means (SUPP). The (PCB) with its components are covered by a rear cover (CA).

To operate this kind of sensor, the X-rays representing a portion of the image to be scanned, pass through the front cover (CF) (because it is made of very little absorbent material such as plastic) and tap on the screen scintillator (PH). The plate (Pb), being highly absorbent of X-rays (preferably lead 1-3 mm thick), serves to limit the field of X-rays to the minimum necessary to improve the life of the electronic components of the sensor. The screen (PH) converts the X-ray image into a visible light image which is transported by the fiber optic (FB) block to the CCD device.

The block (FB) plays the role of absorbing X-rays emerging from the screen (PH) while transmitting the image of this screen without enlargement or reduction. Absorption is necessary to protect the CCD device. The glass used for the manufacture of the fibers of this block is a common glass because in the absence of strong space constraints, it is possible to put all the necessary thickness. A thickness of 10 to 15 mm will be chosen for the fiber (FB). Even if there are still some X-rays reaching the CCD, this has no significant consequence since the influence on the reliability of CCD-s begins at large doses. The visible light image which arrives on the sensitive surface of the CCD (CCD) is converted into an electron image which is digitized using the electronic diagram of the sensor, described below.

The variant presented in Figure 4 has the advantage of using a single CCD. but this advantage can be limited in certain cases because the realization of a very long true CCD, (for example 15 to 30 cm), as many applications require it, is completely impossible. It is then necessary to manufacture shorter CCDs with inactive regions on the edges as small as possible and extreme columns of reduced width to allow precise juxtaposition at the pixel level without causing a discontinuity of periodicity in the total line. More precisely, the width of the extreme columns added to the width of the neighboring dead region must be equal to a normal column width. The role of the support means (SUPP) is to make the structure of the sensor modular. Thanks to him, we can entrust to the semiconductor industry the manufacture of a very long CCD image sensor device which is actually a series of CCD-s of medium lengths, mounted in juxtaposition on this support.

Having fiber blocks of sufficient length is economically difficult. It is also possible to glue pieces very precisely, but this again leads to a significant increase in costs.

The housing (B) must be very rigid to minimize the accidental mechanical forces transmitted to the sandwich (CCD) - (FB). At the same time, this box must facilitate the dissipation of the heat given off by the electronic components.

Second example of sensor arrangement

Another variant of the sensor, which attempts to overcome the drawbacks mentioned above, is presented in Figure 5. The main difference compared to the previous version lies in the use of several CCD devices (CCD) (four in this example ), identical and mounted in a zigzag (two parallel rows and offset so that each CCD has no opposite). The second difference is that each CCD is covered with an individual optical fiber (FB). The fibers (FB) are covered by scintillating screens. The fibers have the particularity of being inclined (with respect to the plane of the CCD-s and in a direction perpendicular to the direction of their lines) so as to compensate for the distance between the two rows of CCDs, by bringing the two rows as close as possible corresponding scintillator screens (PH). The required angle of inclination of the fibers can vary between 5 and 30 degrees depending on the dimensions of the CCD auxiliary elements (such as the width of the ranges to the cutting safety margin, etc.) and depending on the thickness chosen for the fiber (between 8 and 15 mm).

The distance between two consecutive CCDs in a row is chosen so that it is a little less than the length of the active region of a CCD. Thus, by considering the image explored by the sensor (IS) as being broken down into several bands corresponding to each CCD, it will be observed that the bands overlap by a few pixels. The offset of the CCD-s belonging to the two rows is taken into account by the software which manages the acquisition of the image and compensated in such a way that there are no discontinuities corresponding to the positions of the CCD-s in the image sensor. Of course, the number of CCD-s can vary to obtain the necessary length.

The active surface of such an X-ray image sensor is in the form of rectangles arranged in a zigzag and for this, the X-ray absorbing plate (Pb) must be cut in the same way as in FIG. 5. The other elements of the sensor presented in Figure 5 have the same role as their correspondents in Figure 4, so we do not repeat the description.

We will retain from this example, that we can use standard CCD-s, that is to say not manufactured "to measure" or not only for this sensor. In the event of a breakdown (electrical failure on the CCD, take-off of the fiber, etc.), it is entirely possible to change one element while keeping the rest. This is very important for a correct production yield and easy after-sales service. On the other hand, for the previous example, it is practically excluded to intervene to repair or recover the most expensive elements like CCD or fiber.

Third example of sensor arrangement

Figure 6 shows a third embodiment of the image sensor. Compared to the example described in Figure 5, the CCD devices are mounted on standard supports (DIP), which greatly reduces the dependence of CCD-s users on manufacturers of CCD semiconductor devices with all the advantages of standardization, handling, repair, etc. In this figure, we detail in the circle (D), the relative location of the CCD-s (CCD). The support means (SUPP) does not appear because it is no longer necessary. CCD-s (CCD); already mounted in their standard boxes (DIP) connect with copper (PCB) through openings made in the bottom of the box (B). The inclination of the fibers (FB) is greater than in the previous example because of the increase in the distance between the two rows of CCD-s (which in reality is determined by the existence of the DIP capsule). Despite this, it is possible to stay below 32 degrees, which does not create significant optical problems related to the use of fibers in oblique incidence. The arrangement considered in FIG. 6 is considered to be the most efficient of the three arrangements presented.

Figure 7 shows mounting details for the sensor described in Figure 6. These details are also applicable to the three previous examples of sensor; they have not been drawn on the corresponding figures to make them clearer.

As the blocks of optical fibers (glass) (FB) are quite heavy, there is a risk of causing damage in the event of accidental impact, strong vibration, or deformation of the housing (B). To avoid this, we present in Figure 7 the fixing of the fibers (FB) using a hard plate (RE) made of hard resin cast so that the fiber blocks (FB). are embedded there at their center of gravity. Contact with fibers should be reduced to between

1 and 3 mm thick all around each block It is preferable to choose an epoxy resin loaded to have good dimensional stability and polymerizing at room temperature to reduce thermal stress on the CCD.

Before pouring the hard resin plate (RE), a layer of soft resin (SIL) is poured to isolate the CCD-s housings from any mechanical contact with the hard resin and to obtain the seal necessary for the casting of the hard resin.

In the event of strong acceleration (for example a shock), the movement is transmitted directly to the heaviest parts, the fibers. The transmission of mechanical forces through the connection pins of the CCD boxes is limited due to the elasticity and plasticity of these and also because of their connection to the printed circuit (PCB). It is preferable to use tulips or standard means of contact connection with low insertion force. The details shown in Figure 7 also apply to the arrangements shown in Figures 5 and 6. This structure has the advantage of allowing partial disassembly for repair.

Electronic diagrams of the sensor

We observe in Figure 7, below the printed circuit (PCB) the presence of electronic components (CE) which form the electronic control circuit and signal processing of the sensor. This is described in terms of the electrical block diagram and its functions in Figure 8. In a recommended variant, the sensor is connected to the outside by two cables: a cable called (CI) connected to the connector (BNC1) and a cable called (CS) connected to the connector (BNC2).

Via the image cable (CI), the supply current and the information transmitted in digital form by the modems located at both ends circulate in two directions. different: a modem on the interface (I) sends configuration commands and another present in the diagram of the sensor and noted (RX / TX) sends the scanned image.

The condenser (CC) symbolizes the elements of filtering (blocking) of the direct current allowing the coupling of the modem with the line. The coil (LF) and the capacitor (CCF) symbolize a low-pass filter which separates the supply voltage on one side and the modem carriers on the other side. This supply voltage is then applied to a regulation block (REG) which delivers at its output the various stabilized voltages for the other components of the diagram (for example + 5V in the case of logic circuits). The internal power connections in this diagram are not shown in the figure. The diagram also includes the block (A / D) which is an analog-digital converter. It is used to digitize the output video signal from the CCD (s). Most applications of this image sensor will be satisfied with 8-bit precision. The requested speed is quite important and can go up to several Mega-samples / second.

As seen in Figures 5 and 6, you can have multiple CCD-s in an image sensor; they are therefore represented in FIG. 8 by the blocks (CCD1, CCD2, ..., CCDn) corresponding to a sensor equipped with n CCD-s. The analog video signals output from these CCD-s are multiplexed by the multiplexing block (MUX) before being applied to the converter (A / D).

To accomplish their image capture task, the CCD-s are driven by represented drivers grouped in the DRV block. They adapt between, on the one hand, the voltage levels and the output impedance of the logic circuits and, on the other hand, the voltage levels and the input impedance of the CCD-s. The block (POL) symbolizes all the components dedicated to the polarization of the various grids (substrate, level of zero setting of the output, etc.) of CCD during their operation. The vertical displacement synchronization pulses (VSH) enter this diagram through the connector (BNC2) and are applied to the input of the control block (SCANCTL).

The block (SCANCTL) is an algorithmic machine operating on the basis of a high frequency clock (for example 32 MHz) delivered by the oscillator (OSC). Its functions characterize the overall functioning of the sensor. They are a) receive from the computer (C) the configuration parameters via the modem (RX / TX); b) support several image scanning modes which differ by the matching factor of rows and columns (English - "binning factor") and a test mode which allows the generation of a test pattern which replaces the image to digitize; c) receive from the generator (SVPG) the synchronization pulses (VSH) for the vertical displacement of the CCD (s) (scanning modes); d) receive from the converter (A / D) the data representing the pixels of the scanned image (scanning modes); e) send, via the modem (RX / TX), the data representing the pixels of the scanned image (scanning modes); send, via the modem (RX / TX), the data representing a test pattern (test mode); g) deliver the control signals to the converter (A / D) (scanning modes); h) issue the selection commands to the multiplexer (MUX) (scanning modes); i) deliver the displacement pulses, via the drivers (DRV), to the CCD (-s), in accordance with the established configuration and synchronous to the pulses received from the generator (SVPG) (scanning modes) ; j) send, if necessary, pulses to the block (POL) of bias circuits of the CCD (s) (digitization modes);

In a simpler variant of the image sensor, it is possible to dispense with the direction of communication between the computer (C) and the image sensor (IS,) by using a unidirectional modem. To have several configurations anyway, use mini-switches installed on this diagram and accessible by an opening in the rear cover (CA) or in the housing (B). We did not draw a separate figure because the modification is quite obvious. The details which can be given on the production of each functional block are very numerous, but a skilled person with a certain experience in digital and analog electronics can imagine a detailed diagram taking into account the particularities of the CCD devices to be used and the functional specifications. above. It is recommended to choose for the assembly of the components, the CMS technology (Components

Surface mounted), to choose the components with a minimum weight and size and to choose an integrated circuit of the area-programmable type for the synthesis of the (SCANCTL) (this can include the communication functions) or even to realize a "ASIC" ("Applications Specifies Integrated Circuit" = application-specific integrated circuit). All the considerations on the diagram presented in figure 8, remain valid to accompany and complete the three examples of sensors described previously in figures 4,5 and 6

First example of using the SCRX system The structures of radiography devices can be, as explained, kinematically grouped into two classes: those where the generator does not move relative to the object and those where the generator moves.

Figure 9 is a kinematic diagram showing how the SCRX operates when it includes an electromechanical subsystem (MEC) having a kinematic diagram of the first category (fixed generator-object). Such a system can be used for mammograms, chest x-rays, angiographies, etc. in the medical field, or for checking baggage, checking welds, etc. in other fields.

The generator (G) illuminates the object to be radiographed (O) through a diaphragm (DF). The X-rays emerging from the object and which represent its image, touch the sensor (IS). The image sensor (IS) is installed on the carriage (CHR) which runs on the rails (RY). It is driven by the motor (M) using a wire (F) which goes back and forth around the pulley (P) and which is wound on the drum (TB) located on the motor axis. The limits of the travel of the carriage (CHR) are detected by the limit switches of the travel (Ll, L2). The movement of the sensor (IS) is followed by the incremental movement translator

(MT) placed on the motor axis (M). There is also the lever (L) which can tilt around the axis (A). The diaphragm (DF) is rigidly mounted using a support rod (S) on the lever (L).

When the carriage (CHR) carrying the image sensor (IS) moves, it switches the lever (L) because of the connection that exists between the two and thus causes the diaphragm (DF) to move.

The opening and the movement of this diaphragm allow the limitation of the beam of rays

X to the sensor field, therefore limiting the dose of radiation applied to the object to the minimum necessary. The limits of the X-ray beam are shown in dotted lines.

We will recognize, in relation to Figure 2, the elements of the SCRX present in this diagram: (IS, MT, L1, L2). The other elements of the system and their interconnections are no longer shown to make the drawing simpler. The electric motor control (M) is not shown for the same reason.

In these applications where the generator does not move relative to the object to be radiographed and where it is necessary to obtain the image in planar projection, the table contained in the generator of the servo signals (SVPG) is filled by a constant and the postman

(SF) chosen as well as the frequency of the vertical displacement pulses (VSH) of the CCD (s) of the image sensor (IS) is that which gives a displacement speed (-V) of the integrated image relative to the sensor, equal in size and opposite sign to the speed (V) of translation of the sensor relative to the object to be radiographed. This rule of enslavement of the sensor (IS) to the movement imparted by the motor (M) (the relation of equality of speeds and their opposite signs) makes the integrated electronic image fixed relative to the plane of translation of the sensor. The image sensor (IS) is oriented so that it moves in translation in its plane and perpendicular to its lines (which are longer than its columns).

The diaphragm (DF) is made of absorbent material such as lead. The representation of the carriage (CHR) which rolls on rails (RY) by the motor (M) is only one example: it is necessary to understand any mechanical means allowing a plane translational movement of the sensor during the radiographic shooting. Same for the existence of the lever (L): it is necessary to understand any means of mechanical or electromechanical correlation, between the movement of the image sensor and the movement of the diaphragm (DF) to obtain the permanent limitation of the X-ray field to the minimum necessary. The incremental movement translator (MT) can be of the rotation type as in FIG. 9 or of the translation type and in this case, being mounted differently; The important thing is that it can translate the translational movement of the image sensor (IS) into electrical pulses.

Second example of use of the SCRX system

FIG. 10 is a kinematic diagram showing how the SCRX operates when it comprises an electromechanical subsystem (MEC) having a kinematic diagram of the second category (mobile generator-object). Such a system can be used to take panoramic dental x-rays. It makes it possible to deliver images of all the dental arches by unrolling them from one temporomandibular joint to the other. The example was chosen as a generic example of all devices with relative generator-object movement whatever the trajectories or means of motorization used and whatever their fields of application.

The beam generator (G), the diaphragm (DF) and the image sensor (IS) are fixed together by an arm (BS). The arm (BS) is attached perpendicularly to the central axis

(AC) of the cam (CM). The movement of the cam is restricted by the lateral slides (GLC) so that it can rotate and / or slide by translation (see the arrows on the drawing). The motor (M) is mounted on a carriage (CHM) rolling on rails (GLM) and allowing its translation perpendicular to the translation of the cam (CM) (see the arrows on the drawing). Its axis (AM) is eccentrically coupled to the cam (CM). At the other end of the axis (AM) is the incremental position translator (MT). The limit limit switches (L1, L2) detect the movement of the cam (CL) attached to the central axis (AC) of the cam (CM).

When the motor (M) is running, the arm (BS) is driven in rotation and translation at the same time. The sensor (IS) moves relative to the object to be radiographed (for example the head of a patient, which is here the object to be radiographed) on an elliptical trajectory. The X-ray beam is permanently delimited in the field corresponding to the active surface of the sensor. The images of the various parts of the head are simultaneously projected onto the sensor. They move on the active surface of the sensor at different speeds. An exact mathematical correspondence can be established between the position of the image sensor on its trajectory and the speed of the image of the region of interest (for example the dental arches) on the sensitive surface of the sensor. This correspondence is given by a formula of the type:

Vi = Vm * K * g (a) where

Vi is the image speed of the region of interest with respect to the sensitive surface of the sensor; K is a constant of dimensional proportionality;

Vm is the speed of the motor in one-to-one correspondence with the speed of the image sensor (IS); g (a) is a function of the position a of the sensor, parameterized by the shape of the jaw, the generator-sensor distance, the position of the jaw relative to the device and the distance between the axis AC and the AM axis.

We can reformulate this expression as follows: Vi = Vm * k / f (a) where f (a) = l / g (a) is said form factor. This function f (a) must be pre-calculated and put in the table of the generator of the servo signal (SVPG) which makes the servo of the displacement of the electron image integrated by the sensor. Thanks to this enslavement, the integrated image of electrons follows the projected image of interest, which is essential, as this makes it possible to sharpen this image and blur the others without further processing after scanning. It is important to note that this function makes it possible to adapt the SCRX to variations in the profiles of the regions of interest, in our example jaw profiles. This is the case when one passes from a child to an adult or from a woman to a man or even in some cases from one man to another.

With other types of kinematic diagrams, the generator table (SVPG) filled with pre-calculated values makes it possible to make various tomographies to study an object in three dimensions.

Claims

R E V E N D I C A T I O N S

1 - X-ray image scanning system, comprising:

- an X-ray image sensor (IS) based on a CCD multiline charge transfer device, and on a block of optical fibers and a scintillator screen (PH) arranged opposite the CCD charge transfer device to allow an image of interest, in X-rays, of the object to be radiographed which is projected on the screen to be converted by this one into an image in visible light and to be integrated by the CCD device;

- electronic circuits for controlling and processing the sensor signal (IS);

- an electromechanical subsystem (MEC) allowing the sensor (IS) to scan the object (O) exposed to X-rays; characterized in that:

- the X-ray image sensor (IS) comprises several structures each composed of an elementary CCD device, preferably rectangular and of a scintillating screen end, the structures (PH, FB, CCD) being arranged, on a length of the sensor, preferably in the same plane, and, at least for part of each structure, zigzag over at least two parallel rows. 2 - System according to claim 1, characterized in that each elementary CCD device has lines and columns, these lines are oriented along the length of the sensor, these lines are preferably longer than these columns, and in that the spacing between two neighboring structures in the direction of the length of the sensor is equal to the active length of the line of a CCD device elementary minus a few pixels, the offset of the structures of one row relative to the structures of the other row being substantially equal to said spacing, so that the elementary CCD devices used in the sensor do not produce any shaded area between them in the acquired image.

3 - System according to claim 1 or claim 2, characterized in that each elementary CCD device is covered on its sensitive face by a parallelepipedic block (FB) of optical fibers which conduct, preferably faithfully without reduction or enlargement, the image from the end of the screen to him.

4 - System according to one of claims 1 to 3, characterized in that it comprises

- a computer (C) equipped with an interface (I) with the sensor through which the computer receives from the sensor (IS) digital information representing the image of interest of the object to be radiographed.

5 - System according to one of claims 2 to 4, characterized in that the electronic circuits for the control and for the processing of the sensor signal (IS) are provided so that the displacement of the lines (N, N + l ,. ..) of the photo-electron image on the surface of each elementary CCD device, occurs line by line towards the output of the CCD and is controlled, according to the image of interest to be selected, on the scanning of the object (O) by the image sensor (IS), so that, among all the images of the various regions projected simultaneously on the sensor (Is), only one image region of interest is followed with precision and is left clear by integration while the others are blurred. 6 - System according to one of claims 1 to 4, characterized in that the electronic circuits for controlling and for processing the sensor signal (IS) are provided so that the displacement of the lines (N, N + 1 ,. ..) of the photo-electron image on the surface of each elementary CCD device, occurs line by line continuously towards the output of the CCD, at the same time and in the same place as the integration of the photo-electrons corresponding to the projection of the image of all the regions of the object and therefore at the same time as the scanning of the object (0) by the image sensor (IS).

7 - System according to one of claims 1 to 6, characterized in that the electro-mechanical subsystem (MEC) comprises means (MT) for translating or controlling the movement of the image sensor (IS) by pulses electrical (IPT) frequency dependent on the speed of the sensor relative to the object (O) and means (L1, L2) of detection and translation into electrical signals of the limit of scanning stroke.

8 - System according to claim 7, characterized in that it comprises a servo signal generator (SVPG) which performs one or more of the following functions: a) receiving and storing a pre-calculated function having as variable the position of the sensor (IS) on its trajectory, a function called form factor, to adjust the servo of the displacement of the image integrated by the image sensor (IS) so that the system makes the desired region of the object clear ( O) to X-ray; b) supplying, during the scanning of the object (O) to be radiographed, the displacement pulses (VSH) of the image integrated by the image sensor (IS); the instantaneous frequency of these pulses being the result of the multiplication of the instantaneous frequency, provided by the means (MT) to translate, by a fixed scale factor (SF) during the shooting, divided by the form factor corresponding to the current position of the sensor (IS ) and written in his memory; c) receiving the position information provided by means of limit of travel detectors (L1, L2), installed in the electro-mechanical subsystem (MEC), allowing it to determine the period in which the pulses must be sent (VSH) vertical displacement towards the sensor (IS); d) transmit, on request, to the computer (C), the position of the image sensor (IS) on its trajectory; e) be supplied by the same single-circuit cable (CV) which is used for communication with computer C;

9 - System according to one of claims 7 to 8, characterized in that the electronic circuits for controlling and for processing the sensor signal (IS) comprise:

- a high frequency oscillator (OSC), stabilized with quartz, which serves as a precise time base; a synchronization circuit (SYNC) of the electrical pulses (IPT) coming from the means (MT) for translating, which synchronization circuit delivers pulses (SIPT) synchronous to the time base;

- a first block of circuits (SCALER) which receives at its inputs said synchronous pulses (SIPT) and a digital constant called scale factor (SF) and delivers at its output first output pulses (SCIPT) with a frequency equal to multiplying the frequency of said synchronous pulses (SIPT) by the digital constant (SF); - a second block of circuits (DIV) which receives said first output pulses (SCIPT), divides the frequency by a number (RMD) which is a current value of the form factor, and to provide as output second pulses of output (VSH) necessary for moving the image integrated by the sensor (IS);

- a memory (MEM), which can be read and written, in the form of a table representing the pre-calculated values of a function of position of the sensor on its trajectory, function called form factor.

10 - System according to claim 4 and claim 9, characterized in that it comprises a third block (SCTL) of circuits ensuring several of the following functions: a) communicating with the computer (C) to receive functions called factors of shape and scale factors specific to the profile of the object to be radiographed; b) transmit, on request, to the computer (C), the position of the image sensor (IS) on its trajectory; c) receive the position information supplied by the limit switches of the travel limit (L1, L2), take it into account to determine the period in which it is necessary to send said vertical displacement pulses (VSH) to the sensor (IS) ; d) counting the pulses (SIPT) to know the current position of the image sensor (IS) on its trajectory therefore determining the memory addressing index (MEM); d) dialogue with memory (MEM); e) providing the multiplication block (SCALER) with the multiplication constant (SF) to be used; f) provide the division block (DIV) with the current form factor (RMD) to be used for the division. 11 - System according to one of claims 1 to 10, characterized in that ^' the image sensor (IS) comprises means for operating in one of several different scanning modes by a pairing factor and / or for operating in a test mode when it sends digital image signals instead of test pattern signals to test the operation of the system.

12 - System according to one of claims 1 to 11 and according to claim 4, characterized in that the interface (I) of the computer (C) comprises means for performing some or all of the following functions: a ) send the electric current supplying the image sensor (IS); b) receiving the image sent by the image sensor (IS); c) send configuration signals to the image sensor (IS); d) send the electric current supplying the generator of the control signals (SVPG); e) send programming signals to the servo signal generator (SVPG); f) rapidly transfer the data stream representing the digital image from the image sensor (IS) to the computer;

13 - System according to one of claims 4 and 1 to 12, characterized in that it comprises a single single-circuit cable (CI) for connecting the image sensor (IS) to the computer interface (I).

14 - System according to claim 13, characterized in that it comprises a single single-circuit cable (CV) for connecting a generator of the control signals (SVPG) to the computer interface (I). 15 - System according to one of claims 1 to 14, characterized in that an elementary CCD device is long and is mounted and connected to an intermediate support means (SUPP) which in turn is connected to a printed circuit (PCB ) bearing the electronic components (CE) of the electronic circuit of the sensor.

16 - System according to claim 15, characterized in that said charge transfer device (CCD) is not a single very long semiconductor chip but that it is composed of a succession of shorter chips juxtaposed end to end with high precision on a support means (SUPP) to achieve the functional equivalent of a chip at the length requested by the use of the sensor without breaking the periodicity of the pixels in the total line.

17 - System according to one of claims 1 to

16, characterized in that each elementary CCD device is associated with an inclined parallelepiped fiber block (FB) to reduce practically to zero the difference between two rows of scintillator screen ends (PH) corresponding to two adjacent rows of CCD devices elementary. 18 - System according to one of claims 1 to

17, characterized in that each elementary CCD device is mounted and connected to a support means (SUPP) which in turn is connected to a printed circuit (PCB) carrying electronic components (CE) of the electronic control and digitization circuit of the sensor.

19 - System according to one of claims 1 to

18, characterized in that each elementary CCD device is mounted and connected on an individual support (DIP) provided with connection pins which in turn are connected to a printed circuit (PCB) carrying the electronic components (CE) of the electronic control and digitization sensor circuit.

20 - System according to one of claims 1 to 19, characterized in that the structures of elementary CCD device type, block of optical fibers

(FB) and scintillator screen end (PH) are placed in a rigid case (B), said case

- is covered by a front cover (CF) made of material that absorbs little X-rays; the case is filled with a layer of elastic resin (SIL) at the level of the CCD chips, making it possible to isolate them from the mechanical point of view and on which is a second layer of rigid resin (RE), at the level of the centers of gravity blocks (FB) of optical fibers encasing these blocks of fibers (FB) and fixing them relative to the housing (B);

- includes

- a plate (Pb) of X-ray absorbent material, arranged in parallel and close to the scintillator screens (PH) and having one or more openings allowing free access of the X-ray image only to these screens and preventing other rays to progress in the sensor; - electrical connectors (BNC1, BNC2) fixed to the housing (B);

- a printed circuit (PCB) carrying the electronic components (CE) of the electronic control and digitization circuit of the sensor which is fixed on the outside of the housing (B) on the side opposite to the front cover (CF) a rear cover (CA ) covering the printed circuit (PCB). 21 - System according to claim 20, characterized in that on the printed circuit (PCB) are mounted several of the following circuits: an analog-digital converter (A / D) which receives at its input a video signal output from a only CCD charge transfer device used in the sensor (IS) or, if there are several CCD elementary circuits, successively the video output signals of all the CCD elementary circuits after having been multiplexed by a multiplexer block (MUX);

- circuits for applying the correct voltage levels to the elementary circuit (s) CCD;

- circuits (POL) for the polarization of the elementary circuits CCD; - a modem (RX / TX) for one-way or two-way communication with the computer (C);

- a high frequency oscillator (OSC) serving as a time base for the entire system; - a supply voltage regulator (REG);

- circuits (CC, CF, LF) for the separation of the modems carriers from the supply current;

22 - System according to one of claims 1 to 21, characterized in that on the printed circuit (PCB), there are mounted electronic circuits comprising a control block (SCANCTL) which performs several of the following functions: a) receiving from from computer C the configuration parameters through the RX / TX modem; b) support several image scanning modes which differ by the matching factor of the rows and columns of the elementary CCD circuits and a test mode which allows the generation of a test pattern which replaces the image to be digitized; c) receive, in digitization mode, from the generator (SVPG) the synchronization pulses (VSH) for the vertical displacement of the elementary circuit (s) CCD); d) receiving, in scanning mode, from the A / D converter the data representing the pixels of the scanned image; e) send, in scanning mode, through the RX / TX modem the data representing the pixels of the scanned image; f) send, in test mode, through the RX / TX modem the data representing a test pattern; g) deliver, in digitization mode, the control signals to the A / D converter; h) issue, in scanning mode, the selection commands to the MUX multiplexer; i) deliver, in scanning mode, the displacement pulses, through drivers (DRV), to the elementary circuit (s) CCD, in accordance with the established configuration and synchronous to the pulses received from the generator (SVPG); j) send, in scanning mode, pulses to the POL block of bias circuits of the

CCD- s. 23 - System according to one of claims 1 to

22, characterized in that the elementary CCD devices are of the solid frame type.

24 - System according to one of claims 1 to

22, characterized in that the elementary CCD devices are controlled by an electronic circuit allowing the displacement of the integrated photoelectron image at a speed proportional to the frequency of servo pulses.