Medical imaging methods are of central importance in patient diagnosis, and the methods are increasingly associated with performing various treatments. Digital X- ray imaging has increased rapidly among the imaging methods, due to digital electronics and computers, and today a substantial proportion of X-ray images are captured digitally. Increasing digital X-ray imaging creates new and different demands for imaging methods and arrangements, in contrast to conventional imaging.

Among the newest imaging methods, digital imaging plate technology enables normal cassette imaging directly in digital form. Instead of a film and an intensifying screen, the imaging cassette includes a reusable plate with a photostimulable phosphor layer (e. g. a europium-activated barium fluorohalide compound). The phosphor crystals in the imaging plate absorb the quantum of the radiation, and the energy of the quantum causes the electrons to be excited to the upper electron orbits. Part of the electrons of the phosphor crystals remain in a seemingly stable status, and thus it is said that a latent picture has been formed. The radiation used may be e. g. X-ray, a, P, or y radiation or ultraviolet radiation.

However, the term X-ray is used here, although the radiation could as well be radiation of any of the types mentioned above.

After the exposure of the X-ray image, the phoshor plate is placed in an imaging plate scanner, in which it is read by scanning over it pixel by pixel e. g. with a helium-neon laser beam with a wavelength of 633 nm. The quanta of the laser beam cause the electrons to de-excite, whereby the excited electrons emit radiation with the energy and wavelength of the energy difference between the excited state and the ground state. The energy stored in the state can be emitted as visible light, for example. This phenomenon is called photostimulated luminescence.

The emitted light is generally guided through focusing lenses and light guides to a photomultiplier tube (PMT), in which the information is converted to an electric signal and further to a digital output signal with an A/D converter. The signal is taken to an image processor, in which a digital image is formed. Because the diameter of the laser beam can be made very small, this technique provides a good resolution, although the image scanning time becomes longer. In the current devices, the matrix size may be e. g. 4096 x 4096 pixels and the contrast area e. g.

212 or 4096 different shades of grey.

Fig. 1 shows a typical prior art X-ray imaging method, in which the object 102 to be imaged is placed between the imaging plate 101 and the X-ray tube 103 exposed to an accurately delimited and directed primary radiation beam 104. The X-ray tube 103 and the generator 105 that is used as its power supply is controlled from the control console 106. The quality and properties of the image captured can be modified already in the imaging situation by means of imaging and radioscopy parameters on the control console 106 and by means of the X-ray tube, anti-scatter grid and focus. The imaging values used depend on the properties of the object to be imaged, such as its absorption, thickness and size, and the age and other details of the patient to be imaged. For the most typical types of images there may be preselections in the control console, consisting of combinations of different parameters. Imaging values can also be selected manually by means of a keyboard in the control console.

Fig. 2 shows a well-known digital imaging plate scanning system 201. Typically the system 201 includes a separate imaging plate scanning unit 202, to which the imaging plate 101 including a latent image is placed for scanning the image in the imaging plate. The scanned image is digitalized 203, after which it can be saved in the memory 204 of the scanning system or alternatively printed on film 205. In addition, the scanning system often includes an extensive menu 206, which may contain e. g. preselections by which the properties of the captured image can be modified. The preselections can be used e. g. to select the anatomic part of which the image has been taken, the direction and the type of imaging plate used. In addition, the menu may include basic operations that have an effect on the image quality, such as selections that modify the darkness of the image. The scanning system often also includes arrangements for image analysis 207, such as forming a histogram of the image signal of the image being processed for a histogram analysis.

The system shown in Fig. 2 essentially corresponds to the prior art solution for scanning a radiation image as presented in the publication US 4 851 675. In this publication, there is disclosed a solution for scanning the image from an imaging plate and for automatic processing of the received image signal with an imaging plate scanning device, in which solution the image signal is saved in the memory of the scanning device and a histogram analysis of the image signal is automatically carried out. In the solution according to the publication, the final image is obtained by forming with the scanning apparatus a compound of the image stored in the memory and the image which has been processed by means of the parameters of the histogram analysis of the original image.

A problem associated with the prior art solutions is the absence of a generally applicable and user friendly digital X-ray image processing method and arrangement. In the prior art arrangements, the basic operations of image processing have generally been integrated into the scanning apparatus, whereby the system has become awkward to use. In addition, it is typical of the prior art solutions that a histogram is automatically formed of the image signal, and the information contained by the image is analyzed on the basis of the histogram. However, the uncertainty factors associated with methods based on a histogram analysis constitute a problem in forming a diagnostically satisfactory digital image.

The objective of the present invention is to provide a user friendly solution for forming diagnostically satisfactory X-ray images. In addition, it is the objective of the invention to implement the solution advantageously so that the whole dynamic range of the input can be utilized in image formation. Finally it is an objective of the invention to effect considerable savings in storage capacity as compared to the prior art solutions, when digital images are stored.

The objectives of the invention are achieved by providing facilities for interactive processing of the input or image matrix and for storing the final result as scaled, whereby the scaled final result typically contains only a certain quantity of significant bits.

The method according to the invention for processing a digital image matrix is characterized in that it comprises steps in which - a digital image matrix is formed, - at least one signal processing parameter is set, - at least one signal processing is performed on the digital image matrix by means of at least one of the signal processing parameters set for obtaining an immediate result, - an immediate result is presented, - the need for signal processing with the changed signal processing parameters is assessed, - when required, a new value is set for at least one signal processing parameter and at least one signal processing is performed with the new value of at least one of the signal processing parameters set for obtaining a new immediate result, and - the final result is scaled from the immediate result obtained.

The arrangement according to the invention for processing a digital image matrix is characterized in that it comprises - means for forming a digital image matrix, - means for setting at least one signal processing parameter, - means for performing at least one signal processing on the digital image matrix by means of at least one of the signal processing parameters set for obtaining an immediate result, - means for presenting an immediate result, - means for setting, when required, a new value for at least one signal processing parameter and for performing signal processing by the new value of at least one of the signal processing parameters set for obtaining a new immediate result, and - means for scaling the final result from the immediate result obtained.

Some preferred embodiments of the invention are described in the dependent claims.

The invention provides considerable advantages as compared to the prior art solutions. The user can process digital image matrices with e. g. a personal computer, CRT monitor, workstation or a corresponding terminal in which a method and arrangement according to the invention have been installed. The processing of a digital image matrix can be performed interactively, whereby the user can immediately see the effect of the image processing measure on the final result of the image being processed. Processing an image matrix is very similar to the processing of a film image, and the final result is diagnostically satisfactory without further processing. In addition, the signal conversion or the so-called LUT (Look-up table) conversion can be selected so that the final result obtained corresponds to different qualities of film and that the brightness adjustment corresponds to the speed of the film in an analogue manner.

In the method according to the invention, the values of image processing parameters determined or used by the user can be saved so as to correspond to a certain object of imaging, such as a certain anatomic part. The parameter values can be saved e. g. in preselection or quick selection menus, in which case-when a corresponding type of image is processed later-the values of the image processing parameters can be implemented by selecting values corresponding to that type of image from the preselected settings or a menu. Alternatively, the parameter values used by the user can be saved so as to correspond to personal habits or preferences, in which case a user who likes to look at a picture in a lighter shade than other users, for example, can select his or her personal preselection settings at the beginning of the image matrix processing. In the method according to the invention, the preselections can advantageously also be arranged so that a combination of the user's preference and a certain image type can be saved in a certain preselection if, for example, the user is accustomed to looking at a certain anatomic image in a different way than other users.

In addition, the parameter values can be arranged to correspond to some other property of the arrangement, such as the properties of the monitor or a certain scanning apparatus, or the lighting conditions of the room, which may be advantageous when operating in a darkroom or a corresponding place with special lighting.

The processing of a digital image matrix can advantageously be carried out semi- automatically, whereby the method can implement certain preselected values, which may be dependent on the object of the imaging, the user or the monitor. When desired, the semi-automatism can also be totally ignored by using the preselected settings directly.

In addition, the invention can help to achieve substantial savings in storage capacity in relation to the size of the received input or image matrix, because the end result can be scaled from the immediate result so that the desired reduction of the dynamic range is realized. The reduction of the dynamic range is preferably performed by defining the most significant bits from the immediate result.

As input data the method according to the invention typically uses a digital grey shade image with a dymamic range wider or equal to that of the end result, such as 16 bits. At the beginning, the input can be shown to the user in reduced size, after which the user can adjust the brightness as well as global and local contrast of the input interactively so that the effects of the adjustments can be seen immediately in the input being processed, the image matrix. The values of the control parameters can be altered in arbitrary order, and the parameter values or the image can be saved or printed in any step of the embodiment.

The selection controllers for the values of the control parameters can be e. g. sliding bars shown on the display, which are moved on the display using a mouse, keyboard or any other user interface. The sliding bars can be advantageously arranged so that one sliding bar corresponds to brightness, one to global contrast and one to local contrast. The brightness controller can be used to select digital amplification of the incoming image matrix, the global contrast controller to select the conversion of the digitally amplified signal or the so-called LUT conversion, and the local contrast controller to select the contrast improvement conversion of the unamplified input.

For processing the image matrix, it is also possible to select predetermined values of control parameters or combinations thereof by means of preselections.

The following section is a more detailed description of preferred embodiments of the invention with reference to the accompanying drawings, in which Figure 1 shows a typical prior art method of X-ray imaging, Figure 2 shows a well-known digital imaging plate scanning system, Figure 3 is a flow chart of an arrangement for the steps of the method according to the invention and solutions related to it for interactive processing of a digital image matrix, Figure 4 is a flow chart of an embodiment of the method according to the invention for interactive processing of an image matrix, Figure 5 shows an arrangement according to the invention for interactive processing of a digital image matrix, Figure 6 shows a user interface according to the invention for interactive processing of a digital image matrix, and Figure 7 shows three different types of LUT functions.

Figures 1 and 2 have been dealt with above in connection with the description of the prior art.

Fig. 3 is a flow chart of an arrangement 300 for the steps of the method according to the invention and the solutions related to it for interactive processing of a digital image matrix, wherein a digital X-ray image is captured in step 301. The image can also be captured in other ways, e. g. as a video image or by means of magnetic imaging apparatus or CCD cells. The image can also be a tomography image or scanned from a film. With regard to the processing of the image matrix, the source of radiation used to capture the image or the type of radiation is of no significance; it can be any radiation that causes an image to be formed on a medium intended for this, such as a film or imaging plate to be processed digitally.

The X-ray image taken is read from the imaging plate in step 302 with a scanning system intended for this purpose. The image received can be transferred directly to a workstation which has the method and arrangement according to the invention for interactive processing of an image matrix. Alternatively, the image can also be stored in the memory of a scanning system or a database of an information network, from which it can be retrieved for processing to a workstation. In step 303, a digital image matrix is formed of the input received, and in step 304 interactive signal processing is performed on the digital image matrix by means of the method and arrangement according to the invention. Interactive signal processing is carried out by adjusting image processing parameters, the effect of which on the image matrix can be seen immediately. The adjustment of the image processing parameters can be continued until the result is as desired.

In step 305, it is possible to choose whether to print the image matrix formed as a result of interactive signal processing on film, and when this alternative is chosen, the image is printed on film with a film printer in step 306. Naturally, the image can also be printed on other printing materials. In some situations, printing the image on film is more advantageous than using just a digital image or analyzing it with a display terminal. As an alternative, the image matrix can be printed on film in other steps of the embodiment, such as steps related to interactive signal processing in particular, although it is not shown in the flow chart. Some steps related to interactive signal processing 304 according to an embodiment are shown later in image 4. In step 307, the image matrix formed as the result of interactive signal processing can be stored in the memory of the workstation or the database of the information network, for example. The storing of the image matrix, as well as printing, can also take place in other steps of the embodiment, although it is not shown separately in the flow chart.

Fig. 4 is a flow chart of an embodiment of a method according to the invention for interactive processing of a digital image matrix, which is step 304 in Fig. 3. At the beginning of the processing in step 401, the user is shown the input received as the result of scanning or the digital image matrix formed, which can be e. g. a digital grey shade image. The image can be shown in miniature, for example. The original, unprocessed input is denoted by the symbol A. The dynamic range of the input A can be advantageously 16 bits, but it can also be higher or lower. Especially the dynamic range of a grey shade image received as input can be higher or as high as that of the final result.

In step 402a, it is possible to choose whether to carry out image processing using ready-made combinations of adjustment values for image control parameters, such as brightness and global contrast (concerning the whole image area) as well as local contrast (concerning a certain location) by means of preselected settings, also called presets, which have been programmed in advance. If it is decided to select presets, the next step is 402b. Otherwise, the next step is 403a. In step 402b it is possible to select e. g. parameters corresponding to a certain anatomic part from a suitable preselection, after which a new, immediate result is formed of the image matrix in step 402c. The immediate result is shown in step 402d. From the immediate result the user can see the effect of the preselection made in step 402b on the image matrix being processed. After the immediate result formed, the procedure returns to step 402a, in which it is assessed if the result is satisfactory. If the result is satisfactory, the next step is 403a, and if the result is not satisfactory, the procedure moves to step 402b again. The preselections chosen in step 402b can correspond e. g. to the following anatomic parts that are often imaged: the skull, ears, sinus, cervical spine, thorax spine, lumbar spine, coccyx, SI joints, shoulder joint, brachium, elbow joint, antebrachium, wrist, palm, finger, pelvis, articulatio coxae, femur, knee, tibia, ankle, metatarsus, toe, ribs, sternum, clavicle, thorax pictures, trachea, urography, kidney, cysto, foetus and pelvimetry. The preselections may also include parameter values corresponding to the user's habits and preferences or other adjustment values or correction coefficients related to the arrangement. In addition, the selections may include a preselection, which comprises a combination of many different types of parameter adjustments, such as a combination of parameters dependent on a certain user, monitor error, an image representing a certain anatomic part or other corresponding matters.

In step 403a, it is possible to choose whether the brightness of the image matrix should be adjusted. If it is decided to adjust brightness, the next step is 403b, in which the brightness of the image matrix can be altered e. g. steplessly with a sliding controller. Otherwise, the next step is 404a. After brightness adjustment, a new immediate result is formed of the image matrix in step 403c, and the result is shown in step 403d, whereupon the user can immediately see the effect of the brightness adjustment on the image matrix being processed. After the immediate result has been shown, the procedure returns to step 403a, in which it is assessed if the result is satisfactory. If the result is satisfactory, the next step is 404a, and if the result is not satisfactory, the procedure returns to step 403b. The brightness adjustment made in step 403b can have an effect on the digital amplification of the original image matrix A so that the new immediate result B to be formed after brightness adjustment in step 403c is of the form B=k>A, where k, is brightness adjustment.

In step 404a, it is possible to choose whether the global contrast of the image matrix should be adjusted. If it is decided to adjust global contrast, the next step is 404b, in which the global contrast of the image matrix can be adjusted e. g. steplessly with a sliding controller. Otherwise, the next step is 405a. After global contrast adjustment, a new immediate result is formed of the image matrix in step 404c, and the result is shown in step 403d, whereupon the user can immediately see the effect of the global contrast adjustment on the image matrix being processed. After the immediate result formed, the procedure returns to step 404a, in which it is assessed if the result is satisfactory. If the result is satisfactory, the next step is 405a, and if the result is not satisfactory, the procedure moves to step 404b again. The global contrast adjustment made in step 404b can e. g. have an effect on the signal conversion of the digitally amplified image matrix B or the so-called LUT conversion so that the new immediate result C to be formed after global contrast adjustment in step 404c is of the form C=Lut (B, k2), where k2 is the global contrast control parameter. By means of the control parameter k2, the desired LUT can be selected from the LUT functions by sliding interpolation, for example. Thus the signal can be processed with a function which the user can modify with a sliding bar, for example.

Advantageously, there can be three of the LUT functions used for forming the desired LUT (low, high, medium), but there may also be more of them when tomography images are processed, for example. In addition, there may be a separate so-called monitor LUT in use, by which it is possible to take into account any distortions caused by the monitor being used and to rectify them. The LUT conversion is preferably performed after the digital amplification adjustment carried out in step 403.

In step 405a, it is possible to choose whether the local contrast of the image matrix should be adjusted. If it is decided to adjust local contrast, the next step is 405b, in which the local contrast of the image matrix can be adjusted e. g. steplessly with a sliding controller. Otherwise, the procedure moves directly to step 406. After local contrast adjustment, a new immediate result is formed of the image matrix in step 405c, and the result is shown in step 403d, whereupon the user can immediately see the effect of the local contrast adjustment on the image matrix being processed.

After the immediate result formed, the procedure returns to step 405a, in which it is assessed if the result is satisfactory. If the result is satisfactory, the next step is 406, and if the result is not satisfactory, the procedure moves to step 405b again. The local contrast adjustment made in step 405b can be used to influence the contrast improvement conversion of the local contrast of the original unamplified image matrix A so that the new immediate result D to be formed after the adjustment in step 405c is of the form D=Ahe (A), where Ahe is the contrast improvement conversion function for local contrast. The local contrast improvement conversion is preferably performed directly on the unamplified input A.

Step 406 comprises forming a combined immediate result T,, which is a combination of the LUT conversion C made on the image matrix and the local contrast improvement conversion D so that TV= (1-k3) C+k3D, where k3 is a figure between [0,1] and which is determined e. g. when local contrast is adjusted with a local contrast controller. Alternatively, the immediate result T, can be formed in other steps of the embodiment, although it is not shown in the flow chart. The formed immediate result T, is shown in step 407, after which it is possible to assess in step 408 whether the combined immediate result T, is satisfactory. If the result is satisfactory, the procedure moves to step 408; otherwise it returns to step 402a for interactive processing of the image matrix. From step 408, it is naturally also possible to move to other image matrix processing steps, although it is not shown in the flow chart.

In step 409, it can be chosen whether to save the parameter values used in image processing, and when this alternative is chosen, the values are saved in step 410.

The parameter values can be saved as a preselection so as to correspond to suitable adjustment values for an image taken of a certain anatomic part, for example.

Saving the parameters is advantageous e. g. if images of a similar type are taken of the same anatomic part later, in which case image processing parameter values suitable for the image type in question can be obtained simply by selecting the preset corresponding to the anatomic part in question. The parameter values can also be saved so that they correspond to the user or his or her habits and preferences. At the beginning of the image processing, the user can be identified e. g. by initials or a corresponding identifier, whereby the preselection values corresponding to the user can be directly activated. In step 410, it can be advantageously decided whether the parameter values should be saved so as to correspond to a user, image type or anatomic part of which the image has been taken or to some other property related to the arrangement or a combination thereof.

Before step 411, the image matrix being processed includes the whole dynamic range of image A received as input. In step 411, the most significant bits can be determined from the processed image matrix according to a predetermined method, and the final result T can be formed in step 412. The final result T can be obtained e. g. by scaling from the final immediate result of interactive signal processing Tv.

The final result T is naturally scaled so that the desired reduction of the dynamic range is achieved. The final result can be scaled e. g. from the original 16-bit image matrix to an 8-bit image matrix, in which case 50% savings of storage space in comparison to the space taken up by the original image matrix are achieved when the scaled image matrix is stored. In step 413, the final result T is presented.

The dynamic range of the immediate and final results presented in steps 402d- 405d, 407 and 413 may be lower or equal to the real dynamic range of the image matrix being processed. The dynamic range of the result to be presented can be reduced to suit the display being used, for example. However, in the image matrix being processed the dynamic range is preferably the same as the dynamic range of the original input, until just before the storage step, when the most significant bits can be determined from the image matrix and the dynamic range thus be reduced and savings in storage capacity achieved.

Fig. 5 shows an arrangement 500 according to the invention for interactive processing of a digital image matrix. The arrangement according to the invention typically comprises means 501 for obtaining an input A formed as the result of image scanning and means 502 for forming a digital image matrix. After scanning, the input can be transferred automatically from the scanner 513 to the image matrix processing system or it can be retrieved from different kinds of archives or sources by means of the processing system. The archives and sources can be, for example, a film scanner 514, a mass storage 515, a database 516 of an information network and the Internet 517. In addition, the arrangement comprises means 503 for presenting the input received or a digital image matrix formed thereof on a CRT display 523, for example.

The arrangement is characterized in that it comprises means 504 for performing digital amplification and brightness adjustment of the image matrix, means 505 for performing a LUT conversion and global contrast adjustment of the image matrix, means 506 for performing a local contrast improvement conversion and means 507 for forming an immediate result Tv. The immediate result formed is a compound for combining the signal conversion performed on the digital image matrix and the contrast improvement conversion performed on the image matrix for forming an -immediate result. The means 504,505 and 506 can be sliding bars, for example.

In addition, the arrangement comprises means 508 for forming a final result T so that it is scaled from the immediate result T,. Furthermore, the arrangement comprises means 509 for presenting the immediate result and the final result by means of a CRT display 523, for example. The arrangement may also include means 510 for storing the digital image matrix in different types of archives or storage locations, such as a mass storage 515, a database 516 of an information network or Internet 517 and means 511 for printing the image matrix with a printer 518, for example.

The arrangement according to the invention typically also comprises means 512 for the use, modification and storage of preselected settings for processing a digital image matrix. The preselected settings may include settings according to the type of image 519, the users'own settings 520, special settings 521 for the display being used and a library of the LUT functions used 522. Settings according to the type of image may comprise e. g. adjustment values concerning a certain anatomic part or children or pregnant women, for example, and the user's settings may comprise image adjustment values corresponding to the user's own habits or preferences. The special settings may include adjustment values concerning scanning apparatus, printers or the display or other corresponding correction or control values related to the arrangement. In addition, the arrangement may comprise combinations of the above mentioned values as presets.

Fig. 6 shows a user interface 600 according to the invention for interactive processing of a digital image matrix, which user interface comprises an area 601 for presenting the digital image matrix, and sliding controllers 602,603 and 604 for adjusting the signal processing parameters. In addition, the user interface may include means of the type of quick selection or menus for using preselections concerning the type of image or the user. It should also be noted that the sliding controllers and the area 601 can be located in a different order in the user interface, there may be varying numbers of them and that sliding controllers can also be replaced by other types of control means, such as numerical fields or command line.

The sliding controller can also be a mechanical controller. The sliding controller can be used to select the parameter value from a set of values, which may include 256 different parameter values, for example. The user interface can also comprise means for performing other known image processing procedures, such as enlarging or decreasing the image, menus for adjusting the angle of view, and means for determining lengths, surface areas and volumes.

Fig. 7 shows three different types of LUT functions that can be used for performing the global contrast adjustment of the method according to the invention or performing a LUT conversion on the digital image matrix being processed. The figure shows LUTs of low 701, high 702 and medium contrast 703, but there may also be a different number of LUT functions used. For example, in the processing of tomography images it may be advantageous to use in addition to the three LUTs mentioned, one more LUT function in the area of low contrast, and yet another LUT function in the area of high contrast. In addition, it is possible to use a separate monitor LUT, in which distortions caused by the monitor, for example, are taken into account. In the method according to the invention, the LUT used can be formed e. g. by linear interpolation from different LUT functions.

Some preferred embodiments of the solution according to the invention have been described above. Naturally, the principle of the invention can be modified within the scope defined by the attached claims e. g. with regard to the details of implementation and the range of use. Especially the various embodiments of the invention are not limited to the above examples or the processing of X-ray images, but in the scope of the invention the method and arrangement according to the invention can also be used for processing other types of image matrices.