WO2004023790A2

WO2004023790A2 - Patient identification method for x-ray film user-interfaces

Info

Publication number: WO2004023790A2
Application number: PCT/CA2003/001285
Authority: WO
Inventors: Wido Menhardt; Yue Shen
Original assignee: Eastman Kodak Company
Priority date: 2002-09-04
Filing date: 2003-09-04
Publication date: 2004-03-18
Also published as: EP1535462A2; US7494045B2; US20060000884A1; CN1679318A; CA2494103A1; WO2004023790A3; JP2005537101A; AU2003260214A8; AU2003260214A1

Abstract

A method for identifying digitized X-ray films using the label that is 'burned' on such films. The label is digitized and can be used as a single identifier for the corresponding X-ray film image. The digitized label can be displayed such as to be readable by the user thereby allowing the selection and retrieval of the image. The digitized label can also be associated with image analysis results, such as CAD analysis, performed on the corresponding X-ray film image allowing easy retrieval of such results.

Description

PATIENT IDENTIFICATION METHOD FOR X-RAY FILM USE -INTERFACES

CROSS -REFERENCE TO RELATED APPLICATIONS This application claims priority on United States provisional application Serial No. 60/408,079. filed

September 4, 2002 entitled "Patient identification method for CΑD user-interfaces" .

TECHNICAL FIELD This invention relates to the field of X-ray film identification and more particularly to identification of digitized X-ray films in computer user interf ces.

BACKGROUND OF THE INVENTION

X-ray films can be scanned by a film digitizer, and converted into digital format. Digitized films can then be stored for future retrieval or analyzed with image analysis algorithms such as Computer Aided Detection

⁽CAD⁾. Film digitizers are available from, for example.

Canon. ETo tek, Kodak, or Vidar. The film digitizer is; connected t a standard computer and a standard user interlace controls scanning, identifying, and storing the digitized films as images on disk. Standaid user interfaces for digitizer control, including quality control, error detection, and error recovery, are available from Vidar, Applicare, or Merge.

In the diagnosis process, quick and reliable identification of the film is required to increase the efficiency of the process and reduce the risk of errors associa_ted with m sidentification. Furthermore digitized X-ray films are often analyzed with software applications such as _CAD and reliable association of the CAD results with the corresponding digitized X-ray film is very important .

Tn a typical scenario, the radiologist normally reads the film mammograms on an alternator or on a light box. Suspicious findings are noted by the radiologist, and recorded in a medical report. However, a non-zero percentage of cancers are typically missed by radiologists. To decrease the rate of false negative, a second radiologist may read the films, and the results of both radiologists are compared.

The results of CAD can be used in much the same way as a second i diologis . In one scenario, also referred to as "second read with CAD", the radiologist first reads the films Lchout the aid of a computer. Once the radiologist has noted her findings, she reviews the suspicious areas identified by CAD analysis, for example on a mojritor ae an overlay over the digitized versions of the mamraogram, or printed on paper. The suspicious areas may confirm the findings of the radiologist. Or, the radiologist may be prompted to re-examine areas she had not found to be» suspicious, but which were highlighted by the computer.

In another scenario, also referred as "cued read with CAD" , the radiologist may read the films and review the CAD results at the same time, moving back and forth between them using the CAD results to "cue" her reading. In either scenario, it is important to ensure that the _CA_D results are properly associated with the X-ray films to ensure correct identification of the results ana avoid "mix-up" of the films and CAD results Commercial products currently use a number of methods for tc achieve

For example a bar code system can be used. When films are digitized, a separator sheet of the same approximate size of the films is also scanned. It serves as a separator between the films of one case and the films of the next case A bar code is printed on the separator sheet. The digitization system reads the bar code, and digitally associates it with the images. The operator enters no patient name, accession number, or other identifier, the bar code information is the identifier of the case. Only the num πcaJ identifiei corresponding to the bar code (a number! identifies the patient. After the digitization, the 111"is are replaced m the jacket, together with the seoarator sheet with the bar code

nen the films are hung on tne alternator, a bar code reader is used to scan the bar code on separator sheets. The alternator contains a computer and logic to a&sociate the CAD results with a particular alternator position (that of the corresponding X-ray films) . That way, the proper digital images and CAD results are displayed on the mini -monitors integrated into the a.ternator depending on the alternator position The numenca.. ldentifier corresponding to the bar code (together with the alternator osition) i en ifies the patient Th s method is deficient in that radiologists and other medical personnel use patient names and patient record identifiers to identify patients, and cross-correlate patients to other medical inform ion, reports, requisitions, and the like. Since the CAD results are identified only with a barcode and its associated numerical identifier, association of patient names or records with CAD results involves an indirect and unreliable cross-referencing method from CAD results to bar code to patient name. This method is not only inconvenient and time consuming, but also prone to errors .

In other systems a bar code sticker is affixed to each individual film (the sticker includes the orientation information of each individual film) . The digitizer uses the bar code to associate it with the films of the patient. When the films are reviewed, a bar code reader is used to read the bar code stickers on the films, and the appropriate linages are displayed on a monitor, together with the CAD results. In this system the CAD resul s are not associated with alternator positions. This method is equally deficient in that the onlv mechanism to correlate CAD results with a patient name ι-3 to cross -correlate the bar code numbers of the CAD results w th a bar code label affixed to an X-ray film.

Yet another method is described in "Hybrid display for simultaneous side-by-side review of radiographs", United States Patent 5,748,173. In the described system, patient information in the form of a n m or a patient number i.s obtained directly from the film that is placed on a view - b -

box ₍alternator or light box) using a bar code scanner or a _CC_D camera. This method is also deficient in that t^he CA_D results have to be identified by other means with the patient name σr patient number before they can be recalled with the method described here.

Other strategies have also been used to identify X-ray films. For example the operator- can enter the patient name and identification. The patient name is then digitally associated with the digitized images. When a case is reviewed, the radiologist enters the patient natie m a computer, and the appropriate images and CAD results are displayed. Alterna ively, the radiologist ca.n select the patient name from a list. A disadvantage of this method is that the patient name has to be entered manually at the time of digitization, and potentially aqaιr> at the time of review

SUMMARY OF THE INVENTION

In one aspect of the present invention there is provided a method whereby human- eadable patient information on a label is automatically obtained from digitized X-ray film. This human-readable patient information s then u&ed to identify the corresponding X-ray film. In one aspect of the invention the label of a digitized X-ra film is digitally extracted and associated with the corresponding Eilru and displayed such as to be readable by a user This advantageously avoids the need to identify the film by indirect means such as by scanning a bar code, affixed to the film or by hav ng to enter the name or identification number of a patient and therefore overcomes several disadvantages of the prior art-

In one em_bodiment of the invention the digitally extracted label can be enhanced to increase the contrast or to remove textured background to facilitate reading by the user of the alpha-numeric information contained in the labe .

The digitized label may also be associated with results of X-ray film image analysis such as Computer Aided Detection Analysis (CAD) . The human-readable label may be used i user- interf ces related to CAD, including but not limited to, quality control of digitized X-ray films, worklist preparation for review of X-ray films with CAD, selection of CAD results for the purposes of review of X- ray films, display of CAD results during or after review of X-ray films, and documentation of CAD results in electronic orm .

Tn a further embodimen , the digitized label can be used to automatically group digitized x-ray films belonging to the same patient. This can be achieved by digitally comparing the labels of different films using image comparison techniques and using pre-determined correlation criteria to relate the films.

In yet a further embodiment, sub- labels can be digitally extracted from a label and associated with corresponding

X-ray film images and or image analysis results such a?;

CAD. By enabling the identification of images using selected information from the label, the method provides increase flexibility for retrieving desired films

BRIEF DESCRIPTION OF THE DRAWINGS

Further features and advantages of the present invention will become apparent from the following detailed description, taken in combination with the appended drawing, m which:

Fig. 1 is a schematic representation of the identification process using digitized X-ray film label according to an embodiment of the invention,

Fig 2 s a schematic representation of the lanel enhancement process according to an embodiment of the invent! on,

Fig. 9 a schematic representa ion of the aesoc atior and displaying of labels with CAD results according to an embodiment of the invention,

Fig 4 is a schematic representation of the digital extraction and displaying of sub-labels according to an embodiment of the present invention,

Fig. 5 is a schematic representation of the labels lists, to select and retrieve corresponding X-ray film images according to an embodiment of the invention, and

Fig 6 is a schematic representation of label/film displav combinations according to possible embodiments of the invention. _DETAILE_D DE_SCRIPTION OF THE PREFERRED EMBODIMENT

_Individual _X-ray films are labeled with patient information that is "burned" into the f lm. This is accompli_shed with a device that exposes a sma.1 (approximately 3x2 inches) area of the film with alpha- n meπc information that typically lists the patient name, birth date, identification number, accession number, and the name or identifier of the technologist.

Depending on the institution and the devices, the patient information area may be m the same place on all films, or it may vary Tt may be on the edge of the film, or t may be somewhere in the background area outside of the anatomical feature shown on the film It may have black on white text, or white on black text, it may be typed, or computer generated, an the fields and their ali nment may differ from institution to institution, imaging device to imaging device, and exam to exam

In one aspect of the present invention, the area of patient information (i.e the label) can be used as the digital identifier of a patient, in lieu of any other digital identifier. Thus there is provided a method ^ δ system for identifying digitized X-ray films comprising = label which contains alpha-numeric information related to the patient and other information such as but not limited to conditions of exposure of the film

Referring to Figure 1, a digitized image of the label 11 is obtained from X-ray film 10 and is digita ly associated, m look-up table 12, with the corresponding digitized X-ray f lm The digitized label 14 is displayed on a user interface 16 such as computer screen to allow the user to identify the corresponding film using the alpha-numeric information on the label. The image of the label can be displayed alone or simultaneously with the X-ray film image.

In some cases the digitally extracted label may have textured background, be blurred or the contrast may not be important enough to allow easy reading of the information by the user. In these cases the label can be digitally enhanced. By label enhancement it s meant that certain procedures are performed to, for example, remove textured background, increase the contrast between the alpha-numeric characters and the background and enlarge the image. These procedures are well known m the art. Thus, and by referring to Figure 2 the label 14 can be enhanced into an enhanced label (EL) 20 and digitally- associated in look-up table 22 with either the corresponding X-ray film or the corresponding (non- enhanced) label or both. It will be appreciated that if the EL is associated with its corresponding label it can therefore be indirectly associated with the corresponding digitized X-ray film.

Image analysis algorithms such as CAD algorithms can be applied to the digitized films, and the results can be associated with the label (or EL) and corresponding X-rav film images (Figure 3) . The patient information area ⁽label or EL) serves as the unique identifier of the patient, the images, and the image analysis results. Thuis the label can be displayed simultaneously with the X-ray film image and the CAD results or, alternative y, t_he la_bel can be displayed with the image analysis results alon .

_In a preferred embodiment the digitized label is obtained by automatically locating and digitally extracting it from the remainder of the digitized X-ray film. To that effect, pattern recognition and image processing methods, which are known in the art, can be used. Such methods may include but are not limited to edge detection and texture segmentation. For example, in edge detection, the; gradient of pixels intensity of an image is determined and used to identify a portion of the image wherein the gradient corresponds to a predetermined gradient , characteristic of the presence of the label. The extracted digital label will be used to identify the images of a case.

In another aspect of the invention, the user is presented with a user interface to interactively identify the location of the patient information area, and this area is then digitally extracted.

In yet another aspect of the invention, standard locations of the patient information area in films o:E different orientation are stored on the computer disk. Depending on the orientation of the film, the standard location of the patient information area on the film can be looked up, and the patient information area can be extracted from that area.

It should be noted that the extraction of the patient information area corresponds to selection of a sub-image m the _digi_tized _X-ray image, but not necessarily actual elimination of the remainder of the image (the complement of the su_b- image) . As is familiar to those skilled in the art, t_he extraction of a sub- image can be achieved both _by copying the pixel elements of a sub-image into a new image ouffer, and maintaining the sub-image as a separate image; or by identifying a rectangular region the original image, ana representing the new sub- mage m the form of a reference to the original image and the c - ordmates of the rectangular region m that original image Tne actual representation of the extracted area and sub- mage is largely determined by software architecture and design of a particular implementation of the present invention

In this respect, and referring to figure 4, the label 14

(or EL) may be subdivided into one or more sub-labels ^ 0.

As described above for the labels, the sαb-labels can be digitally extracted and associated with the corresponding label, digitized X-ray f lm and/or image analysis lesuits m a look-up table 42 Since more than one sub-label can be extracted from the label, different sub-labels may correspond to the same label of X-ray film image 44 ox image analysis results A sub- label may comprise the name of the patient or the voltage used during exposure of the f lm or orientation information (for example, medio lateral-oblique (MLO) or cramo-caudal (CC) ) , ar.d which s de of the breast was imaged (Left or Right or any other information contained m the label It may also contain a combination or such informat on. The subdivision of the label advantageously allows the use- to recall one or more film images based on mformat-on other than information that is exclusive to one patient such as the name of the patient

_A la_bel detection application can be assigned to a digital image according to tne method described m co- pending application US 10/205,599 incorporated herein by reference The method comprises the steps of. selecting a label acquisition model from a plurality of acquisition models based on at least one attribute of the digital label, and selecting a label detection application from s plurality of label detection applications using the selected label acquisition model

In some embodiments, selection of an acquisition model may include the steps of obtaining at least one attribute of the d_gιtal label, and performing a search to identify the acquisition model using the obtained attributes Tne label detection application may be selected by obtaining information concerning the digital' label m accordance with the selected acquisition model and performing a search to identify the laoel detection application using at least the obtained information.

An example of a system that supports the assignment of a label detection application is also described in co - pending application US 10/205,599 incorporated herein by reference. The system includes a case input system for selecting an acquisition model from a plurality o: acquisition models, bat,ed on at least one attribute of the digital label, and for selecting a label detection application from a plurality of label detection applications using the selected acquisition model

_The system may further include an algorithm server for retrieving and applying the selected label detection application to the digital label.

It will be appreciated that other approaches car. be used to assign a label detection application to a digitized X- ray f lm. For example the application could be default application.

In diagnosis procedures, such as screening mammography, almost the same information s applied to all f ilms of the same patient, although m many cases information pertaining to the individual film, such as X-ray current and orientation is also displayed on the label. Therefore, the labels may differ m some minor aspects.

Thus, all films for the same patient contain an area of about 1x2 inches that is very similar m content, but may be subject to variations m exposure, noise, and othei artifacts. Hence the patient label extracted from one image of the case may be used to represent all images oi the case, or all images of the patient. Thus n another aεpect ot the method of the present invention the images belonging to a patient can automatically be grouped using the digitized alpha-numeric information of the label. For grouping purposes, films of several patients are loaded nto a digitizer, and the films are scanned and converted into digital form. No separator sheets are used, and films of several patients can be mixed. The area oE patient information is automatically identified on each f_ilm, an_d correlated with the corresponding information on other films. Pairs of films with high correlation _values _for these patient information areas are considered to beiong to the same patient, and pairs w th low correlation values belong to different patients- Alternatively, character recognition can be performed on the identification area, and the patient name and other information is obtained directly m alpha-numerical form This way, the patient name could be obtained and used as the identifier of a f lm, ana films of each patient can be grouped based on the patient n me.

Similarity between two films can be determined by comparing the patient information areas (labels) of the two films. Correlation methods, known to those skilled m the art, including but not limited to shift/rotation snd intensity/contrast variations can be used to determine the similarity. A predetermined hign correlation indicates that the two films belong to the same patient and a predetermined low correlation indicates that the two films belong to different patients.

Typically, films of consecutive patients are placed in the input tray of the digitizer Films of one patient are grouped together, followed by the films of the next patient, etc. Therefore, the correlation method will be applied to pairs of consecutive films only. In most cases, there will be groups of four films, which exhibit a high correlation, and thus belong to one patient Correlation between the last film of one patient and the first film of the next patient will be poor. Then, the four films of the next patient will exhibit high mutual correlation therefore enabling the grouping of the films that belong to the sa e patient

_Alternatively, it s also possible to scan films of multiple patients without grouping films of individual patients together. In that case, correlation s preferably performed between all pairs of individual ilms

The digitized labelβ of each film on in a group can be digitally associated with any or all digitized films belonging to this group Therefore it is possible to display all the digitized film m a group by seLectmg only one label It will be appreciated that any othe display combination ("I label 2 films, 1 label 3 films 2 labels 3 films etc.) is possible within one group.

The method is not limited to groups of four films per patient Any number of films per patient can be treated and identified with the correlation method

Using the method of the present invention, an X-ray f lm image can be selected and retrieved from a plurality o- stored mages . Now referring to Figure 5, the labels (or EL or = b- abels) corresponding to all or a predetermined number of the stored films can be displayed on user interface to allow the user to select one or ironi desired X-ray film It will be appreciated that irrag analysis results, such as CAD, can also be retrieved according to this embodiment. For example, the Labels (or EL or sub- labels) of all available patients can be displaye_d on a computer monitor m a list . A technologist creates a "worklist" of CAD results (for example, corresponding to the order of X-ray films on an alternator or light box) through manipulation of digital patient identifiers (i.e. the labels) m a list displayed at the user interface For example, this could be a scrollable l st, and "cut-and-paste" within the list could be used to change the order Or, two lists can be used, one for all available patients, and one tor the worklist. The technologist can move the digital identifiers from one list to the other m the desired order using a mouse or a touch screen user interface or example. Displaying of the labels m a list also enables the user to directly select the name of one or several desired patients without the need to cross -refer the name of the patient with an ID number or to scan a bar code on a film.

Figure 6 shows possible embodiments of label/film displays wherein the label can be displayeo simultaneously with the digitized film and/or with image analysis results. It w ll be appreciated that other display arrangements/combinations are possible

Further, the digital labels could be used to arrange cases into a sequence, for example to create an ordered worklist of patients to be reviewed by a radiologist, the films of the patient are hung on an alternator

Also, the digital labels could be used to arrange the cases into a two-dimensional table, corresponding to the layout of the cases on a light box. _During the reading process, the CAD result3 are displayed on a monitor m proximity to the alternator or view bcx _In typical workflow, the CAD results are displayed after (for a "second read scenario") the radiologist has analyzed the images, to obtain prompts for additional suspicious areas After review of the films, the radiologist selects the corresponding patient information area on the monitor (using a mouse, or a touch screen, for example) . Next , the corresponding images and CAD results are displayed on the same monitor. This is possible because the images and CAD results are digitally associated with the patient information area.

The digitally extracted label can be used whenever the digital image is displayed on a monitor or pπ ted or media Furthermore, by visually associating the digital image with the digitally extracted label using graphics, tabular alignment, color coding, or any other such method, the user can recognize that the label and its' associated image are indeed related, and will thus be able to identify the image as belonging to a particular patient

It w ll be appreciated that the above described method of identifying X-ray images of a patient can be used on t e; digitization station, where images are displayed after digitization for qua ty control purposes or on a CAD results display station, which is used by radioicgists to review areas that may have been deemed suspicious o: disease by CAD. In another aspect of the invention there is provided a system for identifying and retrieving digitized X-ray films having a label, the system comprises- an X-ray film digitizer for generating a digital image of each X-ray film, processor means for determining a position of the label on the digital image of the X-ray film; processor means for digitally extracting an image of the label from the digital image of the x-ray film; processor means for digitally associating the image of the label with the digitized image of the X-ray film, a screen for displaying the label. The system may also comprise a processor means for digitally associating the label w h image analysis results of a corresponding digitized image of the X-ray film and processor means for analysing the images of a plurality of labels

The system may further comprise means for displaying images of one or more labels, user interface means for permitting a user to select one of the displayed labeLs, processor means for retrieving a digitized image of an X- ray film associated u/i h the selected label and display- means for displaying the retrieved digitized images of the X-ray films for examination by the user.

The embodiments of the invention described above re intended to be exemplary only. The scope of the invention is therefore intended to be limited solely by the scopt; of the appended claims

Claims

CLAIMS :

₁. A method for identifying a digitized X-ray film comprising at least one label the method comprising:

( ) obtaining a digital image of said label,

(_ID digitally associating sa d digitized image of the label with said digitized X-ray fiirr,

(in) displaying said digital image of sard label thereby allowing a user to identify said digitized x-ray film

2. The methoα as claimed in claim 1 further comprising the step of selecting sa d label to display said associated digitized X-ray film

3. The method as claimed in claim 1 further compris-ng the step of digitally enhancing said image of the label prior to displaying said image of the J ^' abel

4 The method as claimed m claim 3 wherein said step of digitally enhancing comprises removing textured background and increasing contrast between background and alpha-nαmeric characters m saxα image of the label

5 The method as claimed m claim 1 further comprising the step of digitally associating the label with a corresponding image analysis result of the digitized X-ray film _The metho_d as claimed m claim 5 wherein said image analysis is a Computer Aided Detection (CAD) analysis

_The method as claimed m claim 2 wherein said image of said label is displayed simultaneously with sa d digitized X-ray film.

8 The method as claimed m claim 5 wherein said image of said label s displayed simultaneously with εaic image analysis result.

9. The method as claimed m claim 5 wherein said image of said label is displayed simultaneously with said image analysis result and said digitized X-ray film.

10 The method as claimed m claim 1 wherein said step of obtaining a digital image of the label comprises

ι,ι) determining position of said label on said film;

(n) digitally extracting said image of said label .

11 The method as claimed m claim 10 wherein said step of determining is performed automatically

12. The method as claimed m claim 11 wherein said step of determining position is performed using a method selected from edge detection and texture segmentation. _{13 T}he method as claimed m claim 10 wherein said step of determining is performed by a user.

1₄. A method for retrieving, from a plurality of _digitized X-ray films, a digitized X-ray film said film comprising at least one label, the method comprising

( ) obtaining a digital image of said label;

(ιι! digitally associating said digitized image of the label w th said digitized X-ray f lrn,-

(m) displaying a plurality of labels eich associated with a digitized x-ray film,

(iv) selecting a label from said plurality of labels to retrieve said digitized X-ray film

15 Ih method as claimed m claim 14 further comprising tne seep of displaying said X-ray film.

16 The method as claimed m claim 14 further comprising the step of digitally enhancing said image of the¹ label prior to displaying said image of the Label .

17 The method as claimed in claim 16 wherein sa d step of digitally enhancing comprises removing textured background and increasing contrast between background and alpha-numeric characters m said image of the label

₈. _The method as claimed m claim 14 wherein image analysis results are associated with said digitized image and wherein the method further comprises the step of displaying corresponding image analysis results of the digitized X-ray film.

19. The method as claimed m claim 18 wherein said image analysis is CAD analysis.

20. A method for automatically grouping digitized X-iray films comprising at least one label and belonging to a same case and/or patient the method comprising.

d) obtaining a digital image of said label;

(ii) digitall -associating said digitized image of the label with said digitized X-ray film;

(iii) analyzing images of a plurality of labels, associated with respective digitized X-ray films;

(iv) identifying labels within said plurality of labels having a desired degree of correlation sufficient to indicate that said identified labels belong to said same case and/or patient;

(v) digitally grouping digitized X-ray film^'s corresponding to said identified labels.

21. The method as claimed in claim 20 wherein said step of analyzing is performed using shifr/rotation or intensity/contrast variations. The method as claimed m claim 20 wherein each of sai_d i_dentified bels is associated with each film _belonging to said same case and/or patient and wherein any one of said labels can be selected to effect display of all of said films

A method for assigning a label detection application to a digital image the method comprising steps of

( ) selecting a label acquisition model from a plurality of acquisition models based on at least one attribute of the digital label, and

(n) selecting a label detection application from a plurality of label detection applications using the selected acquisition model

The method as claimed m any one of clanms 1 21 wherein said label is one or more sub- label

The method as claimed m claim 24 wherein .said one or more sub- label comprises information selectee fi om name of said patient, exposure voLtage and orientation

The method as claimed m claim 1, 14, 20 or 22 wherein said digitized X-ray f lm is displayed simu aneously with said label and wherein said label and digitized X-ray film are visually associated. 7 _A system for identifying and retrieving digitized X- ray films having a label applied to each f lm, the system comprising:

(_l) an X-ray film digitizer for generating a digital image of each X-ray film;

(n ) means for determining a position of he label on the digital image of the X-ray f^'l'πi;

(m) means for digitally extracting an image of the label from the digital image of the x-ray film;

(iv) means for digitally associating the, image of the label with the digitized image of tne X- r y film,-

(v means for displaying the label

28 The system as claimed n claim 23 further comprising means for digitally associating the label with irrags- analysis results of a corresponding digitized image of the X-ray film.

29. The system as claimed m claim 24 wherein sard image analysis is CAD analysis.

30 The system as claimed m claim 24 further comprising means for analyzing the images of a plurality of labels

1. The system as claimed m claim 24 further comprising means for displaying im ges of o or more labels, means for permitting a user to select one of the displayed labels, means for retrieving a digitized image of an X-ray film associated with the selected label and means for displaying the retrieved digitized images of the X-ray films for examination by the use .