KR100921655B1 - System for processing visual images of digestive system by easiliy inputting comments on diseases and findings for each visual image - Google Patents

Abstract

PURPOSE: A system for processing visual images of a digestive system by inputting comments on diseases and findings for each visual image are provided to allow a user to get a test result in short time by checking an image of a digestive organ taken by a capsule endoscope. CONSTITUTION: A system for processing visual images of a digestive system is composed of data storage unit(72), a display unit(80), a memory(73c), input unit, and a processing unit(73). The memory unit stores images, and the display unit shows images. The names corresponding to various symptoms of a disease are stores in the storage unit in advance. An operator inputs comments of the image through the input unit.

Description

Image processing system of digestive organs with easy input of pathological findings on photographed images {SYSTEM FOR PROCESSING VISUAL IMAGES OF DIGESTIVE SYSTEM BY EASILIY INPUTTING COMMENTS ON DISEASES AND FINDINGS FOR EACH VISUAL IMAGE}

The present invention relates to a system for processing an image of the digestive organs, and more specifically, to examine the image of the digestive organs taken by a capsule endoscope or the like to minimize the unnecessary time wasted in the diagnosis process to diagnose the condition of the examinee The present invention relates to an image processing system of the digestive organs that makes it convenient and quick.

The digestive system of the human body consists of the esophagus, stomach, small intestine (ie, duodenum, jejunum, ileum) and several other organs, including the colon. To date, various endoscopy methods, such as gastroscopy or push enteroscopy colonoscopy, have been used as diagnostic tools for visually viewing the digestive organs of the human body. However, these conventional methods do not meet the requirements for a precise diagnostic test in visual examination of the small intestine.

One limited approach to addressing this problem has been to use a Wireless Capsule Endoscope (WCE) that incorporates video and video technology into wireless transmission. In particular, the wireless capsule endoscope method uses a small capsule of about 10-15mm in diameter and 20-30mm in length. As shown in Figs. 1 and 2, the capsule endoscope 50 is equipped with a light emitting diode (LED) and a micro camera to brighten the inner wall of the organ 41 by the light emitting diode, The image is taken and wirelessly transmitted to the receiver 60 worn by the examinee 40 (S11). The transmitted image data is transmitted to the processing system 1, and the expert interprets the captured image to diagnose whether there is a symptom, and the transmitted image data is stored in the recording medium of the processing system 1 (S12). ).

Then, the diagnostic expert examines the stored images one by one and captures images of suspected diseases or images having symptoms (ie, symptoms) associated with the diseases (S13). Then, as shown in FIG. 3, the diagnostic expert examines the captured images 81 displayed on the display unit 80, one by one, and suspects the image 81 in the comment field 86. Create a disease or condition (S14). At this time, the comment about the condition or disease is input as text, so it is difficult not only to look at the overall image of what symptoms in which image, but also to determine which symptoms are most serious. Therefore, the next step is to review the images on which the condition or condition is annotated, and to grasp the seriousness of how severe the disease or condition is (S15), and finally the diagnosis report on the subject. (S16).

However, as shown in FIG. 2, since the wireless capsule endoscope 50 has to repeatedly review the images and input the diagnosed symptoms or diseases in reading and interpreting the images of the organs 41, the wireless capsule endoscope 50 has to be repeatedly performed. It is tedious and time consuming, and as a result, the diagnostics do not write an annotation to avoid repetitive tasks, resulting in a large number of captured images that are skipped. Is included, there is a risk of inaccurate diagnostic results. In addition, there is no way for the diagnoser to input the degree of the condition on the captured images, and even if the weight of the condition is written in the annotation, the weight of the condition cannot be grasped at the same time through the annotation. It was accompanied by cumbersome work to look at.

Therefore, there is a need for a new method and a system for implementing the same, which can more accurately and quickly process images of the digestive organs to more effectively and conveniently diagnose the condition of a subject.

The present invention to solve the above problems, to provide an image processing system of the digestive organs to examine the image of the digestive organs taken by the capsule endoscope and the like to be able to perform the diagnosis of the subject's condition in a short time. The purpose.

That is, the present invention solves the conventional problem that it takes a long time for tedious and unnecessary work as the examiner repeatedly inputs in the process of inputting any symptoms or diseases while looking at the image taken inside the organ of the examinee, It is an object of the present invention to provide an image processing system of the digestive organs, which enables the rapid and convenient input of diagnosis of a subject's condition.

The present invention has been made to achieve the above object, it is used for diagnosing the disease in the digestive organs from a plurality of images including the first image and the second image of the digestive organ taken by the capsule endoscope A processing method comprising: an image display step of displaying photographed images; A first condition input step of inputting a condition for the first image which is one of the images by a diagnosis person; A condition input for the first image is selectably displayed for the second image displayed next to the first image, and a selected condition among the displayed symptoms is input as a condition for the second image Entering the second condition; It provides a photographing image processing method of the digestive organs, comprising.

This means that images of the digestive organs taken by the capsule endoscope are taken from one to several tens of images of one condition at a certain position of the digestive organs, so that when the condition is inputted to the first image suspected of having a disease, The second image to be displayed next is very likely to be the same condition as the first image, so that the same condition as the one entered in the first image can be selected for selection, so that the diagnosis is performed on the second image among the listed conditions. By selecting only whether the condition is a corresponding condition, it is possible to easily input a finding about the condition of the second image which is continuous with the first image. In this way, the diagnostic person does not input the symptom by typing every few to several dozen images taken for the same symptom case, and does not select any or all of the simply listed symptom lists or select any of them. Thereby, the diagnosed condition for the second image subsequent to the first image can be input simply and quickly without boring.

Here, the capsule endoscope photographing the digestive organs moves at an irregular speed by peristalsis of the digestive organs. However, in order to ensure the diagnosis accuracy by the capsule endoscope, even when the capsule endoscope moves quickly and the camera of the capsule endoscope does not face the position where the disease occurred for a while, the disease generated in the gastrointestinal tract must be photographed. The image must be secured so that accurate diagnosis can be made, so take enough images per unit time (for example, 2 to 4 times per second). Therefore, when the digestive organs are photographed by the capsule endoscope, the pathology of any one of the digestive organs may be photographed by several to several tens of images.

However, since it is cumbersome and takes a long time for a diagnosis person to make comments about repeated symptoms for several to several dozen images, the patient has made comments about the symptoms for any one or two of the images. The risk of neglecting the details of the pathology appearing in the image by repeated typing boredom, or overlooking the uncommented image of several to dozens of images, has been missed.

However, in the present invention, instead of repeatedly inputting by typing the annotation of the diagnosis, the condition previously entered for successive images is listed as a list of symptoms for the next image, and the second list of the listed symptoms is listed. By simply selecting / clearing the disease corresponding to the image with a mouse click etc., it is possible to quickly and easily input the disease to the second image. Thus, a diagnosis task that examines images without distracting attention to the input itself as in the prior art Because more attention can be focused on, the diagnosis of diseases and diseases of the digestive organs can be made more accurately and quickly. Moreover, since there are not many kinds of symptoms generally found in one subject, it is easy to input a diagnosis diagnosed for one image into the same condition for another image in order to shorten the overall diagnosis time. Very efficient.

In addition, the present invention not only selectively causes the symptoms inputted for the first image to be listed in the second image, but also the annotations inputted for the first image are also displayed on the second image, which is the same as the first image. Annotation of the content can also be entered by the examiner's choice.

On the other hand, the term "pathology" described in the present specification and claims has a dictionary meaning of "symptoms of the disease", but the term "pathology" in this specification and claims means a dictionary meaning of "symptoms of the disease" In addition, it includes the meaning of 'disease' itself, such as cancer, and also includes all the symptoms to be remembered in the diagnosis process even if it is not directly related to the disease, as well as food, foreign body, etc. Likewise, it is used as a term that includes all the meanings of 'finding' regarding general matters necessary for the diagnosis process even if it is not related to the disease.

At this time, the first path input step, if a comment that corresponds to the name of the condition is stored in advance in the state, the annotation is included in the annotation, if the annotation is included in the comment, stored in advance about the condition The diagnostically selectable display of the condition of the unified name corresponding to the noted condition name, and the condition selected by the diagnoser among the displayed symptoms is inputted as the condition for the first image. In this way, when the examiner examines the first image and writes the findings about the condition in a comment in various terms and forms, that is, even if the diagnoser describes the findings about the first image in various forms of similarity, By automatically converting and inputting a single symptom name corresponding to various types of terms written in the comment, it is possible to extract or search statistics on the diagnosed symptom based on the unified symptom name. In the process of inputting the symptom of the second image displayed subsequent to the first image, the condition inputted in the annotation in the first image is changed to a single name of the single symptom and is selectively displayed. The diagnosed condition can be entered by simply selecting a unified name for the second image. It is.

As used herein, the term 'uniform disease' or 'unified disease name' used in the present specification and claims means that the diagnosis is expressed as a single symptom, even though the terminology described in the notes related to the disease is expressed by various similar words. Refers to a condition expressed by a unified name. Thus, a condition of the same name as that described in the comment may correspond to 1: 1 or many to 1 depending on the type of condition.

On the other hand, similarly to the first disease input step of inputting the disease for the first image, in the second disease input step of inputting the disease for the second image subsequently displayed next to the first image, the first image is inputted. In addition to the conditions shown in, other conditions can be easily entered. To this end, the second condition input step may include the step of inputting a comment by a diagnosis person; If a term coinciding with the disease name previously stored in relation to the disease is included in the annotation for the second image, the disease name of the unified name corresponding to the disease name can be selectively selected. Displaying together; And selecting the selected one of the symptoms being displayed as a condition for the second image.

In other words, the first image inputted as having one or more symptoms usually has only the same condition as the condition inputted in the first image, but a new condition that does not appear in the first image (eg, For example, if there is a bleeding in the ulcer symptoms) in the second image, it is necessary to enter a new condition in the second image. To this end, as in the case of inputting an annotation on the first image, a diagnosis can be made by the diagnoser newly inputting the second image by allowing an annotation on the second image to be made while the condition entered in the first image is displayed on the list of symptoms. If a comment contains a term that matches the name of the condition previously stored in memory, a new, unified name of the corresponding name corresponding to the new name of the condition is optionally displayed, allowing further entry of the diagnosed new condition.

Similarly, in the second image subsequent to the first image, when the condition that appeared in the first image no longer appears, when the condition entered for the first image is displayed as selectable for the second image, the second image By not selecting for, it can be excluded as a condition not corresponding to the second image.

On the other hand, the present invention is a processing method used for diagnosing the disease in the digestive organs from a plurality of images including a first image of the digestive organs, comprising: an image display step of displaying the photographed images; Inputting a comment from the diagnoser on the first image being displayed; If a term is included in the annotation that matches a disease name previously stored in relation to the disease, selectively displaying a disease of a unified name corresponding to the disease name; Inputting the selected condition among the displayed symptoms as the condition for the first image; It provides a photographing image processing method of the digestive organs, comprising.

That is, as described above, when the examiner writes the findings of the condition in a comment in various terms and forms while looking at the first image, the name corresponding to the condition and its analogous words are automatically converted into the corresponding name of the corresponding condition and inputted. By doing so, even if various terms are inputted by various diagnosers, statistics on the condition can be extracted with the name of the unified condition, and the process of inputting the pathological findings on other images displayed next to the first image. In the first image, it is possible to selectively display the condition inputted in the annotation as a single name of the disease, so that the symptom of the next image similar to the first image can be selected without any comment. Allow input.

In addition, the present invention is a processing method used to diagnose whether there is a disease in the digestive organs from a plurality of images of the digestive organs, an image for displaying a plurality of images of images taken on a plurality of screens separated A display step; An image display step of displaying the photographed images; Assigning, by the diagnoser, two or more images into one group; If an annotation is input by the diagnoser for the group, inputting the same condition for all images of the group; It provides a photographing image processing method of the digestive organs, comprising.

This is because, when capturing the digestive organs using a capsule endoscope, the pathology of the same case appears in several images. Therefore, by designating them as a group, inputting the diagnosed pathology for the designated group is performed only once. This is to help the diagnosis of the diagnosis by allowing the patient to collectively input all the images belonging to the group.

In the imaging image processing methods of the digestive organs, not only the input diseases for the captured images of the digestive organs are input, but also the weight of the symptoms indicated in the images is additionally input. Therefore, after inputting all of the symptoms of the images of the digestive organs, when examining the result of the input of the images, not only what kind of symptoms were input to the subjects' photographed images, but also how serious the symptoms were. As a result, it is possible to eliminate the conventional cumbersome work of looking at the images again in order to determine the severity of the condition after inputting annotations on the images.

At this time, in the process of inputting the condition of the condition inputted to the image, if the diagnostic person brings the serious evaluation of the condition performed on the previous image to the next image as it is, the diagnoser may mistakenly If left as it is, it may cause distortion in the evaluation of the seriousness of the condition, so it is better to leave 'Undefined' as the default setting when no seriousness of the condition is entered.

And, if the same condition for the images are continuous in a range within a predetermined interval (L), the images are recorded as one condition, any one or more of the symptoms, paths entered for the images It further includes the step of statistically summarizing. Through this, even when the input of the symptoms for the image is completed, even if a plurality of images are taken for one condition, it is easy to grasp what symptoms in the digestive organ of the examinee as a whole.

In addition, when an input of a desired condition is input and searched while the input of the symptoms for the images is completed, an image corresponding to the input path among the images may be displayed. That is, since the names of the diseases inputted in the annotations in various forms are data as one unified disease name for each disease, the statistics can be calculated or searched using them.

On the other hand, the images displayed in the image display step may be images captured by the diagnosis of the image suspected as a disease. That is, in general, in diagnosing images of the digestive organs of a subject by a capsule endoscope, the processing methods may be applied for diagnosing a symptom by examining all images taken one by one, unlike in the related art, and may be associated with a symptom. The processing methods may be applied for diagnosing a condition while only looking at the images automatically extracted. have.

The term "capture" described in the present specification and claims is not limited to selecting and storing a portion of images separately in a separate memory or recording medium, and storing the portions of the images, and then searching for a portion of the images to find next. It is used with the meaning of remembering that it was selected for viewing.

On the other hand, according to another field of the invention, the present invention provides a computer readable storage medium encoded with a computer program for executing the above processing methods.

According to another aspect of the present invention, there is provided an apparatus for processing images including a first image of a digestive organ for diagnosis of a subject, the apparatus comprising: a data storage unit for storing the photographed images; A display unit for displaying the photographed images; A memory which stores names related to the diseases in advance; An input unit for inputting a comment or condition for the first image by a diagnostic person; If the term written in the annotation for the first image matches a name stored in the memory, a condition corresponding to the name is selectably displayed on the display unit from the memory, and the condition selected by the input unit is selected from the first unit. A processing unit for inputting the condition for the image; It provides a processing system of the captured image of the digestive organs, characterized in that configured to include.

The present invention provides an apparatus for processing images including a first image and a second image of a digestive organ for diagnosis of a subject, comprising: a data storage unit storing the photographed images; A display unit for displaying the photographed images; A memory which stores names related to the diseases in advance; An input unit for inputting a comment or condition for the first image by a diagnostic person; The condition inputted to the first image is displayed on the display unit selectably with respect to the second image displayed next to the first image, and the condition selected by the input unit among the symptoms being displayed on the display unit A processing unit for inputting the condition for the second image; It provides a processing system of the captured image of the digestive organs, characterized in that configured to include.

As described above, according to the present invention, the images of the digestive organs taken by the capsule endoscope are suspected of having a condition by using several to dozens of images of one condition at a certain position of the digestive organs. When a condition is input for the first image, the second image to be displayed next is very likely to be the same condition as the first image, and thus displays the same condition as that entered in the first image so as to be selectable and diagnosed. Imaging of the digestive organs, allowing the user to easily and quickly enter the findings of the symptoms of the second and subsequent images by selecting only whether the symptoms correspond to the second image among the listed symptoms. Provide an image processing system.

In addition, when an annotation is made from a diagnosis person with respect to the first image being displayed, when the term corresponding to the disease name previously stored in relation to the disease is included in the created annotation, the present invention corresponds to the disease name. By selectively displaying the symptoms of a unified name, the selected symptoms among the displayed symptoms are configured to be input into the symptoms for the first image, so that the examiner may look for the symptoms in various terms and forms. By writing in the comment, the name corresponding to the disease and its similar words can be automatically converted into the corresponding name of the corresponding disease so that it can be inputted. Statistics can be extracted and images needed after diagnosis The image processing system of the digestive organs, which makes it possible to search and examine the image, and to easily input the diagnosis result with the unified name of the disease in case of diagnosing the same condition as the previous image with respect to successive images. To provide.

In this way, the present invention, without having to input the information about the symptoms by typing in the comment column for the diagnosis of several to dozens of images taken for the same case of the case, only one note for one condition listed, By selecting any one or all of the list of symptoms or not selecting any symptoms, the result of the diagnosis of the same condition as the first image can be easily and quickly inputted to the second image following the first image and the like. It is possible to facilitate the diagnosis and to diagnose the images of the digestive organs more quickly and accurately.

Hereinafter, a first embodiment of the present invention will be described in detail with reference to the accompanying drawings.

However, in describing the present invention, the same or similar reference numerals are given to known functions or configurations, and detailed description thereof will be omitted for clarity.

As shown in FIG. 4, the method S100 of processing an image of a digestive organ according to a first embodiment of the present invention includes an image capturing step S110 of capturing the inside of an organ of a subject from a capsule endoscope 50; An image storage step (S120) of receiving photographing data from the capsule endoscope 50 and storing the images in a data storage unit 72 such as a hard disk, and an image capturing step of capturing images suspected of being sick while looking at the stored images. (S130), a diagnosis input step (S140-S180) of inputting the symptoms of the images while looking at the captured images, and the diagnosis of the symptoms inputted in the symptoms input step (S140-S180) or the diagnosis and display to the diagnosis It consists of the results verification step (S190-S199).

In the image capturing step S110, when the examinee 40 swallows the capsule endoscope 50, the capsule endoscope 50 moves inside the organ by the peristaltic movement of the examinee's organ 41, and the capsule endoscope 50 is moved to the capsule endoscope 50. The inside of the organ revealed by the mounted light emitting diode is photographed by a camera mounted on the capsule endoscope 50.

The image storing step (S120) transmits the image data photographed inside the organ 41 by the capsule endoscope 50 to the receiver 60 worn by the subject 40 through a human body or by a radio frequency method. The image data received by the receiver 60 is transmitted back to the processing system 1, and received by the photographing data receiving unit 71 of the main body 70 of the processing system 1, so that a hard disk or This is done by storing in a data storage unit 72 such as an optical disc.

The image capturing step S130 is performed by storing and capturing all images showing abnormal signs while looking at the images stored in the data storage unit 72 one by one or several at the same time. Thus, the image capture step looks at many images, but it takes relatively short time to look at the unit image because it is very short. Although the first embodiment of the present invention has been described that the image capture step (S130) is included as an example, it is not necessary to go through the image capture step (S130) of inferring images with abnormal signs of a number of images, the data storage unit ( While looking at the images stored in the 72) it may proceed to the step (S140-190) of directly diagnosing and inputting the symptoms of the image.

The disease input step (S140-S180), a step (S140) for writing the diagnosis information on the display of the first image (115a) being displayed in the annotation column 120, and previously stored in the annotation created in the annotation column 120 If a term coinciding with the name of the disease is included, a step of calling up the unified disease name corresponding to the term from the memory 73c and automatically displaying it on the disease list 130 (S150); Selectively inputting the diagnosed condition of the first image 115a by selecting or not selecting 'V' in the check box 132 corresponding to each condition 131, and the list of symptoms 130. Step S170 of inputting the weight of the disease with respect to the selected disease 131 of FIG. 2) and the second image 115b displayed after the input of the disease for the first image 115a is finished. The symptoms memorized in one image 115a are listed in the symptom list 130. In step S180, the diagnosed patient inputs the diagnosed symptom for the second image 115b by selecting or not selecting 'V' on the check box 132 of the symptom list 130. .

In order to implement such a disease input step (S140-S180), in the first embodiment of the present invention, the image display control unit 73a of the processing unit 73 embedded in the main body 70 of the processing system 1 is provided. By using the screen displayed on the display unit 80 as shown in FIG. That is, an image 115a to be diagnosed is displayed on a large screen on the image display window 110, and a plurality of images 115z, 115b..., Adjacent to the image 115a are thumbnailed around the image 115a to be diagnosed. Listed in a manner 110 is displayed. In this case, the number of screens displayed in a thumbnail manner may be displayed by clicking the button indicated by 112 as shown in FIG. 5, or by displaying the 15 images by clicking the button indicated by 113. It may be displayed. In addition, although not shown in the drawings, the number of images displayed in a thumbnail manner is variously adjustable. In addition, the images diagnosed and inputted are displayed on the location image bar 141 of the status display unit 140, and the location 142 is displayed in response thereto. Together with the image 143 and the corresponding disease 144 is displayed in association with the disease image position display bar (141). Reference numeral 141a in the drawing indicates the time required for slowly playing back the entire images.

In order to diagnose and input a symptom for a continuous first image and a second image among a plurality of images captured in step S130 by using the displayed screen 80 as shown in FIG. As described above, the diagnoser writes the symptom in the annotation column 120 for the diagnosis of the symptom of the first image 115a being displayed (S140).

At this time, if the comment made by the diagnoser in the comment column 120 includes a term corresponding to the name of the condition previously stored, the unified condition name corresponding to the term is read from the memory 73c, and the list of symptoms 130 is read. The name of the unified disease is automatically displayed (S150).

That is, in FIG. 7, the comment inputted by the diagnoser to the comment column 120 includes a condition 121 called polyp, and the comment column as shown in FIG. 13 is included in the memory 73c of the processing system 1. Since the name coinciding with the polyp associated with the condition created in 120 is included, the condition 131 called polyp is displayed on the condition list 130 from the name list 77 of the memory 73c.

Also, as shown in Fig. 8, the diagnoser may make another annotation on the first image 115a, in which the term written in the annotation coincides with another term in the disease name list 77 stored in the memory 73c. Then, another matched disease name is further displayed in the disease list 130. In this case, the same symptom may be described in another similar expression for each diagnoser. For example, as shown in FIG. 13, when bleeding occurs, blood, bleed, bleeds, bleeding, or the like can be written in various ways, so that all of these similar terms are stored in the disease name list 77 of the memory 73c. By placing them on the symptom list 130 with a single name for these similar terms (e.g., "bleeding" for bleeding), thereby annotating the patient with different expressions for the same symptom. Even if we write, we can statistically and database one unified disease name. That is, as shown in FIG. 8, if the diagnoser writes an annotation including terms of erosion and hyperemic matching the names stored in the disease name list 77 of the memory 73c, The erosion and bleeding, which are the unified disease names corresponding to these terms, are called by the name calling unit 73d and loaded into the disease list 130. For this purpose, as shown in Fig. 13, the memory 73c stores a disease name list 77 in which similar names of various diseases are stored, and a unified disease name list 99 corresponding to these disease names.

After all comments on the condition for the first image 115a being displayed are written, the condition list 130 lists the conditions 131 loaded from the memory 73c. Among these, as shown in FIG. 9, the diagnosis item that the diagnostician wants to input into the first image 115a is placed on the check box 132 by clicking the mouse cursor 92a to check "V". If the symptoms listed in 130 are inadequate as the symptoms of the first image 115a, the "V" is not checked in the check box 132 of the corresponding disease, and thus, "V" among the symptoms displayed on the disease list 130. Only the selected diseased checker is input to the disease input storage unit 73b as the disease of the first image 115a (S160).

On the contrary, the check box 132 selects the "V" check mark as the default setting, and clicks the mouse cursor 92a to release the "V" check to select the symptoms of the disease list 130. You can also choose to input. In general, since the symptoms tend to appear in a plurality of consecutive images, this method has an advantage of minimizing the input of diagnosed symptoms for the images.

And, as shown in Figure 9, for each of the symptoms 131 diagnosed and inputted by the diagnoser as shown in the first image 115a, the diagnoser has additional information such as the degree or shape of each of the symptoms 131. Is input by dragging the scroll bar 134a 'with the mouse 92 to input 134a or by clicking the button form 134b 134b' (S170).

At this time, before the diagnoser inputs additional information such as the seriousness of the disease 131, each information is basically displayed as “undefined”. In this way, in the process of inputting the pathology 131 for the pathology 131 inputted to the first image, the diagnosis of the pathology performed by the diagnoser on the previous image 115z is carried on the next image as it is. If the diagnosis is made by mistake and the patient forgets to check the seriousness of the condition, it may cause distortion in the serious evaluation of the condition. Therefore, it is better to leave it as 'not input' when the condition is not entered.

As described above, unlike the diagnosis by the diagnosis person, the additional information necessary for the entry of each condition 131 is displayed in a separate window 134 so that the diagnosis can be quantitatively inputted, thereby providing information about the seriousness of the condition. It can be a database that can be easily organized by searching or searching statistically, and it is possible to identify where and how serious the disease is later without having to look at each captured image again.

Then, the memory button 133 is stored by clicking on the memory button 133 for a symptom for which there is a condition that the second image 115b consecutive to the first image 115a is expected to overlap with the first image 115a. Do it. At this time, the shape of the memory button 133 changes as indicated by reference numeral 133 '. Accordingly, as shown in FIG. 10, when the second image 115b subsequent to the first image 115a is displayed in the image display window 110, the second image 115b is displayed in the image display window 110. Erosion and bleeding, which are stored as the list of symptoms 130 of), are displayed.

Therefore, if the diagnostician wants to input only erosion and bleeding, which are the symptoms diagnosed in the first image 115a with respect to the second image 115b, the symptoms displayed in the list of symptoms 130 of FIG. By simply clicking the check box 132 of the field 131 and displaying a "V", inputting the diagnosed condition for the second image 115b is simply completed (S180).

In general, since one condition tends to be repeated in several to several tens of images continuously, the memory button 133 is a diagnosis screen of the second image 115b as shown in the previous first image 115a. 10), when the diagnosis of the image after the second image 115b is no longer desired to display the current condition (i.e., erosion and bleeding) in the disease list 130, click the memory button 133 again. To release the memory state.

After inputting the diagnosis result of the first image 115a, the second image 115b is to be diagnosed, but no condition is input to the previous image 115z of the first image 115a. A, if a condition (ie, polyps, erosion, bleeding) is input in the first image 115a, the condition image of the condition display unit 140 at the moment of switching to the diagnosis screen (FIG. 10) of the second image 115b. The location 142 of the first image 115a diagnosed as having a condition is displayed on the location bar 141, and the condition of the erosion of the first image 115a is the most severe. ) And its image 115a are newly displayed 143 "and 144" in association with the disease image location bar 141 as the representative image 143. Then, when the diagnosis result of the second image 115b is worse than the first image, the pathology of the second image 115b is immediately followed by completion of the diagnosis diagnosis input to the second image 115b. The representative image 143 appearing in the image position bar 141 is replaced with the second image 115b.

As such, in the process of diagnosing the condition by examining the captured images one by one, the database is databased with one unified name of the related condition regardless of the terms used in the comment of the diagnosis person, and at the same time, the weight of the condition is quantified. It is usually databased, and the diagnostic practitioner only needs to comment on the newly appearing images, and check boxes for each of the symptoms 131 shown in the list of symptoms 130 for similar or identical images. Just click 132 to enter simply and quickly.

On the other hand, the reference numeral 143 'in the drawing indicates that although the condition is input by the diagnosis person, the image placed as "not input" for the seriousness of the condition is displayed in association with the condition position display bar 141.

After repeating the above process and inputting the symptoms for all the captured images, a statistics button 114 written as " statistics " is generated at the bottom right of the condition display 140 of FIG. Then, when the diagnoser clicks the statistics button 114, as shown in FIG. 15, the statistical result of the input condition on the captured images is displayed on the statistical display window 150. FIG.

More specifically, when a plurality of images are diagnosed as one condition in succession, such as the first image 115a and the second image 115b, or may be smaller than the predetermined interval L, the same condition may occur. If diagnosed, it is regarded as one condition, and these images are classified into one group (S190).

This is because, as described above, one disease case is common to several to several tens of images, so seeing them as one fits the reality. And, if the distance is smaller than the predetermined interval, since the capsule endoscope 50 may take a different direction for a short time by the peristalsis of the digestive organs, even if the images showing the same symptoms are not separated from each other and slightly apart, in fact one To be considered. The larger the number of images grouped into one group, the wider the area in which the condition appears, and thus, the weight is given in proportion to the number of images.

When the symptom data is summarized and displayed on the statistical display window 150 of FIG. 15, the examinee receiving the diagnosis result as shown in FIG. 15 generally has an ulcer (Ulcer) corresponding to the weight 3 at the central position of the digestive organ. In addition, it can be seen at a glance that bleeding (Bleeding) occurred at a position corresponding to approximately 4/5 of the digestive organs by weight 4 (S199).

When using the processing method (S100) according to the first embodiment of the present invention configured as described above, the diagnoser inputs the symptom by typing in the comment column for the several to several dozen images taken for the same case. By not selecting any one or all or none of the listed symptom list, it is possible to enter the symptom input for a plurality of images simply and quickly without boredom, for the convenience of the diagnosers. can do.

Hereinafter, a second embodiment of the present invention will be described in detail with reference to the accompanying drawings.

However, in describing the second embodiment of the present invention, the same or similar reference numerals are assigned to the same or similar components as those of the above-described first embodiment, and the detailed description thereof will be directed to the gist of the second embodiment. It will be omitted for clarity.

As shown in FIG. 14, the method of processing an image of the digestive organs according to the second embodiment of the present invention includes, in addition to or in place of the image processing method S100 according to the first embodiment, a plurality of images. It is configured to be able to input the condition at a time while all of them have been designated in advance. That is, as shown in FIG. 14, a plurality of thumbnail-type screens displayed on the image display window 110 around the first image 115a to be diagnosed currently are dragged to the mouse cursor 92 or the keyboard 92. After designating the block as a group by specifying a block with a specific key of), an annotation for the designated group is input to the comment column 120, and the diagnosis of all images of this group is performed as in steps S150 to S180 of FIG. You can enter all at once.

Although the preferred embodiments of the present invention have been described above by way of example, the scope of the present invention is not limited to these specific embodiments, and may be appropriately changed within the scope described in the claims.

That is, in the embodiment of the present invention, the medically unrelated diseases were described as being shown to one subject, and the "hyperemia" symptom was described as being able to be expressed by a unified disease name "bleeding". Although medically inadequate, the present invention has been exemplified in order to more clearly describe the configuration of the processing method and the system according to the present invention, and the scope of the present invention is not limited to the exemplary embodiments described.

Figure 1 is a schematic diagram showing the configuration of obtaining an image of the digestive tract using a capsule endoscope

FIG. 2 is a flow chart sequentially illustrating a conventional processing method for diagnosing a disease or a disease of a subject from an image of a digestive organ

3 shows a display screen executing the method of FIG.

4 is a flowchart sequentially illustrating a processing method of a captured image of a digestive organ according to an embodiment of the present invention;

5 to 6 are diagrams illustrating a screen for inputting an annotation for a first image by the processing method of FIG. 4.

7 and 8 illustrate a screen on which a list of symptoms of a first image is displayed by inputting an annotation on the screen shown in FIG.

FIG. 9 is a view showing a screen for selecting a condition to be input for the first image in the list of symptoms of FIG. 8 and displaying the weight of the condition input for the first image.

FIG. 10 illustrates a screen on which a list of symptoms for a second image following a first image is displayed by the processing method of FIG. 4; FIG.

FIG. 11 is a diagram illustrating a screen for selecting a condition to be input for the first image in the list of symptoms of FIG. 10 and displaying the weight of the condition input for the first image;

12 is a block diagram showing the configuration of a processing system for implementing the processing method of FIG.

FIG. 13 is a diagram showing a configuration for calling a disease related name stored in a memory of the processing unit of FIG. 12 into a disease list. FIG.

14 is a diagram showing a screen implementing the processing method of the captured image of the digestive organs according to another embodiment of the present invention;

Figure 15 is a diagram showing a screen showing the statistics of the distribution of symptoms diagnosed throughout the digestive organs after diagnosing the symptoms of the digestive organs of the examinee according to the present invention.

*** Explanation of main parts of drawing ***

1: Processing System 50: Capsule Endoscope

60: receiver 71: recording data receiver

72: data storage unit 73: processing unit

77: List of disease names 99: List of disease names

80: display unit 90: diagnostic input unit

110: image display window 120: annotation field

130: disease list display window 140: disease status display unit

Claims (14)

A digestive organ imaging image processing system used for diagnosing a disease in the digestive organs from a plurality of images including a first image and a second image of the digestive organ taken by a capsule endoscope. A data storage unit for storing the images; A display unit for displaying the images; A memory for storing unified disease names corresponding to various names associated with the disease in advance; An input unit for inputting annotations on the images by the diagnoser; If a comment inputted to the first image includes terms that match the various names stored in the memory, a unified disease name corresponding to the name is selectable from the memory and displayed on the display unit, and the displayed unification If a diseased disease name is selected for the first image, the unified disease name selected for the first image is automatically selectable for the second image, even for the second image that is displayed after the first image. A processing unit for displaying; Processing system of the captured image of the digestive organs, characterized in that configured to include. A digestive organ imaging image processing system used for diagnosing a disease in the digestive organs from a plurality of images including a first image and a second image of the digestive organ taken by a capsule endoscope. A data storage unit for storing the images; A display unit for displaying two or more images simultaneously; An input device for assigning two or more images to a group by a diagnostic person and inputting an annotation for the group; A processing unit which, when an annotation for the group is input by the input device, inputs the same condition input for all images of the group; Processing system of the captured image of the digestive organs, characterized in that configured to include. The method according to claim 1 or 2, And the processing unit is configured to further input a degree of the condition relative to the condition input for the image. The method of claim 3, wherein And the processing unit defaults to 'undefined' in a state in which a degree of the disease is not input to the disease input to the image. The method according to claim 1 or 2, And the processing unit processes the images as one condition if the same condition for the images continues in a range within a predetermined interval. The method according to claim 1 or 2, When the processing unit inputs and searches for a desired condition while the input for the symptoms for the images is completed, the processing unit displays an image corresponding to the entered condition among the images on the display unit. Processing system of taken images of the organ. delete delete delete delete delete delete delete delete

Images