JP2009131319A - Image processing apparatus and image processing system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve such a problem that work of visually understanding an organ region where a capsule type endoscope is located by images inside a body imposes a burden on an observer. <P>SOLUTION: An image processing system 100 relating to an embodiment is equipped with the capsule type endoscope 110 for capturing the images inside the body while moving the inside of the body of a subject to be examined and a receiving device 120 for receiving signals transmitted by the capsule type endoscope 110. The receiving device 120 is equipped with: an organ determining part 123 which determines the organ region where the capsule type endoscope 110 is positioned when capturing the image inside the body from the amount of characteristics of the image inside the body captured by the capsule type endoscope 110; and an informing part 124 for informing the observer of the determination information based on a result of the determination by the organ determining part 123. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a technique for processing an in-vivo image in which a body is imaged, and more particularly to a technique for processing an in-vivo image captured by a capsule endoscope.

  In the field of endoscopes, swallowable capsule endoscopes have been proposed and put into practical use. After being swallowed from the subject's mouth for observation (examination), the capsule endoscope moves along the body according to its peristaltic movement along with other contents until it is naturally discharged from the human body. Take in-vivo images. The in-vivo image data is sequentially transmitted to the outside of the body by wireless communication and stored in the memory of the receiving device carried by the subject. Observers such as doctors and nurses, after the imaging operation of the capsule endoscope is completed, transfer a series of in-vivo image data stored in the memory to a workstation or the like and subsequently view the in-vivo images. On the other hand, in recent years, a simple display device capable of browsing an in-vivo image captured by a capsule endoscope in real time has been developed. For example, Patent Document 1 discloses a simple display device that displays in-vivo images in real time by receiving in-vivo image data transmitted from a receiving device or a capsule endoscope.

JP 2006-293237 A

  By the way, in general, since a capsule endoscope does not originally have a self-running function, for example, when a functional abnormality such as a decrease in gastric emptying due to diabetes occurs, a capsule endoscope is used at a specific organ site. May stagnate. For example, if the capsule endoscope is stagnant in the stomach, the battery will run out before the capsule endoscope reaches the small or large intestine, and it becomes impossible to image these organ parts. appear. In this case, the observer needs to grasp the movement state of the capsule endoscope and take measures such as administration of a gastrointestinal motility promoting agent as soon as possible. However, when the capsule endoscope is moving in the body, the movement position of the capsule endoscope, for example, the work of visually grasping the organ part where the capsule endoscope is located by the in-vivo image, This is a heavy burden for the observer.

  The present invention has been made in view of the above, and when the capsule endoscope is moving in the body, the burden on the observer who observes the in-vivo image captured by the capsule endoscope is reduced. A feature of the present invention is to provide an image processing technique that can be reduced.

  In order to solve the above-described problems and achieve the object, an image processing system according to an aspect of the present invention includes a capsule endoscope apparatus that captures an in-vivo image while moving in a body to be examined, and the capsule-type interior. An image processing system including a receiving device that receives a signal transmitted from an endoscope device, the body based on a feature amount indicating a feature of an in-vivo image obtained by the capsule endoscope device An organ determination unit that determines an organ site where the capsule endoscope device is located when the image is captured, and a determination by the organ determination unit when the capsule endoscope device is moving inside the body. And a notification unit for notifying determination information based on the result. Here, the in-vivo image is an image showing an organ having a unique function in the body, such as the stomach, the small intestine, the esophagus, and the blood vessel.

  According to the image processing system according to this aspect, determination information based on the determination result by the organ determination unit, for example, information indicating an organ site where the capsule endoscope exists, is transferred to the body by the capsule endoscope. When an observer observes an in-vivo image, it is no longer necessary to visually grasp the organ site where the capsule endoscope device is located. As a result, the capsule endoscope moves in the body. This can reduce the burden on the observer when doing so.

  An image processing system according to another aspect of the present invention includes a capsule endoscope apparatus that captures an in-vivo image while moving in a body to be examined, and a receiving apparatus that receives a signal transmitted by the capsule endoscope apparatus. And a notification unit for notifying position information indicating the position of the capsule endoscope apparatus when the capsule endoscope apparatus is moving inside the body.

  An image processing apparatus according to another aspect of the present invention is configured to determine an organ site where a capsule endoscope apparatus is located when an in-vivo image is captured based on a feature amount indicating a feature of the in-vivo image in which the inside of the body is captured. An organ determination unit for determination and a notification unit for notifying determination information based on a determination result by the organ determination unit when the capsule endoscope apparatus is moving in the body.

  ADVANTAGE OF THE INVENTION According to this invention, the image processing technique which can reduce the observer's burden when the capsule endoscope is moving in the body can be provided.

  Hereinafter, an image processing system which is the best mode for carrying out the present invention will be described with reference to the accompanying drawings. In addition, this invention is not limited by each embodiment. In the description of the drawings, the same or corresponding parts are denoted by the same reference numerals.

(Embodiment 1)
FIG. 1 is a block diagram showing a schematic configuration of an image processing system 100 according to the first embodiment of the present invention. The image processing system 100 includes a capsule endoscope 110 that captures and transmits an in-vivo image while moving in the subject's body, and a receiving device that receives, stores, and stores the in-vivo image transmitted by the capsule endoscope 110. 120. During imaging of the in-vivo image by the capsule endoscope 110, the receiving device 120 is always carried by the subject and sequentially receives the in-vivo image.

  The capsule endoscope 100 is a capsule-type device that can be introduced into the body, and includes an imaging unit 111 that images the inside of the body, a transmission unit 112 that transmits the in-vivo image to the receiving device 120, an in-vivo image, and a capsule. A storage unit 113 that stores a program that executes each process performed by the endoscope 100 and a control unit 114 that controls each unit are provided. The imaging unit 111 is realized by a CCD camera, an LED, and the like, and takes an in-vivo image under a control of the control unit 114 at a constant imaging rate, for example, 2 to 4 frames / sec. The transmission unit 112 is realized by an RF antenna or the like, and transmits the in-vivo image to the reception device 120.

  The receiving apparatus 120 includes a receiving unit 121 that receives in-vivo images, a storage unit 122 that stores programs for executing various processes, and an organ site where the capsule endoscope 110 is located when each in-vivo image is captured. An organ determination unit 123 to be determined, a notification unit 124 that notifies information (hereinafter referred to as “determination information”) based on a determination result by the organ determination unit 123, and a control unit 128 that controls each unit are provided. The receiving unit 121 receives the in-vivo image transmitted by the transmitting unit 112 via a plurality of receiving antennas (not shown) arranged on the body surface of the subject. The storage unit 122 stores various types of information and includes an image storage unit 122a to store in-vivo images. The image storage unit 122a is realized by a portable storage medium such as a compact flash (registered trademark), and is used to transfer an in-vivo image between the receiving device 120 and a workstation (not shown) after the in-vivo image is captured. . Note that the control unit 128 performs compression processing such as DCT encoding represented by JPEG on the received in-vivo image, and then stores it in the image storage unit 122a. Based on the file size of the in-vivo image after compression, the organ determination unit 123 captures the organ part captured in each in-vivo image, that is, the organ part where the capsule endoscope 110 is located when each in-vivo image is captured. Are sequentially determined.

  The notification unit 124 includes a notification control unit 125 that controls the notification unit 124, a display unit 126 that displays determination information, and a transmission unit 127 that transmits the determination information to the communication apparatus 1 outside the image processing system 100. The notification control unit 125 includes a memory 125a that stores determination information and the like. The communication device 1 is realized by a mobile terminal outside the image processing system 100, such as a mobile phone, and includes a reception unit 2 that receives the determination information transmitted by the transmission unit 127 and a display unit 3 that displays the determination information.

  Next, the determination information notification process performed by the image processing system 100 will be described with reference to FIG. FIG. 2 is a flowchart illustrating a processing procedure when the image processing system 100 notifies the determination information. First, the control unit 114 controls the imaging unit 111 to capture an in-vivo image (step S101). Thereafter, the control unit 114 transmits the in-vivo image outside the body by the transmission unit 112 (step S102). On the other hand, the control unit 128 receives the in-vivo image via the receiving unit 121 (step S103). The control unit 128 controls the organ determination unit 123 to determine the organ site where the capsule endoscope 110 is located when the in-vivo image is captured (step S104). Thereafter, the control unit 128 controls the notification unit 124 to notify the determination information (step S105). The control unit 128 determines whether or not the number of in-vivo images that have undergone the processing of steps S103 to S105 has reached a preset scheduled number of images to be captured (step S106). When the number of in-vivo images that have been processed exceeds the scheduled number of images to be captured, the control unit 128 ends the process of notifying the determination information (step S106: Yes). On the other hand, when the number of in-vivo images that have been processed is equal to or less than the scheduled number, the control units 114 and 128 repeat the processing of steps S101 to S105 (step S106: No). In addition, the control parts 114 and 128 are good also as repeating the process of step S101-S105 until the capsule endoscope 110 is discharged | emitted out of the body. As described above, the image processing system 100 processes the in-vivo image in real time by the receiving device 120, determines the organ part through which the capsule endoscope 110 is passing, and notifies information based on the determination result.

  Here, the processing performed by the organ determination unit 123 under the control of the control unit 128 in step S104 will be described with reference to FIGS. 3, 4, and 5. FIG. The organ determination unit 123 takes advantage of the fact that the file size of the in-vivo image after DCT encoding and the amount of change in the file size differ from organ to organ, when the organ part imaged in the in-vivo image, that is, when capturing the in-vivo image The organ site where the capsule endoscope 110 is located is determined.

  The reason why the file size of the in-vivo image after DCT encoding differs depending on the imaged organ site is that the shape of the mucosal surface of each organ differs. For example, the mucosal surface of the stomach has fewer irregularities than the small intestine with villi. For this reason, the in-vivo image obtained by imaging the stomach has a higher correlation between each pixel and its surrounding pixels in the image region than the in-vivo image obtained by imaging the small intestine. In addition, when the image is converted to frequency space, the frequency component of the in-vivo image obtained by imaging the stomach is biased toward low frequencies and the amount of information is small, and the frequency component of the in-vivo image obtained by imaging the small intestine is distributed from low to high frequencies. The amount of information increases. Therefore, as shown in FIG. 3, the file size after DCT encoding of the in-vivo image obtained by imaging the esophagus or stomach is smaller than the file size after DCT encoding of the in-vivo image obtained by imaging the small intestine. Further, since the mucosal surface of the large intestine is less uneven than the mucosal surface of the small intestine, as shown in FIG. 3, the file size after DCT encoding of the in-vivo image obtained by imaging the large intestine is the same as that of the in-vivo image obtained by imaging the small intestine. It becomes smaller than the file size after DCT encoding.

  In addition, the amount of change in the file size of the in-vivo image after DCT encoding differs depending on the imaged organ site because the moving speed of the capsule endoscope 110 in each organ differs. In the large intestine, the capsule endoscope 110 is stagnated because the contents are clogged. Therefore, when the in-vivo images are compared in the time series of imaging, the change between the images is small. On the other hand, since the capsule endoscope 110 moves more smoothly in the small intestine than in the large intestine, the change between images is large. Therefore, as shown in FIG. 4, the change amount of the file size after DCT encoding of the in-vivo image obtained by imaging the large intestine is small, and the change amount of the file size after DCT encoding of the in-vivo image obtained by imaging the small intestine is large.

  First, the organ determination unit 123 is based on the file size after DCT encoding, and the organ site where the capsule endoscope 110 is located when each in-vivo image is captured is the esophagus or stomach or the small intestine. Determine whether. Further, the organ determination unit 123 performs post-DCT encoding on in-vivo images that are captured in time series later than the in-vivo image in which it is determined that the organ site where the capsule endoscope 110 is located is the small intestine. Based on the change amount of the file size, it is determined whether the organ part where the capsule endoscope 110 is located is the small intestine or the large intestine.

  FIG. 5 is a flowchart illustrating the procedure of the organ determination process performed by the organ determination unit 123 in step S104. As shown in FIG. 5, the organ determination unit 123 calculates a file size after DCT encoding of the received in-vivo image (step S201). The control unit 128 stores the file size calculated by the organ determination unit 123 in the storage unit 122 (step S202). The organ determination unit 123 determines whether or not the received in-vivo images have reached a predetermined number (step S203). When the number of in-vivo images reaches the predetermined number (step S203: Yes), the organ determination unit 123 reads the file size of the predetermined number of in-vivo images and changes the file size between the in-vivo images before and after that are adjacent in time series. The amount is calculated (step S204). Thereafter, the organ determination unit 123 determines the organ site where the capsule endoscope 110 is located when each in-vivo image is captured based on the file size of the in-vivo image and the amount of change in the file size, and ends the processing. (Step S205). If the in-vivo images have not reached the predetermined number (step S203: No), the process ends without performing the processes of steps S204 and 205. As described above, the organ determination unit 123 determines the organ site where the capsule endoscope 110 is located when the in-vivo image is captured, for each predetermined number.

In step S204, the organ determination unit 123 first determines whether the file size F _size of the in-vivo image is smaller than a preset threshold value T _Fsize as shown in the following equation (1). The threshold value T _Fsize is a threshold value for determining whether the organ part imaged in the in-vivo image is “esophagus or stomach” or “small intestine”, and is based on the sex, age, past history, etc. of the subject. It is set in advance.
F _size <T _Fsize (1)

When the file size F _size is smaller than the threshold value T _Fsize , the organ determination unit 123 determines that the organ part where the capsule endoscope 110 is located is the esophagus or the stomach, and when the file size F _size is larger than the threshold value T _Fsize. The organ site where the capsule endoscope 110 is located is determined to be the small intestine.

Further, the organ determination unit 123 corresponds to each in-vivo image as shown in the following formula (2) for the in-vivo image captured after it is determined that the organ site where the capsule endoscope 110 is located is the small intestine. It is determined whether or not the attached file size change amount F _sizeDiff is smaller than the threshold value T _FsizeDiff . The threshold value T _FsizeDiff is a threshold value for determining whether the organ part imaged in the in-vivo image is “small intestine” or “large intestine”, and is set in advance based on the sex, age, medical history, etc. of the subject.
F _sizeDiff <T _FsizeDiff (2)

When the file size change amount F _sizeDiff is smaller than the threshold value T _FsizeDiff , the organ determination unit 123 determines that the organ site where the capsule endoscope 110 is located is the large intestine, and the file size change amount F _sizeDiff is the threshold value T. _When larger than _FsizeDiff , it is determined that the organ part where the capsule endoscope 110 is located is the small intestine.

  FIG. 3 is a diagram showing the transition of the change in the file size with respect to the imaging time. The file size shown in the figure is an average file size calculated for each predetermined number of time series. FIG. 4 is a diagram showing the change of the change amount of the file size with respect to the imaging time. The change amount of the file size shown in the figure is an average file calculated for each predetermined number of time series. The amount of change in size. When using the formulas (1) and (2), the organ determination unit 123 replaces the file size of each in-vivo image and calculates the average calculated for each predetermined number of time series as shown in FIGS. Using the file size and the amount of change thereof, the organ site where the capsule endoscope 110 is located when the in-vivo image is captured may be determined for each predetermined number of images. Note that the organ determination unit 123 may perform the organ determination processing using either the file size of the in-vivo image or the change amount of the file size.

  Next, processing performed by the notification unit 124 under the control of the control unit 128 in step S105 will be described. The notification unit 124 includes organ part information indicating an organ part where the capsule endoscope 110 is currently located, that is, an organ part determined by the organ determination unit 123, and the capsule endoscope 110 has reached a specific organ. Later time information, which is information indicating the time during which the specific organ is continuously imaged, excess information, which is information indicating that the time during which the specific organ is continuously imaged exceeds the threshold, and a capsule endoscope Four types of information including arrival information, which is information indicating that 110 has reached a specific organ, are notified as determination information. Here, the organ part information is information for specifying an organ part, for example, the name or identification code of the organ part. This specific organ may be designated by the observer or may be designated in advance in the image processing system. In this case, the image processing system is further provided with a reception unit (not shown) and a coincidence determination unit (not shown). The accepting unit accepts designation of an organ part from an observer as the specific organ, for example. The coincidence determination unit is provided in the notification control unit 125, and determines whether or not the organ part specified by the reception unit matches the organ part determined by the organ part part. The notification unit 124 notifies arrival information, which is information indicating that the capsule endoscope apparatus has reached the designated organ site, when it is determined by the match determination unit that they match. In that case, of course, the notification unit 124 may also notify other three types of information. In any case, the notification unit 124 performs four processes in parallel in the process of step 105 and notifies four types of information. In the first embodiment, the determination information is displayed on the display unit 126, and the determination information is notified by transmitting the determination information to the communication device 1 carried by the observer via the transmission unit 127.

  FIG. 6 is a flowchart illustrating a procedure of processing for notifying organ part information. As illustrated in FIG. 6, the notification control unit 125 controls the display unit 126 and the transmission unit 127 to notify the organ part information based on the latest determination result by the organ determination unit 123 and ends the process. (Step S311).

  The observer and subject carrying the communication device 1 can grasp the organ part through which the capsule endoscope 110 is currently passing by receiving notification of the organ part information. Therefore, for example, when the same organ site is imaged for a long time, the observer determines that the capsule endoscope 110 is stagnant, calls back the subject, and moves the capsule endoscope 110. It can be carried out. Further, the subject can visit the observer himself and inform the observer that the capsule endoscope 110 continues to image the same organ part for a long time.

FIG. 7 is a flowchart showing a processing procedure for notifying post-arrival time information. As illustrated in FIG. 7, the notification control unit 125 determines whether the organ part where the capsule endoscope 110 is located is a specific organ based on the latest determination result by the organ determination unit 123. (Step S321). When the organ site where the capsule endoscope 110 is located is a specific organ (step S321: Yes), the notification control unit 125 determines whether or not the capsule endoscope 110 has imaged the specific organ for the first time (step S321: Yes). Step S322). When the specific organ is imaged for the first time (step S322: Yes), the notification control unit 125 sets the time when the capsule endoscope 110 first images the specific organ, and the time T when the capsule endoscope 110 reaches the specific organ. _S is stored in the memory 125a (step S323). When the specific organ is being imaged (step S322: No), the notification control unit 125 calculates the time T _S when the capsule endoscope 110 reaches the specific organ from the time T _N when the latest in-vivo image is captured. The time T during which the capsule endoscope 110 continues to image the specific organ is calculated by subtraction (step S324). Thereafter, the notification control unit 125 controls the display unit 126 and the transmission unit 127 to notify the time T during which the specific organ is continuously imaged, and ends the process (step S325). When the organ part where the capsule endoscope 110 is located is not a specific organ (step S321: No), the notification control unit 125 ends the process.

  The observer and the subject carrying the communication device 1 can easily grasp the time T during which the capsule endoscope 110 continues to image the specific organ by receiving notification of time information after arrival. For example, when a stomach in which the capsule endoscope 110 is likely to stagnate is set as a specific organ, an observer can easily determine whether or not the capsule endoscope 110 is stagnant in the stomach.

FIG. 8 is a flowchart illustrating a procedure of processing for notifying excess information. As illustrated in FIG. 8, the notification control unit 125 determines whether the organ part where the capsule endoscope 110 is located is a specific organ based on the latest determination result by the organ determination unit 123. (Step S331). When the organ site where the capsule endoscope 110 is located is a specific organ (step S331: Yes), the notification control unit 125 determines whether or not the capsule endoscope 110 has imaged the specific organ for the first time ( Step S332). When an image for the first time a specific organ (Step S332: Yes), the notification control unit 125 reads the subject thresholds _{T A} and organ threshold _{T B,} by multiplying the subject thresholds _{T A} and organ threshold _{T B,} the capsule endoscope A threshold value T _MAX that is a measure of the time required for the mirror 110 to normally pass through the specific organ is calculated (step S333). Then, the notification control unit 125, the capsule endoscope 110 stores the time _{T S} has reached the specific organs in the memory 125a (step S334). When the specific organ is being imaged (step S332: No), the notification control unit 125 calculates the time T _S when the capsule endoscope 110 reaches the specific organ from the time T _N when the latest in-vivo image is captured. The time T during which the capsule endoscope 110 continues to image the specific organ is calculated by subtraction (step S335). Thereafter, the notification control unit 125 determines whether or not the time T during which the specific organ is continuously imaged is greater than or equal to the threshold value T _MAX (step S336). When the time T during which the specific organ continues to be imaged is equal to or greater than the threshold value T _MAX (step S336: Yes), the notification control unit 125 controls the display unit 126 and the transmission unit 127 to identify the capsule endoscope 110. Information indicating that there is a high possibility of stagnation in the organ is notified, and the process is terminated (step S337). On the other hand, when the organ site where the capsule endoscope 110 is located is not a specific organ (step S331: No) and when the time T during which the specific organ is continuously imaged is less than the threshold T _MAX (step S336: No), notification is made. The control unit 125 ends the process. It should be noted that, subject threshold T _A is a variable that determined the subject of gender, age, such as by history. Moreover, organ threshold T _B is a variable determined by the particular organ preset.

The observer and subject carrying the communication device 1 can recognize that the capsule endoscope 110 is stagnating in a specific organ by a threshold value T _MAX or more by receiving notification of excess information. Therefore, in this case, the observer can perform a treatment of recalling the subject and moving the capsule endoscope 110. In addition, the subject can visit the observer himself and inform the observer.

  FIG. 9 is a flowchart illustrating a processing procedure for notifying arrival information. As illustrated in FIG. 9, the notification control unit 125 determines whether the organ site where the capsule endoscope 110 is located is a specific organ based on the latest determination result by the organ determination unit 123. (Step S341). When the organ site where the capsule endoscope 110 is located is a specific organ (step S341: Yes), the notification control unit 125 controls the display unit 126 and the transmission unit 127 to identify the capsule endoscope 110. The fact that the organ has been reached is notified and the process is terminated (step S342). When the organ site where the capsule endoscope 110 is located is not a specific organ (step S341: No), the notification control unit 125 ends the process. Note that this processing may not be performed after the capsule endoscope 110 reaches a specific organ.

  An observer and a subject carrying the communication device 1 can grasp that the capsule endoscope 110 has reached a specific organ by receiving notification of arrival information. For example, when the specific organ is set as the small intestine, the observer may pass through the stomach without stagnation or stay in the stomach based on the presence or absence of notification. Can be determined.

  According to the image processing system according to the first embodiment, when the capsule endoscope apparatus is moving in the body, not the image of the organ part itself but the determination information as a result of processing the image is notified. Thus, the observer can easily know the organ site where the capsule endoscope apparatus is located in real time. In particular, even if the capsule endoscope device is stagnated in an organ site or is stopped and transmission of an in-vivo image fails, the organ site when the capsule endoscope device is stagnant is determined. The observer can easily grasp. Therefore, the burden on the observer when the capsule endoscope apparatus is moving inside the body can be reduced.

  Further, according to the image processing system according to the first embodiment, the in-vivo image is processed in real time to determine the organ site where the capsule endoscope apparatus is located, so that the signal generated by the capsule endoscope apparatus Compared with the case where the organ site where the capsule endoscope apparatus is located is determined from the body surface based on the strength, the determination can be made more accurately. Furthermore, according to the first embodiment, it can be understood that the capsule endoscope apparatus is performing imaging processing in the body without any abnormality on the apparatus.

  Furthermore, according to the image processing system according to the first embodiment, the organ site where the capsule endoscope apparatus is located can be grasped without viewing the in-vivo image in real time with a simple display device or the like. The subject can ensure freedom of action. Further, according to the first embodiment, the subject can also grasp the organ site where the capsule endoscope apparatus is located, so that anxiety about the examination is reduced.

  In the first embodiment, the notification unit 124 includes the display unit 126 and the transmission unit 127 and notifies the determination information. Instead of or in addition to these, the notification unit 124 notifies the determination information by sound, vibration, or the like. A notification device may be provided. Further, the notification unit 124 may include only one of the display unit 126 and the transmission unit 127. For example, even when only the display unit 126 is provided, the observer can grasp the movement information of the capsule endoscope 110 based on the report of the subject. Furthermore, in Embodiment 1, the notification unit 124 notifies four types of determination information. However, the notification unit 124 may notify one or more types of determination information designated among the four types of determination information.

(Modification)
By the way, in the first embodiment, it is determined whether the capsule endoscope 110 has reached the specific organ and notified of the determination information. However, in the present modification, whether the capsule endoscope 110 has passed the specific organ. And determining information is notified.

  The image processing system according to the present modification has the same configuration as that of the image processing system 100, and the processing procedure when notification of determination information is performed is the same as the steps S101 to S106 shown in FIG. The contents of the determination information notified in S105 are different. Specifically, in the process of step S105, the notification unit 124 indicates the organ part information and the time during which the capsule endoscope 110 continues to image the organ part below the specific organ after passing through the specific organ. It shows that it has passed the specific organ, the post-passage time information that is the information to show, the post-passage excess information that is the information that indicates that the time of continuing to image the organ part below the specific organ has exceeded the threshold, and Four types of information are reported, including passage information that is information.

FIG. 10 is a flowchart illustrating a procedure of processing for notifying post-passage time information. As shown in FIG. 10, the notification control unit 125, based on the latest determination result by the organ determination unit 123, the organ part where the capsule endoscope 110 is located is an organ part below the specific organ. It is determined whether or not (step S421). When the organ part where the capsule endoscope 110 is located is an organ part below the specific organ (step S421: Yes), the notification control unit 125 is the first time that the capsule endoscope 110 is below the specific organ. It is determined whether or not an organ site has been imaged (step S422). When the organ part lower than the specific organ is imaged for the first time (step S422: Yes), the notification control unit 125 indicates the time when the capsule endoscope 110 last imaged the specific organ. The time T _E passing through the specific organ is stored in the memory 125a (step S423). Also, if the middle imaging specific organs (step S422: No), the notification control unit 125, the most recent time the capsule endoscope 110-vivo images from the time T _N imaged has passed a specific organ T _E Subtraction is performed to calculate a time T ′ during which the capsule endoscope 110 continues to image an organ part below the specific organ (step S424). Thereafter, the notification control unit 125 controls the display unit 126 and the transmission unit 127 to notify the time T ′ during which the organ part below the specific organ is imaged, and ends the process (step S425). . When the organ part where the capsule endoscope 110 is located is not an organ part below the specific organ (step S421: No), the notification control unit 125 ends the process.

  The observer and subject carrying the communication device 1 can easily grasp the time T ′ during which the capsule endoscope 110 continues to image the lower part of the specific organ by receiving the post-passage time information notification. Therefore, it becomes easy for the observer to determine whether or not the capsule endoscope 110 is stagnant in an organ part below the specific organ. For example, when the specific organ is designated as the esophagus, the time when the organ lower than the esophagus, that is, the stomach is continuously imaged is notified, so the observer can check whether the capsule endoscope 110 is stagnant in the stomach. It becomes easier to judge whether or not.

FIG. 11 is a flowchart illustrating a procedure of a process of notifying excess information after passing. As shown in FIG. 11, the notification control unit 125 is based on the latest determination result by the organ determination unit 123, and the organ part where the capsule endoscope 110 is located is an organ part below the specific organ. It is determined whether or not (step S431). When the organ site where the capsule endoscope 110 is located is an organ site below the specific organ (step S431: Yes), the notification control unit 125 is the first time that the capsule endoscope 110 is below the specific organ. It is determined whether or not an organ site has been imaged (step S432). When the first lower portion than a specific organ organ captured (step S432: Yes), the notification control unit 125 reads the subject thresholds _{T A} and organ threshold _{T C,} multiplied by the subject thresholds _{T A} and organ threshold _{T C} Then, a threshold T ′ _MAX , which is a measure of the time required for the capsule endoscope 110 to normally pass through an organ site below the specific organ, is calculated (step S433). Then, the notification control unit 125, the capsule endoscope 110 stores the time _{T E,} which has passed through a specific organ in the memory 125a (step S434). Further, when an organ site below the specific organ is being imaged (step S432: No), the notification control unit 125 causes the capsule endoscope 110 to pass through the specific organ from the time T _N when the latest in-vivo image is captured. By subtracting the time T _E , the time T ′ during which the organ part below the specific organ is imaged is calculated (step S435). Thereafter, the notification control unit 125 determines whether or not the time T ′ during which the organ part below the specific organ is imaged is equal to or greater than the threshold value T ′ _MAX (step S436). When the time T ′ during which an organ part below the specific organ is continuously imaged is equal to or greater than the threshold T ′ _MAX (step S436: Yes), the notification control unit 125 controls the display unit 126 and the transmission unit 127, Information indicating that there is a high possibility that the capsule endoscope 110 is stagnant in an organ part below the specific organ is notified, and the process is terminated (step S437). On the other hand, when the organ part where the capsule endoscope 110 is located is not an organ part below the specific organ (step S431: No), the time T ′ during which the organ part below the specific organ is continuously imaged is a threshold value. If it is less than T ′ _MAX (step S436: No), the notification control unit 125 ends the process. Incidentally, the organ threshold T _C is a variable than a specific organ preset corresponding to the lower organ.

The observer and subject carrying the communication device 1 receive the notification of excess information after passing, thereby confirming that the capsule endoscope 110 is stagnating at a threshold T ′ _MAX or more in an organ part below the specific organ. I can grasp. Therefore, in this case, the observer can perform a treatment of recalling the subject and moving the capsule endoscope 110. In addition, the subject can visit the observer himself and inform the observer.

  FIG. 12 is a flowchart illustrating a processing procedure for notifying passage information. As shown in FIG. 12, the notification control unit 125, based on the latest determination result by the organ determination unit 123, the organ part where the capsule endoscope 110 is located is an organ part below the specific organ. Is determined (step S441). When the organ part where the capsule endoscope 110 is located is an organ part below the specific organ (step S441: Yes), the notification control unit 125 controls the display unit 126 and the transmission unit 127 to perform capsule type. The notification that the endoscope 110 has passed the specific organ is notified, and the process is terminated (step S442). When the organ part where the capsule endoscope 110 is located is not an organ part below the specific organ (step S441: No), the notification control unit 125 ends the process as it is. Note that this processing may not be performed after the capsule endoscope 110 passes through the specific organ.

  The observer and subject carrying the communication device 1 can confirm that the capsule endoscope 110 has passed through the specific organ by receiving notification of passage information. For example, if a specific organ is set as the stomach, the observer and the subject have passed through the stomach without stagnation or stagnation in the stomach based on the presence or absence of notification. Can determine if there is a possibility.

  According to this modified example, as in the first embodiment, the observer can easily know in real time the organ site where the capsule endoscope device is located. The burden on the observer when moving is reduced. Note that the image processing system according to the present invention may notify all of the seven types of determination information shown in the first embodiment and the modified example, and one or more types of determination specified among the seven types of determination information. Information may be notified.

(Embodiment 2)
In the first embodiment, the receiving device 120 includes the organ determination unit 123. However, in the second embodiment, the capsule endoscope includes the organ determination unit. FIG. 13 is a block diagram of a schematic configuration of the image processing system 200 according to the second embodiment. The image processing system 200 includes a capsule endoscope 210 and a receiving device 220.

  As shown in FIG. 13, the capsule endoscope 210 includes a transmission unit 212 and a control unit 214 instead of the transmission unit 112 and the control unit 114 of the capsule endoscope 110, and also includes an organ determination unit 215. Other configurations of the capsule endoscope 210 are the same as other configurations of the capsule endoscope 110. The organ determination unit 215 processes the in-vivo image captured by the imaging unit 111 in real time, and determines the organ site where the capsule endoscope 210 is located when the in-vivo image is captured. The transmission unit 212 transmits the in-vivo image and the determination result by the organ determination unit 215 to the outside of the body. The control unit 214 controls processing of each unit of the capsule endoscope 210.

  As illustrated in FIG. 13, the reception device 220 includes a reception unit 221 and a control unit 228 instead of the reception unit 121 and the control unit 128 of the reception device 120, and does not include the organ determination unit 123. Note that other configurations of the receiving device 220 are the same as the other configurations of the receiving device 120. The receiving unit 221 receives the in-vivo image and the determination result by the organ determining unit 215. The control unit 228 controls processing of each unit of the receiving device 220.

  Next, the determination information notification process performed by the image processing system 200 will be described with reference to FIG. FIG. 14 is a flowchart illustrating a processing procedure when the image processing system 200 notifies the determination information. First, the control unit 214 controls the imaging unit 111 to capture an in-vivo image (step S501). Thereafter, the control unit 214 controls the organ determination unit 215 to determine the organ site where the capsule endoscope 210 is located (step S502). Thereafter, the control unit 214 transmits the in-vivo image and the organ region determination result to the outside of the body by the transmission unit 212 (step S503). The control unit 228 receives the in-vivo image and the organ part determination result transmitted by the transmission unit 212 via the reception unit 221 (step S504). Next, the control unit 228 controls the notification unit 124 to notify the determination information (step S505). Thereafter, the control unit 228 determines whether or not the number of in-vivo images that have undergone the processing in steps S240 and S250 has reached a preset number of scheduled imaging (step S506). When the number of in-vivo images for which processing has ended exceeds the scheduled number of images to be captured, the control units 214 and 228 end the processing (step S506: Yes). On the other hand, when the number of in-vivo images that have been processed is equal to or less than the planned number of captured images, the control units 214 and 228 repeat the processing of steps S501 to S505 (step S506: No). Note that the control units 214 and 228 may repeat the processes in steps S501 to S505 until the capsule endoscope 210 is discharged from the body.

  In step S502, the organ determination unit 215 performs the same process as the process performed by the organ determination unit 123 in step S104 illustrated in FIG. In step S505, the notification unit 124 performs a process similar to the process performed by the notification unit 124 in step S105 illustrated in FIG.

  Similar to the image processing system 100, the image processing system 200 determines an organ site where the capsule endoscope 210 is located when an in-vivo image is taken, and notifies information based on the determination result. Therefore, according to the image processing system according to the second embodiment, the observer can easily know in real time the organ site where the capsule endoscope device is located. It is possible to reduce the burden on the observer when moving the camera.

  In the first and second embodiments, the organ determination units 123 and 215 perform post-DCT encoding when performing determination processing of the organ site where the capsule endoscopes 110 and 210 are located when each in-vivo image is captured. Although the file size of the in-vivo image is used, information regarding the frequency when the in-vivo image is converted into the frequency space may be used.

  Since the properties and colors of the mucosal surface of each organ differ from one organ to another, when an in-vivo image is converted into a frequency space, the distribution of frequency components differs depending on the organ imaged in the image. Further, for example, comparing the movement status of the capsule endoscopes 110 and 210 in the small intestine and the large intestine, it can be seen that the small intestine moves smoothly and the large intestine is stagnated because the contents are clogged. Therefore, in the small intestine, the change between images is large, and the amount of change in the frequency distribution is large. On the other hand, there is little change between images in the large intestine, and the amount of change for the frequency is small. Therefore, it is possible to determine the organ imaged in the in-vivo image using the information relating to the distribution of frequency components and the information relating to the amount of change in the frequency distribution. Examples of the information relating to the frequency include a power spectrum calculated by Fourier transform, a DCT coefficient calculated at the time of decoding after DCT encoding, and the like. Hereinafter, an organ determination process using a DCT coefficient will be described.

  FIG. 15 is a flowchart showing the procedure of the organ determination process when the DCT coefficient is used. For example, in step S140 shown in FIG. 2, the organ determination unit 123 first calculates a DCT coefficient of the in-vivo image (step S601). The control unit 128 stores the DCT coefficient calculated by the organ determination unit 123 in the storage unit 122 (step S602). The organ determination unit 123 determines whether or not the received in-vivo images have reached a predetermined number (step S603). When the in-vivo images reach the predetermined number (step S603: Yes), the organ determination unit 123 reads the DCT coefficients of the predetermined number of in-vivo images, and changes in the DCT coefficients between the in-vivo images adjacent in time series. The amount is calculated (step S604). Thereafter, the organ determination unit 123 determines the position of the capsule endoscope 110 when the in-vivo image is captured based on the DCT coefficient of the in-vivo image and the amount of change in the DCT coefficient. The organ part to be determined is determined, and the process is terminated (step S605). If the in-vivo images have not reached the predetermined number (step S603: No), the organ determination unit 123 ends the process without performing the processes of steps S604 and 605. In this way, the organ determination unit 123 determines the organ site where the capsule endoscope 110 is located when the in-vivo image is captured for each predetermined number. In the image processing system 200, the organ determination unit 215 and the control unit 214 perform the processes in steps S601 to S605 on the in-vivo image captured by the imaging unit 111.

  Here, the DCT coefficient will be described. The DCT coefficient is a coefficient calculated by discrete cosine transform for each image area corresponding to a pixel block of 8 × 8 pixels, and is a coefficient indicating a frequency component of the image area corresponding to each pixel block. Specifically, as shown in FIG. 16, the DCT coefficient of each pixel block is a set of 64 DCT coefficients that are each element of the quantization matrix of 8 × 8 matrix. Of these 64 DCT coefficients DCT1 to DCT64, DCT1 corresponds to a DC component, and DCT2 to DCT64 correspond to an AC component. Further, as shown in FIG. 16, DCT2 to DCT64 indicate frequency information from low frequency components to high frequency components, respectively. In JPEG, RGB values are converted into Y values indicating luminance information and U and V values indicating color information, and then discrete cosine conversion is performed for each Y value, U value, and V value. , DCT coefficients for each V value are calculated. That is, each pixel block has 3 × 64 = 192 DCT coefficients.

  The DCT coefficient calculated by the organ determination unit 123 in step S601 is an average value of the DCT coefficients of the in-vivo images. The average value of DCT coefficients means a set of average DCT1 to average DCT64 calculated based on each of DCT1 to DCT64 of all pixel blocks on the image area. Since the DCT coefficients for each of the Y value, U value, and V value are calculated, the average value of the DCT coefficients of each in-vivo image is also a set of 192 DCT coefficients. The change amount of the DCT coefficient calculated by the organ determination unit 123 in step S604 is a set of change amounts of the average DCT1 to the average DCT64. Note that the organ determination unit 123 may not calculate the average value for all of the DCT1 to DCT64 but may calculate the average value for a specific DCT coefficient and use this for the processing in steps 604 and S605.

  In step S605, the organ determination unit 123 first calculates a feature vector of the in-vivo image based on the DCT coefficient, compares the feature vector with reference data, and captures the in-vivo image. It is determined whether the organ site where 110 is located is the stomach, the small intestine or the large intestine. Further, the organ determination unit 123 calculates a feature vector based on the amount of change in the DCT coefficient for the in-vivo image determined to have captured the small intestine or the large intestine, and compares the feature vector with the reference data. It is determined whether the organ site where the capsule endoscope 110 is located at the time of imaging is the small intestine or the large intestine. The reference data is dictionary data in which feature vectors of in-vivo images are classified in advance for each imaged organ.

  In this embodiment, the organ determination unit 123 calculates a three-dimensional feature vector having the feature quantities A, B, and C calculated based on the DCT coefficient of the Y value as vector elements. Here, the feature amount A is an average value of a plurality of DCT coefficients indicating a low frequency component among the average DCT2 to the average DCT64 of the Y value of each in-vivo image. The feature amount B is an average value of a plurality of DCT coefficients indicating a high frequency component among the average DCT2 to the average DCT64 of the Y values of the in-vivo images. The feature amount C is an average DCT1 that is an average value of DC components of the Y values of the in-vivo images.

  FIG. 17 is an example of reference data to be compared with feature vectors calculated based on DCT coefficients. As shown in FIG. 17, an image obtained by imaging the stomach has a larger feature quantity A indicating a low frequency component than an in-vivo image obtained by imaging the small or large intestine. On the other hand, in the in-vivo image obtained by imaging the small intestine or the large intestine, the feature quantity B indicating the high-frequency component is larger than the in-vivo image obtained by imaging the stomach. Therefore, as shown in FIG. 17, the feature vectors can be classified for each organ in the feature space. Therefore, the organ determining unit 123 uses the known vector such as the kNN method and the subspace method based on the feature vector and the reference data, and the organ in which the capsule endoscope 110 is located when the in-vivo image is captured. Determine whether the site is the stomach, small intestine or large intestine.

  Further, in this embodiment, the organ determination unit 123 calculates a feature vector having the feature amounts D, E, and F calculated based on the amount of change in the DCT coefficient of the Y value as vector elements. Here, the feature amount D is a change amount of an average value of a plurality of DCT coefficients indicating a low frequency component among the average DCT2 to the average DCT64 of the Y values of the in-vivo images. The feature amount E is a change amount of an average value of a plurality of DCT coefficients indicating a high frequency component among the average DCT2 to the average DCT64 of the Y value of each in-vivo image. The feature amount F is an average DCT1 that is an average value of the DC components of the Y value of each in-vivo image.

  FIG. 18 is an example of reference data to be compared with a feature vector calculated based on the change amount of the DCT coefficient. As shown in FIG. 18, the feature amounts D, E, and F of the in-vivo image obtained by imaging the large intestine are smaller than the feature amounts D, E, and F of the in-vivo image obtained by imaging the small intestine. Therefore, as shown in FIG. 18, feature vectors can be classified for each organ in the feature space. Therefore, the organ determination unit 123 determines the organ site where the capsule endoscope 110 is located when the in-vivo image is captured by a known identification method based on the feature vector and the reference data.

  Note that the organ determination unit 123 replaces the DCT coefficient of each in-vivo image and the amount of change in the processing of steps S604 and 605, and calculates the average value of the DCT coefficient calculated for each predetermined number of time-series images and the DCT coefficient. Using the amount of change, an organ captured in the in-vivo image may be determined every predetermined number.

  Further, the organ determination unit 123 may calculate the change amount of the DCT coefficient only for the in-vivo image captured after it is determined that the organ site where the capsule endoscope 110 is located is the small intestine or the large intestine. In this case, the calculation processing amount of the organ determination unit 123 can be reduced, and the processing time is shortened. In addition, the organ determination unit 123 may perform the organ determination process using only one of the DCT coefficient or the change amount of the DCT coefficient. The organ determination unit 123 combines the file size of the in-vivo image, the DCT coefficient, and these change amounts, A determination process may be performed. In addition, the organ determination unit 123 may perform an organ determination process based on the color of the mucous membrane surface of each organ.

1 is a block diagram showing a schematic configuration of an image processing system according to a first embodiment of the present invention. It is a flowchart which shows the process sequence in case the image processing system shown in FIG. 1 notifies notification of determination information. It is a figure which shows transition of the change of the file size with respect to imaging time. It is a figure which shows transition of the change of the variation | change_quantity of the file size with respect to imaging time. It is a flowchart which shows the procedure of the organ determination process which the organ determination part shown in FIG. 1 performs. It is a flowchart which shows the procedure of the process which notifies organ site | part information. It is a flowchart which shows the procedure of the process which notifies the time when the capsule endoscope shown in FIG. 1 continues imaging a specific organ. FIG. 3 is a flowchart showing a procedure for performing a process of notifying that when the time during which the capsule endoscope shown in FIG. 1 continues to image a specific organ exceeds a threshold value. FIG. 2 is a flowchart showing a procedure for performing processing for notifying that when the capsule endoscope shown in FIG. 1 reaches a specific organ. FIG. 12 is a flowchart illustrating a procedure of a process for notifying a time during which the capsule endoscope according to the modification of the first embodiment continues to image an organ part below a specific organ. 10 is a flowchart showing a procedure for notifying that when the time during which the capsule endoscope according to the modification of the first embodiment continues to image an organ part below a specific organ exceeds a threshold value. 12 is a flowchart showing a procedure for performing processing for notifying that when a capsule endoscope according to a modification of the first embodiment passes a specific organ; FIG. 3 is a block diagram illustrating a schematic configuration of an image processing system according to a second embodiment. It is a flowchart which shows the process sequence in case the image processing system shown in FIG. 14 notifies notification of determination information. It is a flowchart which shows the procedure of the organ determination process at the time of using a DCT coefficient. It is a figure which shows the DCT coefficient of an 8x8 pixel block. It is a figure which shows an example of the reference data compared with the feature vector calculated based on the DCT coefficient. It is a figure which shows an example of the reference data compared with the feature vector calculated based on the variation | change_quantity of a DCT coefficient.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Communication apparatus 2,121,221 Reception part 3,126 Display part 100,200 Image processing system 110,210 Capsule type endoscope 111 Imaging part 112,127,212 Transmission part 113,122 Storage part 114,128,214, 228 Control unit 120, 220 Reception device 122a Image storage unit 123, 215 Organ determination unit 124 Notification unit 125 Notification control unit 125a Memory

Claims (18)

An image processing system including a capsule endoscope device that captures an in-vivo image while moving in a body to be examined, and a receiving device that receives a signal transmitted by the capsule endoscope device,
Organ determination for determining an organ site where the capsule endoscope apparatus is located when the in-vivo image is captured based on a feature amount indicating a feature of the in-vivo image obtained by the capsule endoscope apparatus. And
A notification unit for notifying determination information based on a determination result by the organ determination unit when the capsule endoscope apparatus is moving in the body;
An image processing system comprising:   The image processing system according to claim 1, wherein the notification unit notifies organ part information indicating an organ part determined by the organ determination unit as the determination information.   The image processing system according to claim 1, wherein the notification unit notifies that the capsule endoscope apparatus has reached a specific organ site as the determination information.   The notification unit notifies, as the determination information, a time during which the same capsule part is imaged after the capsule endoscope apparatus reaches a specific organ part. The image processing system according to any one of the above.   The image processing system according to claim 1, wherein the notification unit notifies that the capsule endoscope apparatus has passed a specific organ site as the determination information.   6. The notification unit according to claim 1, wherein, as the determination information, after the capsule endoscope apparatus passes through a specific organ part, the notification unit notifies the time during which the same organ part is imaged. The image processing system according to any one of the above.   The notification unit, when the capsule endoscope apparatus reaches a specific organ part as the determination information, and when the time during which the same organ part is imaged exceeds a predetermined time, the capsule endoscope The image processing system according to any one of claims 1 to 6, wherein information indicating that the mirror device is highly likely to be stagnant is notified.   The notification unit, as the determination information, when the capsule endoscope apparatus has passed a specific organ part and the time taken to image the same organ part exceeds a predetermined time, the capsule endoscope The image processing system according to claim 1, wherein information indicating that there is a high possibility that the device is stagnant is notified.   The image processing system according to claim 1, wherein the notification unit notifies the determination information by display display.   The image processing system according to claim 1, wherein the notification unit notifies the determination information by sound information and / or vibration information.   The image processing system according to claim 1, wherein the notification unit notifies the determination information by transmitting the determination information to an external portable terminal.   The image processing system according to claim 1, wherein the notification unit is provided in the receiving device.   The organ determination unit, as the feature quantity of the in-vivo image, includes at least a compression-encoded file size of the in-vivo image and a DCT coefficient calculated when decoding the DCT-encoded in-vivo image. The image processing system according to claim 1, wherein an organ site is determined based on any one of the above.   The organ determination unit divides the compression-encoded in-vivo images into one or more predetermined in-vivo image groups, and the in-vivo image groups before and after adjacent in time series as the feature amount of the in-vivo images. And determining an organ site based on at least one of a change amount of a file size between the two and a change amount of a DCT coefficient calculated when the in-vivo image subjected to DCT encoding is decoded. The image processing system according to claim 1.   The image processing system according to claim 1, wherein the organ determination unit is provided in the receiving device.   The image processing system according to claim 1, wherein the organ determination unit is provided in the capsule endoscope apparatus. An image processing system including a capsule endoscope device that captures an in-vivo image while moving in a body to be examined, and a receiving device that receives a signal transmitted by the capsule endoscope device,
An image processing system comprising: a notification unit that notifies position information indicating a position of the capsule endoscope apparatus when the capsule endoscope apparatus is moving inside the body. An organ determination unit that determines an organ site where the capsule endoscope device is located when the in-vivo image is captured based on a feature amount indicating a feature of the in-vivo image in which the body is captured;
A notification unit for notifying determination information based on a determination result by the organ determination unit when the capsule endoscope apparatus is moving in the body;
An image processing apparatus comprising:

Images