JP2013085593A - Medical image display apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce not only the diagnostic reading time required in diagnostic reading but also the laborious load of a diagnostic reading person.SOLUTION: The medical image display apparatus 4 includes an acquiring part 4c for acquiring a plurality of images imaged by a capsule endoscope 2 being one example of an intraorganic imaging apparatus and the imaging position and imaging posture of the capsule endoscope 2 at every image, a region dividing part 13 for dividing an organ along the stretching direction thereof, an image classifying part 14 for determining the position of the image to the organ using the imaging position and imaging posture of the capsule endoscope 2 at every image, a preferential degree determining part 15 for extracting the feature quantity at every image and determining the preferential degree of respective images in the regions corresponding to the feature quantity of the image at every region, an image forming part 16 for forming a display image wherein a plurality of the images in the regions are arranged in the region corresponding to the preferential degrees of the respective images in the regions, and a display part 4e for displaying a distribution image thereof.

Description

  Embodiments described herein relate generally to a medical image display device, for example, a medical image display device that displays a medical image captured by a capsule endoscope.

  In recent years, capsule endoscopes targeting the digestive tract, which is an example of a tubular organ, have been put into practical use. This capsule endoscope mainly sends an imaging unit such as a CCD camera that captures an image (medical image) in a body cavity, an illumination unit such as an LED that brightens the inside of the body cavity, and captured image information to a receiving device outside the body. And a power supply unit such as a button battery for supplying power to these units.

  Such intracapsular visual field is an intracavity of several pieces / second between about 8 hours and 10 hours from the oral administration to a subject until it is naturally discharged from the anus mainly by peristaltic movement of each digestive tract. Take a picture. For example, assuming that two images are taken for 8 hours per second, as many as 57,000 600 images are taken in one inspection.

  Thus, the number of images acquired in one inspection is large and tends to increase further in the future. For this reason, since the burden on the radiogram interpreter becomes larger than it is now, the development of display devices and software that support radiogram interpretation has been carried out. For example, for the purpose of improving the operability and efficiency of interpretation, a method of generating a diagnostic image based on the position of the endoscope from an endoscope image captured in time series, or the endoscope image is divided into regions There has been proposed a method of generating a diagnostic image by deforming and pasting the deformed image on a two-dimensional or three-dimensional model.

JP 2010-240000 A

  However, as described above, generating a diagnostic image based on a time-series endoscope position or generating a diagnostic image by dividing an endoscope image into regions and generating a diagnostic image is the number of images used for diagnosis. On the other hand, it makes it difficult for the radiogram reader to confirm the feature region and feature amount in the diagnostic image. In addition, it is impossible to confirm whether there is any omission of pathological findings other than the image used for generating the diagnostic image and the specified feature. For this reason, the interpretation time required for interpretation becomes long, and the labor burden on the interpreter also increases.

  The problem to be solved by the present invention is to provide a medical image display apparatus capable of reducing the interpretation time required for interpretation and the labor burden on the interpreter.

  The medical image display device according to the embodiment includes a plurality of images captured by an intra-organ imaging device that moves in a tubular organ and images in time series, and an imaging position and an imaging posture of the intra-organ imaging device for each image. Determining the position of the image with respect to the organ using the acquisition unit, the region dividing unit that divides the organ into a plurality of regions along the extending direction, and the imaging position and imaging posture of the intra-organ imaging device for each image; An image classification unit that classifies a plurality of images for a plurality of regions, and extracts image feature amounts for each image, and priorities that determine the priority of each image in the region according to the image feature amounts for each region A degree determination unit, an image generation unit that generates a display image in which a plurality of images in the region are arranged according to the priority of each image in the region, and a display unit that displays the display image.

It is a block diagram showing a schematic structure of a capsule endoscopy system concerning an embodiment. It is a flowchart which shows the flow of the image display process which the medical image display apparatus of the capsule endoscopy system shown in FIG. 1 performs. It is explanatory drawing for demonstrating the calculation of the locus | trajectory and imaging direction of a capsule endoscope. It is explanatory drawing for demonstrating the production | generation of a digestive tract model, and calculation of the imaging region of a capsule endoscope. It is explanatory drawing for demonstrating the area division | segmentation of a digestive tract model. It is a figure which shows an example of the endoscopy screen containing a 1st histogram image. It is a figure which shows an example of the endoscopy screen containing a 2nd histogram image.

  Embodiments of the present invention will be described with reference to the drawings.

  As shown in FIG. 1, a capsule endoscopy system 1 according to the present embodiment includes a capsule endoscope 2 that is an example of an intra-organ imaging device that images a tubular organ, and the capsule endoscope 2. Receiving device 3 that receives the data of the above-mentioned data, and a medical image display device 4 that displays a medical image in the organ based on the data of the receiving device 3.

  The capsule endoscope 2 includes an illumination unit 2a that illuminates the inside of a tubular organ (for example, the digestive tract), an imaging unit 2b that images the inside of the tubular organ, and a position detection unit that detects an imaging position of the capsule endoscope 2 2c, a posture detection unit 2d that detects the imaging posture of the capsule endoscope 2 at the imaging position, and a transmission unit 2e that transmits information such as the captured image, the imaging position, and the imaging posture to the reception device 3 outside the body. The power supply unit 2f that supplies power to each unit and the control unit 2g that controls each unit are provided. Each of these parts is enclosed in a capsule housing.

  The illuminating unit 2a is an illuminating device that irradiates light into a tubular organ that can be imaged by the imaging unit 2b. As this illumination part 2a, LED (light emitting diode) etc. are used, for example. The capsule housing is formed with a transparent tip portion, for example, so that the internal wall of the organ can be irradiated with light by the built-in illumination unit 2a.

  The imaging unit 2b is an imaging device that performs imaging in a time series with a constant time interval (a constant cycle). For example, a CCD (charge coupled device) camera or the like is used as the imaging unit 2b. In addition to the light irradiation of the illuminating unit 2a described above, the capsule housing is formed so that the inner wall of the organ can be imaged by the built-in imaging unit 2b.

  The position detection unit 2c is a detection device that detects the imaging position of the capsule endoscope 2 in synchronization with the imaging of the imaging unit 2b. This imaging position is a spatial position of the capsule endoscope 2 and is determined by the X axis, the Y axis, and the Z axis. Regarding this spatial position detection, for example, there are various position detection means using the strength of electromagnetic waves, or position detection means using a small GPS (global positioning system) built in the capsule endoscope 2. It is possible to use a simple position detecting means.

  The posture detection unit 2d is a detection device that detects the imaging posture of the capsule endoscope 2 at the imaging position together with the above-described imaging position in synchronization with the imaging of the imaging unit 2b. As the posture detection unit 2d, various posture detection units such as a posture detection unit using an acceleration sensor, a gyro sensor, or the like can be used.

  The transmission unit 2e includes the image captured by the imaging unit 2b, the imaging position of the capsule endoscope 2 detected by the position detection unit 2c, and the imaging posture of the capsule endoscope 2 detected by the posture detection unit 2d. This is a transmission device that wirelessly transmits the included information to the reception device 3. For example, a transmission antenna is used as the transmission unit 2e.

  The power supply unit 2f is a power supply device that supplies power to each unit. For example, a button-type battery is used as the power supply unit 2f. Note that a plurality of batteries are provided according to the required capacity in order to cope with an inspection time of about 8 to 10 hours.

  The control unit 2g performs control of imaging the inner wall of the organ at a constant time interval (a constant cycle) by the imaging unit 2b while irradiating the inside of the organ by the illumination unit 2a. Further, the control unit 2g sequentially transmits information including the image captured by the imaging unit 2b, the imaging position detected by the position detection unit 2c, and the imaging posture detected by the posture detection unit 2d to the reception device 3 by the transmission unit 2e. Control to send.

  When the capsule endoscope 2 is swallowed by a patient, the capsule endoscope 2 performs imaging at a constant time interval (a constant cycle) while moving in a tubular organ by a peristaltic motion. In this imaging, irradiation light is irradiated from the illumination unit 2a to the inner wall of the organ by the power from the power supply unit 2f, and the reflected light reflected by the inner wall is photoelectrically converted by the imaging unit 2b to generate image data. At this time, the position detection unit 2c detects the imaging position of the capsule endoscope 2 in synchronization with the generation of the image data. Similarly, the posture detection unit 2d also detects the capsule endoscope in synchronization with the generation of the image data. 2 imaging postures are detected. The image data, the imaging position and the imaging posture of the capsule endoscope 2 are transmitted to the receiving device 3 outside the patient's body by wireless communication by the transmission unit 2e. The generation of image data, the detection of the imaging position by the position detection unit 2c, the detection of the imaging posture by the posture detection unit 2d, and the transmission of data are controlled at a constant time interval (for example, several frames / second) under the control of the control unit 2g. Done in

  The receiving device 3 includes a receiving unit 3a such as a receiving antenna that wirelessly receives information transmitted from the capsule endoscope 2, and a storage medium 3b such as a flash memory that stores the received information. The receiving device 3 receives information transmitted from the capsule endoscope 2, that is, image data, information including the imaging position and the imaging posture of the capsule endoscope 2, and sequentially stores them in the storage medium 3b. . The storage medium 3b is detachable. The storage medium 3b is attached to the medical image display device 4, and all the information in the storage medium 3b is input to the medical image display device 4.

  The medical image display apparatus 4 includes a control unit 4a that centrally controls each unit, a storage unit 4b that stores various programs and various data, an acquisition unit 4c that acquires various data from the storage unit 4b, and various processes. 4d, a display unit 4e for displaying various images, and an operation unit 4f for receiving an input operation from a user (user) such as a doctor. This medical image display device 4 functions as a workstation.

  The control unit 4a controls each unit based on various programs and various data stored in the storage unit 4b. Further, the control unit 4a executes a series of data processing based on various programs and various data stored in the storage unit 4b.

  The storage unit 4b includes a memory for storing various programs and various data in addition to a memory functioning as a work area for the control unit 4a. As these memories, for example, a ROM, a RAM, a magnetic disk device (HDD), a flash memory (semiconductor disk device), or the like is used. The storage unit 4b stores all information in the storage medium 3b of the receiving device 3, that is, image data for all images, information including the imaging position and the imaging posture of the capsule endoscope 2.

  The acquisition unit 4c acquires information such as image data and the imaging position and imaging posture of the capsule endoscope 2 for each image from the storage unit 4b, and transmits the information to the processing unit 4d. The acquisition of information is performed in response to, for example, a user such as a doctor performing an input operation on the operation unit 4f and instructing the processing unit 4d to execute the process.

  The processing unit 4d includes an imaging trajectory calculation unit 11 that calculates the trajectory and imaging direction of the capsule endoscope 2, an image position calculation unit 12 that calculates the position of the image with respect to the tubular organ, and a plurality of organs along the extending direction. A region dividing unit 13 that divides the image into regions, an image classifying unit 14 that classifies a plurality of images (endoscopic images) for each divided region, and a priority determining unit that determines image priority for each region 15 and an image generation unit 16 that generates various images such as a histogram image and a composite image (diagnostic image).

  The imaging trajectory calculation unit 11 receives information about the imaging position and orientation of the capsule endoscope 2 for each image from the acquisition unit 4c, and calculates the trajectory and imaging direction of the capsule endoscope 2 according to the information. And output to the image position calculation unit 12. Note that the information regarding the imaging position and the imaging orientation is time-series position data corresponding to the number of captured images.

  The image position calculation unit 12 calculates the position of the image with respect to a tubular organ (for example, the digestive tract) based on the trajectory and the imaging direction of the capsule endoscope 2 input from the imaging trajectory calculation unit 11. For example, the image position calculation unit 12 is based on the trajectory of the capsule endoscope 2 input from the imaging trajectory calculation unit 11 and is an organ model (for example, digestion) that is a virtual organ having an arbitrary tube diameter with the trajectory serving as a core line. A tube model or the like) is created, an imaging area on the inner wall of the organ is obtained from the imaging position and imaging direction of the capsule endoscope 2, and the position of the image with respect to the organ model is specified from the position of the imaging area.

  The region dividing unit 13 includes a plurality of regions at an arbitrary distance (for example, a predetermined interval or a desired interval) along the extending direction of the organ model created by the image position calculating unit 12, that is, the trajectory of the capsule endoscope 2. Divide into In this region division, the organ model is divided on the basis of the position of the organ model in the extending direction using the locus of the capsule endoscope 2 calculated by the imaging locus calculation unit 11.

  The image classification unit 14 uses the position information of the image with respect to the organ model obtained by the image position calculation unit 12 to extract an image that exists (corresponds) in each region divided by the region division unit 13. Then, it is classified as an image group belonging to the area. Note that only one image may or may not exist in the area.

  The priority determination unit 15 performs feature extraction on the image based on the feature designated by the user, and the priority of each image group (each image in the region) for each region described above according to the feature amount. (For example, priority for displaying an image), that is, a priority order is determined. For example, when the redness (color tone: red) used as an index of bleeding inside the body cavity is designated as a feature, the image priority of the image is set higher as the redness that is the feature amount is larger. The As the feature amount, in addition to the color condition (degree), in addition to the color condition (density), it is also possible to use the area of the color portion (color region), and the values thereof. It is also possible to use a numerical value obtained by multiplying (density × color region).

  The image generation unit 16 rearranges the image group, which is a plurality of images classified into the region, in the unit of region divided by the region division unit 13, in the order of the position of the organ in the traveling direction or in the order of the time taken. The image group for each area is arranged in the order in which each area is arranged to generate a list image. That is, the position of the organ in the extending direction is set on the horizontal axis, and the image group (or the number of sheets) classified into each area is set on the vertical axis. The first histogram image is generated by arranging in a histogram. In addition, the image generation unit 16 changes the above-described rearrangement in the order of high priority and performs the same processing as described above to generate a list image, that is, a second histogram image (display image) based on the priority. Further, the image generation unit 16 uses the image group in the region designated by the user to synthesize images in any direction within the region, and generates a composite image that becomes a diagnostic image.

  Note that the processing unit 4d described above may be configured by hardware such as an electric circuit, or may be configured by software such as a program for executing these functions, or a combination of both. It may be configured.

  The display unit 4e is a display device that displays various images such as a histogram image, a composite image, and other images transmitted from the processing unit 4d in color. For example, a liquid crystal display or a CRT (Cathode Ray Tube) display is used as the display unit 4e.

  The operation unit 4f is an input unit that is input by a user such as a doctor and receives various input operations such as image display, image switching, and setting change. As the operation unit 4f, for example, an input device such as a mouse or a keyboard is used.

  Next, image display processing performed by the medical image display device 4 described above will be described in detail. When a user such as a doctor performs an input operation on the operation unit 4f to instruct execution of an image display process, the following process is executed. Here, processing for the digestive tract (for example, intestinal tract) will be described as an example of a tubular organ.

  As shown in FIG. 2, the acquisition unit 4c first reads a captured image captured by the capsule endoscope 2 from the storage unit 4b, an imaging position and an imaging posture of the capsule endoscope 2 (step S1).

  In this step S1, all images captured by the capsule endoscope 2 are read from the storage unit 4b, and further, the imaging position and imaging posture of the capsule endoscope 2 are also read from the storage unit 4b for all the images.

  After the processing of step S1, the imaging trajectory calculation unit 11 of the processing unit 4d uses the imaging position and imaging orientation of the capsule endoscope 2 read by the acquisition unit 4c to determine the trajectory and imaging direction of the capsule endoscope 2. Calculate (step S2).

  In step S2, as shown in FIG. 3, first, the imaging position of the capsule endoscope 2 read by the acquisition unit 4c is traced in time series, the locus L1 of the capsule endoscope 2 is calculated, and further acquired. Using the imaging posture of the capsule endoscope 2 read by the unit 4c, the direction in which the capsule endoscope 2 is oriented at each imaging position along the locus L1 of the capsule endoscope 2 is calculated as the imaging direction D1. To do.

  Returning to FIG. 2, after the processing in step S <b> 2, the image position calculation unit 12 of the processing unit 4 d uses the locus L <b> 1 and the imaging direction D <b> 1 of the capsule endoscope 2 input from the imaging locus calculation unit 11. The position of the image is calculated (step S3).

  In step S3, as shown in FIG. 4, first, a gastrointestinal tract model M1 having an arbitrary tube diameter is set with the locus L1 of the capsule endoscope 2 calculated by the imaging locus calculating unit 11 as a core line (center line). create. Next, the imaging region F1 on the inner wall of the digestive tract model M1 is obtained from the imaging position of the capsule endoscope 2 (position when the image is captured) and the imaging direction D1 at the imaging position, and the position of the imaging region F1 Accordingly, it is estimated where the captured image is located on the digestive tract model M1, and the position of the image with respect to the digestive tract is specified by associating the position of the digestive tract model M1 with the image.

  In the generation of the aforementioned digestive tract model M1, in order to calculate the diameter of the digestive tract model M1, for example, the thickness of the digestive tract (average thickness according to age and physique, or obtained from other examinations) Information such as patient specific thickness) is used. This information is stored in advance in the storage unit 4b. In addition, in the calculation of the imaging region F1, information on the viewing angle of the capsule endoscope 2 is used in addition to the information such as the thickness of the digestive tract described above. The information on the viewing angle is stored in advance in the storage unit 4b together with the information such as the thickness of the digestive tract described above. For example, when calculating the imaging region F1, the distance (subject distance) along the imaging direction D1 from the imaging position to the inner wall of the digestive tract is calculated using information such as the thickness of the digestive tract described above. Based on the subject distance and the aforementioned viewing angle, an imaging field, that is, an imaging area F1 is obtained.

  Here, in the creation of the aforementioned digestive tract model M1, when the capsule endoscope 2 stays in the same position and changes its direction, and the locus L1 moves up and down and left and right at the same location, the average position or the center position is set. You may recalculate and use as a point on a core line. Also, when calculating the tube diameter of the digestive tract model M1, the estimation is based on the movable upper and lower and left and right ranges when the capsule endoscope 2 moves in the same place, instead of the thickness information of the aforementioned digestive tract. You may do it. In calculating the position of the captured image, the position may be corrected using features such as the surrounding image and the shape in the image.

  Returning to FIG. 2, after the processing in step S3, the region dividing unit 13 of the processing unit 4d divides the digestive tract model M1 created by the image position calculating unit 12 into regions at an arbitrary distance along the extending direction (step S4). ).

  In this step S4, as shown in FIG. 5, the created digestive tract model M1 is divided into regions at an arbitrary distance, for example, a predetermined equal interval along the extending direction, that is, the locus L1 of the capsule endoscope 2. . Thus, the digestive tract model M1 is divided into a plurality of regions An (n is a natural number). Note that this region division is performed again when the division position (for example, the division interval) is changed by the user.

  Returning to FIG. 2, after the processing in step S4, the image classification unit 14 of the processing unit 4d classifies the image read by the acquisition unit 4c for each region An divided by the region dividing unit 13 (step S5). .

  In this step S5, based on the position information of the image with respect to the digestive tract model M1 obtained by the image position calculation unit 12, an image existing in the area An is extracted for each divided area An, and the area An is extracted. As an image group belonging to However, there may be a case where only one image exists or does not exist in the area. Note that this image classification is performed again when the above-described region division is performed again.

  Here, in the above-described image classification, when the imaging region F1 of a certain image belongs to a plurality of regions An (for example, when the region spans two regions An), it is classified into both or the region occupying the larger region An. The processing such as being performed is performed. The boundary condition for this classification may be arbitrarily specified by the user.

  After the processing of step S5, the image generation unit 16 of the processing unit 4d generates a first histogram image G1 that shows the images G1a arranged for each region An on the basis of the position in the extending direction of the digestive tract, and the display unit 4e The first histogram image G1 is displayed (step S6).

  In this step S6, for each area An divided by the area dividing unit 13, the image group classified into the area An is rearranged in the order of the position in the advancing direction of the digestive tract or in the time of imaging, and further the area An. A list image is generated by arranging each image group in the order in which each region An is arranged. That is, the position in the extending direction of the digestive tract is set on the horizontal axis, and the image group classified into each region An is arranged on the vertical axis. As shown in FIG. 6, the first histogram image G1 is generated.

  Here, FIG. 6 shows an example of the first histogram image G1, in which the images G1a for each region An are arranged in the vertical direction, and the image group for each region An is each region An in the horizontal direction. Are arranged in the order in which they are arranged. In FIG. 6, the direction from the left to the right is the traveling direction of the digestive tract. In the endoscopic examination screen of FIG. 6, a gastrointestinal model image G2 showing the gastrointestinal model M1 is shown together with the first histogram image G1. In this digestive tract model image G2, the locus L1 of the capsule endoscope 2, the division position of the digestive tract, and the like are also shown.

  Thereafter, the user performs an input operation on the operation unit 4f while looking at the above-described endoscopic examination screen, for example, operates the operation panel in FIG. 6 with a mouse or the like, specifies a feature to be confirmed, and further performs the first operation. A range (number of images) of images used for diagnosis is selected and designated from the histogram image G1. These designated values are read by the acquisition unit 4 c and used by the priority determination unit 15 and the image generation unit 16. In the above-described image range specification, for example, the user performs an input operation on the operation unit 4f to select an image to be directly displayed from the first histogram image G1, or the digestive tract on the digestive tract model image G2. It is possible to specify an arbitrary range in the model M1 and select an image to be displayed.

  Returning to FIG. 2, after the process of step S6, it is determined whether or not the target feature and image are specified (step S7), and the determination is repeated until the feature and image are specified (NO in step S7). . If it is determined that the target feature and image have been specified in accordance with the specification by the user (step S7), the priority determination unit 15 targets the specified image based on the specified feature. Feature extraction is performed, and the priority of the image is determined for each region in accordance with the feature amount (step S8).

  In step 8, the feature is extracted from the designated image, and the priority for displaying the image is determined according to the feature amount. For example, when the red color (color tone: red) used as an index of bleeding inside the body cavity is designated as a feature, the image priority of the image of each region An increases as the red color, which is a feature amount, increases. Highly determined. In other words, the priority is determined so that the image with the highest red condition is displayed in the foreground, and the images with the lower red condition are displayed in order from the foremost image to the back.

  Depending on the type of disease, the tissue may change color to yellow or become dark. Therefore, when it is desired to check the characteristics, the color tone yellow or black may be designated as the characteristic. In addition, as a feature of the shape, specify the contour of the tube contour on the image showing a cross section perpendicular to the extending direction of the gastrointestinal tract (such as a dent from a circle) as a feature, and the one with a lot of dent as an index of stenosis It is good also as an image with high priority. In addition, the residence time of the capsule endoscope 2 in the divided area An may be specified as a feature, and may be presented to the user as an area difficult to pass through the capsule endoscope 2 or an index of stenosis of the tube. When a plurality of features are designated, priority may be determined using a plurality of feature amounts by an arbitrary calculation method, and the degree may be presented by displaying a barometer image without rearranging the images. . However, the display of the barometer image is a display of the barometer image based on the position in the extending direction of the digestive tract, unlike the existing method of displaying the barometer image along the time axis.

  After the processing of step S8, a second histogram image G3 is generated that shows the images arranged in descending order of priority for each region on the basis of the position in the extending direction of the digestive tract, and the display unit 4e displays the second histogram image G3 is displayed (step S9).

  In this step S9, for each region An divided by the region dividing unit 13, the image group classified into the region An is rearranged in the order of the priority calculated by the priority determining unit 15, and the region An is further sorted. A list image is generated by arranging each image group in the order in which the regions An are arranged. That is, the position in the extending direction of the digestive tract is taken as the horizontal axis, and the image group classified into the regions An and arranged in order of priority is taken as the vertical axis Arranged in a histogram form, as shown in FIG. 7, a second histogram image G3 based on priority is generated. When a plurality of features are designated, the individual priority of the image group is determined for each feature, and then the image group in the region An is classified for each feature and the second histogram is obtained in the same manner as described above. An image G3 is generated.

  Here, FIG. 7 shows an example of the second histogram image G3. In the second histogram image G3, a display range exists for each of two different features (feature 1 and feature 2). There is a display range R1 related to the feature 1 and a display range R2 related to the feature 2. Such a display range is automatically set according to the number of specified features. The images G1a for each region An are arranged in the vertical direction in the order of priority based on the feature 1 in the display range R1, and are arranged in the vertical direction in the order of priority based on the feature 2 in the display range R2. Furthermore, the image group for each area An is arranged in the order in which the areas An are arranged in the horizontal direction. In FIG. 7, the direction from the left to the right is the advancing direction of the digestive tract, and an image having a higher priority for each feature is present in the foreground, and the lower the priority is, the deeper the image is located. .

  The endoscopic examination screen of FIG. 7 shows a gastrointestinal model image G2 showing the gastrointestinal tract model M1 together with the second histogram image G3 described above. Further, priority is given to the extending direction of the gastrointestinal tract. A barometer image G4 showing the degree or the number of images as a color density difference is shown. The gastrointestinal model image G2 is a virtual organ image showing the gastrointestinal model M1 as a virtual organ.

  Here, when two different features (feature 1 and feature 2) are designated, a barometer image G4 relating to either feature 1 or feature 2 is displayed, and when the user performs an input operation on the operation unit 4f. The barometer image relating to the other feature is switched to. This is effective when there is no space for displaying two barometer images at the same time on the endoscopic examination screen, and both the barometer images of feature 1 and feature 2 can be confirmed. Of course, instead of this switching display, if there is a space on the endoscopy screen for displaying both the feature 1 and feature 2 barometer images at the same time, both the feature 1 and feature 2 barometer images are displayed in advance. Anyway.

  Thereafter, the user performs an input operation on the operation unit 4f while viewing the above-described endoscopic examination screen, for example, operates the operation panel in FIG. This designated value is read by the acquisition unit 4 c and used by the image generation unit 16. In the designation of an arbitrary direction in the area An, for example, the user performs an input operation on the operation unit 4f, selects a group of images in the area An to be displayed from the second histogram image G3, and within the area An. Select an arbitrary direction.

  Returning to FIG. 2, after the process of step S9, it is determined whether or not an arbitrary direction in the area An is specified (step S10), and the determination is repeated until an arbitrary direction in the area An is specified (step S10). NO). If it is determined that an arbitrary direction has been specified in accordance with the above-mentioned specification by the user (YES in step S10), the image generation unit 16 combines the image group in the area An according to the specified arbitrary direction. A composite image G5 is generated, and the display unit 4e displays the composite image G5 together with the above-described second histogram image G3 (step S11), and returns the process to step S7.

  In this step S11, when the user designates the desired direction in the area An, the images in the desired direction are connected using the images existing in the area An. As shown in FIG. G5 is generated. At this time, the image to be used may be an image selected by the user, and when a plurality of images of the same area An are selected, an image having a high priority of the designated feature or a feature amount A composite image G5 may be created using pixels having a high height.

  Here, the endoscopic examination screen of FIG. 7 shows an example of a composite image G5 that is a diagnostic image, and a red feature T1. Furthermore, the feature quantity of feature 1 is shown as a numerical value in the composite image G5. This feature amount is a numerical value obtained by the priority determination unit 15. The composite image G5 is an image having a sense of depth, and the image can be enlarged. Note that the composite image G5 is regenerated in response to the change in the desired direction. For this reason, the composite image G5 functions as a panoramic image that shows the inner wall of an arbitrary region An 360 degrees around the core line of the digestive tract, so that the user can freely change the desired viewing direction and perform interpretation. is there.

  Thereafter, when the user performs an input operation on the operation unit 4f while looking at the endoscopic examination screen, for example, by operating the operation panel in FIG. 7 with a mouse or the like, the feature to be confirmed or the image used for diagnosis is changed. The processes after S7 are performed, and a new histogram image G3 and composite image G5 are generated and displayed. This re-display process is repeated according to the feature to be confirmed and the change of the image used for diagnosis. Further, when the user inputs an operation on the operation unit 4f to instruct the display end, the above-described image display process is ended.

  Thus, in the image display process, the second histogram image G3 is displayed by the display unit 4e together with the composite image G5 serving as a diagnostic image and presented to the user. That is, the priority of the image is determined based on the feature that the user wants to confirm, and the image that seems to be abnormal is displayed with priority. Furthermore, a composite image G5 that significantly shows the feature to be confirmed is also created and presented. As a result, the number of images to be interpreted can be reduced, and the feature to be confirmed and its value can be confirmed on one composite image G5. As a result, it is possible to perform a capsule endoscopy with less interpretation time required for interpretation and less labor burden on the interpreter.

  As described above, according to the present embodiment, the image groups in the region An are arranged according to the individual priorities of the image groups in the region An for each region An, and the image groups for each region An are further arranged. A second histogram image G3 is generated in the order of arrangement of the regions An, and the second histogram image G3 is displayed on the display unit 4e together with the composite image G5 serving as a diagnostic image. This makes it easy to check the feature region and feature amount in the diagnostic image, as well as confirming that the image used to generate the diagnostic image and the pathological findings outside the specified feature are not leaked. Therefore, the interpretation time required for interpretation and the labor burden on the interpreter can be reduced.

  In addition, a plurality of feature amounts of different types in the image are extracted for each image, the priority of each image group in the region An is determined according to the feature amount of the image for each region An and each type of feature amount, For each type of An and feature amount, the image groups in the region An are arranged according to the priority of the individual images in the image group in the region An, and the image groups for each region An are arranged in the order in which the regions An are arranged. By generating the histogram image G3, even when a plurality of different features are specified, the images are arranged according to the priority of the image for each feature, so that the interpretation efficiency can be improved. Time and labor of the reader can be reduced.

  In addition, a plurality of images classified into the same region An among the plurality of regions An are combined, and a combined image G5 in an arbitrary direction is generated and displayed, thereby generating a combined image G5 that clearly shows the specified feature Therefore, it is possible to improve the interpretation efficiency. As a result, it is possible to reduce the interpretation time and the labor burden on the interpreter. Note that it is desirable to create a composite image G5 in an arbitrary direction using an image having a higher priority.

  Further, the core line of the organ extending in the extending direction of the tubular organ is obtained based on the imaging position of the capsule endoscope 2 for each image, the tubular virtual organ whose axis is the obtained core line is obtained, and the obtained virtual organ is obtained. By dividing into a plurality of regions An along the extending direction, it becomes possible to perform division according to the actual organ path, so that more accurate region division can be realized.

  Moreover, the imaging region F1 of the capsule endoscope 2 with the imaging position as a reference is obtained for each image using the imaging position and the imaging posture of the capsule endoscope 2, and the imaging region of the capsule endoscope 2 for each obtained image. By determining the position of the image with respect to the organ in accordance with F1 and classifying a plurality of images for each region An, each image can be accurately classified into each region An, and more accurate interpretation is performed. be able to. As this classification method, in addition to using the imaging region F1 of the capsule endoscope 2, an imaging direction H1 based on the imaging position is obtained based on the imaging posture of the capsule endoscope 2, and the imaging position is obtained. An intersection of a straight line (imaging direction vector) extending in the imaging direction H1 from the inner wall of the tubular organ may be used. However, since the tubular organ has a large number of curved portions, the classification accuracy is higher when the above-described imaging region F1 is used.

  In addition, the residence time of the capsule endoscope 2 in the divided region An is designated as a feature, and the residence time is set as a position difficult to pass through the capsule endoscope 2 or an index of the stenosis of the organ. When the dwell time at each imaging position is displayed as a reference and presented to the user, the interpretation efficiency can be improved, and as a result, the interpretation time and the burden on the interpreter can be reduced. In the image display processing at this time, the processing unit 4d functions as a residence time calculation unit that calculates the residence time of the capsule endoscope 2 for each region An, and the image generation unit 16 follows the organ extension direction for each region An. A residence time image indicating the residence time is generated, and the display unit 4e displays the residence time image. Thereby, the residence time at each imaging position of the capsule endoscope 2 is displayed based on the position of the organ in the extending direction.

  In the above-described embodiment, the locus L1 and the division position of the capsule endoscope 2 are shown in the digestive tract model image G2 in addition to the digestive tract model M1, but the present invention is not limited to this. A portion where an image exists or does not exist on the tube model M1 may be indicated by display of color, gradation, pattern, or the like. In this case, since it is possible to easily grasp the imaging region and the non-imaging region for the digestive tract model M1, it is possible to improve the interpretation efficiency. As a result, the interpretation time and the labor burden on the interpreter are reduced. Can be reduced. Further, the current image range designated in step S7 may be shown on the digestive tract model M1 by displaying colors, gradations, patterns, and the like. In this case, since the specified current image range can be known at a glance, it is possible to easily grasp the interpretation target area, and to improve the interpretation efficiency. As a result, the interpretation time and the effort of the interpreter Burden can be reduced.

  Although one embodiment of the present invention has been described above, this embodiment is presented as an example and is not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 2 Capsule endoscope 4 Medical image display apparatus 4c Acquisition part 4d Processing part 4e Display part 13 Area division part 14 Image classification part 15 Priority determination part 16 Image generation part An area | region F1 Imaging area G1a Image G2 Gastrointestinal model image G3 Histogram Image G5 Composite image M1 Gastrointestinal model

Claims (9)

A plurality of images captured by an intra-organ imaging device that moves in a tubular organ and images in time series; an acquisition unit that acquires an imaging position and an imaging posture of the intra-organ imaging device for each of the images;
A region dividing unit that divides the organ into a plurality of regions along the extending direction;
An image classification unit that determines the position of the image with respect to the organ using the imaging position and the imaging posture of the intra-organ imaging device for each image, and classifies the plurality of images with respect to the plurality of regions;
A priority determining unit that extracts the feature amount of the image for each image and determines the priority of each image in the region according to the feature amount of the image for each region;
An image generation unit that generates a display image in which a plurality of images in the region are arranged according to the priority of each image in the region for each region;
A display unit for displaying the display image;
A medical image display device comprising:   The image generation unit arranges a plurality of images in the region according to the priority of each image in the region for each region as the display image, and sets the image group for each region in the order in which the regions are arranged. The medical image display apparatus according to claim 1, wherein the display images are arranged. The priority determination unit extracts a plurality of feature amounts of different types in the image for each image, and determines each image in the region according to the feature amount of the image for each region and each type of feature amount. Determine the priority,
The image generation unit arranges a plurality of images in the region according to the priority of each image in the region for each type of the region and the feature amount as the display image, and an image group for each region The medical image display device according to claim 1, wherein a display image in which the regions are arranged in the order in which the regions are arranged is generated. The image generation unit generates a composite image in an arbitrary direction by combining a plurality of images classified into the same region among the plurality of regions.
The medical image display apparatus according to any one of claims 1 to 3, wherein the display unit displays the composite image. A residence time calculation unit that calculates the residence time of the intra-organ imaging device for each region,
The image generation unit generates a residence time image indicating the residence time along the extending direction of the organ for each region,
The medical image display apparatus according to claim 1, wherein the display unit displays the residence time image.   The region dividing unit obtains a core line of the organ extending in the extending direction of the organ based on an imaging position of the intra-organ imaging device for each image, and obtains a tubular virtual organ having the obtained core line as an axis. 6. The medical image display apparatus according to claim 1, wherein the obtained virtual organ is divided into the plurality of regions along the extending direction.   The image classification unit obtains an imaging region of the intra-organ imaging device based on the imaging position using an imaging position and an imaging posture of the intra-organ imaging device for each image, and the organ for each obtained image The position of the said image with respect to the said organ is determined according to the imaging area | region of an internal imaging device, and the said some image is classified with respect to these several area | regions. Medical image display device. The image generation unit generates a virtual organ image indicating the virtual organ, the core line of the organ, and a division position of the virtual organ,
The medical image display apparatus according to claim 6, wherein the display unit displays the virtual organ image.   The image generation unit, as the virtual organ image, in addition to the virtual organ and the core line of the organ, the division position of the virtual organ, a virtual organ image indicating a location where the image exists or does not exist on the virtual organ The medical image display device according to claim 8, wherein the medical image display device is generated.