JP2010284405A - Medical image processor, medical image diagnostic device and medical image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To efficiently perform proper medical observation by enabling a user to recognize a region difficult to observe in a generated image for observation. <P>SOLUTION: An image processing part 27 generates an image for endoscopic display of a form of an inner wall surface of an intestine based on volume data representing the form of the intestine. The image processing part 27 imparts an evaluation point representing the difficulty in observing in the above image to each of a plurality of intestine inner wall surface voxels corresponding to the intestine inner wall surface. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a medical image processing apparatus and a medical image diagnostic apparatus having a function of generating a morphological image of the inner wall surface of a tubular organ based on volume data, and a medical image processing program for causing a computer to realize the above functions. .

  Technology for generating images similar to observing the inner wall surface of a tubular organ with an endoscope based on volume data collected by a computed tomography (CT) diagnostic imaging device or magnetic resonance imaging (MRI) diagnostic imaging device It has been known.

  For example, in a colon examination (CT Colonography, hereinafter referred to as CTC) using a CT image diagnostic apparatus, endoscopic display using a perspective projection method (PVR) or fish-eye projection method, and in addition to the intestinal lumen surface Intestinal screening for colorectal cancer detection is performed by using an expanded display.

  In the endoscopic display of the intestinal tract, in order to prevent overlooking of the cancer due to curvature of the intestinal tract and meniscus, it is also observed to reciprocate through the intestinal tract. In addition, oversight occurs due to the shape of the intestinal tract in reciprocal observation, so that the display of undisplayed areas (areas that did not enter the field of view) on the inner wall of the intestinal tract when endoscopic display is performed, Setting may also be performed (see Non-Patent Document 1).

  The problems associated with the above-mentioned prior art in the large intestine screening test by CTC are described below.

  (1) In endoscopic display by PVR, the viewing angle is generally set to around 90 degrees, and an unnatural image is produced even if the angle is too large or too small. When the viewing angle is 90 degrees, the other range of 270 degrees is a blind spot. In particular, in the case of the large intestine, as shown in FIG. 17, the meniscus is raised in a direction perpendicular to the line of sight, so the back side of the meniscus tends to be a blind spot. In order to prevent oversight of cancer due to this blind spot, reciprocal observation of the intestinal tract is performed. there were. In addition, by reciprocating the intestinal tract, problems such as increased burden on the observer and decreased throughput also occur.

  (2) Endoscopic display by fisheye projection may be performed to deal with blind spots of PVR. Endoscopic display by the fisheye projection method has an advantage that a display image similar to that of an actual endoscope can be obtained. However, although the field of view is wide, there is a possibility that a small polyp may be overlooked due to the large distortion, and depending on the shape of the intestinal tract and the size, orientation, and position of the meniscus, for example, as shown in FIG. Blind spots may occur and do not completely prevent oversight in screening.

  (3) The intestinal inner wall surface development display can give a bird's-eye view of the entire intestinal inner wall, but because of the large distortion, there is a risk that the spherical polyp will be stretched and look like a meniscus. In addition, if a reduced display is used to look down on the entire intestinal tract, it will be easy to miss a small polyp.

  (4) It seems that the oversight of a polyp of a certain size can be prevented by displaying a non-display area and using a line-of-sight movement route that does not generate a non-display area. However, micropolyps and flat cancers are not always detected only when the wall surface enters the display field of view. A micropolyp is difficult to detect if the projection size on the display screen is small. In the case of PVR or fish-eye projection, the projection size on the screen depends on the distance from the viewpoint to the polyp. In addition, detection may be difficult depending on the angle between the projection direction of the polyp and the line-of-sight direction. Thus, for example, in the situation shown in FIG. 19, it is difficult to correctly identify the polyp.

  (5) Since flat cancer is difficult to detect only by displaying the surface of the inner wall of the intestinal tract, as shown in FIG. 21, the frontal clipping is used together with the front clipping as shown in FIG. In some cases, the accuracy of cancer detection may be improved. In this case, there is a possibility that an appropriate cross-sectional angle cannot be obtained with respect to the intestinal tract and the cancer may be overlooked due to the curvature of the intestinal tract and the core wire. Specifically, if the intestinal tract and the core wire are in the state shown in FIG. 22, the MPR surface can be appropriately set for cancer and the cancer detection accuracy can be improved. However, for example, if the intestinal tract and the core wire are in the state shown in FIG. 23, the MPR surface may not be appropriate for cancer, and the cancer detection accuracy may not be improved.

Yuichiro Hayashi et al., “Development of detection function for unpresented area in virtualized endoscope system”, MEDICAL IMAGING TECHNOLOGY Vol.20 No.5 September 2002

  As described above, in the prior art, an image suitable for observing the structure of the inner wall surface of the tubular organ cannot be generated, and the observer may miss a lesion to be detected. It was.

  The present invention has been made in view of such circumstances, and the object of the present invention is to allow a user to recognize a region that is difficult to observe in the generated image for observation. It is to enable appropriate medical observation to be performed efficiently.

  The medical image processing apparatus according to the first aspect of the present invention includes a morphological image generating means for generating a morphological image representing the morphology of the inner wall surface of the organ based on the volume data representing the morphology of the tubular organ, and the organ And an assigning means for assigning an evaluation point representing difficulty in viewing in the morphological image with respect to each of a plurality of evaluation regions set by dividing the inner wall surface into a plurality of regions.

  According to a second aspect of the present invention, there is provided a medical diagnostic imaging apparatus that collects volume data representing the shape of a tubular organ in a subject, and forms the shape of the inner wall surface of the organ based on the volume data. Giving a morphological image generating means for generating a morphological image to be expressed and an evaluation point indicating difficulty in viewing the morphological image for each of a plurality of evaluation regions set by dividing the inner wall surface of the organ into a plurality of portions Means.

  According to a third aspect of the present invention, there is provided a medical image processing program that reads a computer capable of accessing a storage medium storing volume data representing a tubular organ form, reads the volume data from the storage medium, and stores the volume data in the volume data. Morphological image generation means for generating a morphological image representing the morphology of the inner wall surface of the organ on the basis thereof, and the appearance of the morphological image for each of a plurality of evaluation regions set by dividing the inner wall surface of the organ into a plurality of parts It is made to function as an assigning means for assigning an evaluation point representing the difficulty.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to make a user recognize the area | region where observation becomes difficult in the produced | generated image for observation, and, thereby, appropriate medical observation can be implemented efficiently.

The figure which shows the structure of the principal part of CT apparatus which concerns on one Embodiment of this invention. The block diagram which shows the function of the memory | storage part in FIG. 1, and an image process part. The flowchart which shows the process sequence at the time of implementing the large intestine screening test | inspection by the image process part in FIG. The figure which shows the concept of the process which extracts an intestinal tract inner wall segment from large intestine segment data. The figure which shows the mode of the evaluation of the frequency | count or accumulation time which entered the visual field of endoscopic display. The figure showing the angle of the gaze direction and normal direction in endoscopic display. The figure showing the distance on the screen from the projection screen center in an endoscopic display to the intestinal tract inner wall surface voxel. The figure which shows an example of the clipping surface suitable for observation. The figure which shows an example of the clipping surface unsuitable for observation. The figure showing the distance on the screen from an observer's eyes | visual_axis position with respect to the projection screen in an endoscopic display to the intestine inner-wall surface voxel. The figure which shows an example of a mode that the residue in an intestinal tract exists in an intestinal tract. The figure which shows an example of an evaluation image. The figure which shows the 1st specific example which determines the viewpoint and line-of-sight direction for additional observation. The figure which shows the 2nd specific example which determines the viewpoint and line-of-sight direction for additional observation. The figure which shows the 3rd specific example which determines the viewpoint and line-of-sight direction for additional observation. The figure which shows the 3rd specific example which determines the viewpoint and line-of-sight direction for additional observation. The figure which shows the problem in the colon screening test | inspection by CTC. The figure which shows the problem in the colon screening test | inspection by CTC. The figure which shows the problem in the colon screening test | inspection by CTC. The figure which shows the problem in the colon screening test | inspection by CTC. The figure which shows the problem in the colon screening test | inspection by CTC. The figure which shows the problem in the colon screening test | inspection by CTC. The figure which shows the problem in the colon screening test | inspection by CTC.

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

  FIG. 1 is a diagram illustrating a configuration of main parts of a CT apparatus 100 according to the present embodiment. This CT apparatus includes a scan gantry 1 and a computer apparatus 2. The scan gantry 1 is a component for collecting projection data related to a subject. The projection data collected by the scan gantry 1 is used for processing such as image reconstruction in the computer device 2.

  The scan gantry 1 includes a bed 10, an X-ray tube device 11, an X-ray detector 12, a gantry rotation drive unit 13, a high voltage generation unit 14, a bed drive unit 15, and a data collection unit 16.

  In the scan gantry 1, the X-ray tube device 11 and the X-ray detector 12 are mounted on an annular rotary mount (not shown) in an opposing relationship. The rotating gantry is driven to rotate by the gantry rotation driving unit 13. At this time, the X-ray tube device 11 and the X-ray detector 12 rotate around the same rotation axis RA. The scan gantry 1 forms a cavity (imaging port) inside the rotation trajectory of the X-ray tube device 11 and the X-ray detector 12.

  The X-ray tube device 11 includes an X-ray tube 11a and an X-ray filter 11b. The X-ray tube 11 a receives power supply from the high voltage generator 14 and emits X-rays toward the X-ray detector 12. The X-ray filter 11b removes low energy components to reduce exposure. The high voltage generator 14 includes a high voltage transformer, a filament current generator, and a rectifier. In addition to this, the high voltage generator 14 includes a tube voltage switch, a filament current switch, and the like in order to adjust the tube voltage and the filament current arbitrarily or stepwise.

  The X-ray detector 12 is configured by arranging a plurality of X-ray detection element arrays in a direction along the rotation axis RA. The X-ray detector 12 detects incident X-rays and outputs an electrical signal corresponding to the intensity.

  The subject is placed on the top 10 a of the bed 10. The bed 10 is driven by the bed driving unit 15 to move the top plate 10a in the longitudinal direction (left and right direction in FIG. 1). Usually, the bed 10 is installed such that the longitudinal direction thereof is parallel to the rotation axis RA. Usually, the subject is placed on the top board 10a so that the body axis is along the rotation axis RA. Thus, the subject is inserted into the cavity of the scan gantry 1 as the top plate 10a moves.

  The data collection unit 16 collects the output of the X-ray detector 12 and supplies it to the computer apparatus 2 as raw data. Note that an interface using slip ring, optical communication, or the like is interposed between the X-ray detector 12 and the data collection unit 16. As a result, the data collection unit 16 can collect the output of the X-ray detector 12 while continuously rotating the rotary mount.

  The computer apparatus 2 includes a gantry control unit 21, a preprocessing unit 22, an image reconstruction unit 23, a display unit 24, a console 25, a storage unit 26, an image processing unit 27, and a line-of-sight detection unit 28. These gantry control unit 21, preprocessing unit 22, image reconstruction unit 23, display unit 24, console 25, storage unit 26, image processing unit 27, and line-of-sight detection unit 28 are mutually connected via a data / control bus 29. It is connected.

  The raw data supplied from the scan gantry 1 to the computer apparatus 2 is sent to the storage unit 26 via the preprocessing unit 22 and stored therein. The image reconstruction unit 23 generates volume data based on the raw data. Volume data is stored in the storage unit 26.

  The console 25 is provided for the operator to input various information such as scanning conditions and various instructions. The console 25 includes an operation screen.

  The gantry control unit 21 controls the operation of the scan gantry 1 so that scanning is performed so that projection data necessary for diagnosis is obtained.

  The line-of-sight detection unit 28 detects the position of the observer's line of sight on the display screen of the display unit 24. For the detection of the line-of-sight position, a conventional technique such as scanning the position and direction of a human pupil or cornea to obtain the line-of-sight position on the screen is used. The line-of-sight detection unit 28 outputs line-of-sight position information indicating the line-of-sight position.

  FIG. 2 is a block diagram illustrating functions of the storage unit 26 and the image processing unit 27. The storage unit 26 includes a CT volume / raw data storage function 26a, a display condition storage function 26b, an extracted segment management / storage function 26c, a colon core line management / storage function 26d, an intestinal lining position information management / storage function 26e, and an evaluation / display state. A management / storage function 26f is included. The image processing unit 27 includes a display function / display state output function 27a, a large intestine extraction function 27b, a large intestine core line extraction function 27c, an intestinal inner wall voxel extraction function 27d, a display condition setting function 27e, a display state evaluation function 27f, and a three-dimensional clipping function. 27g, a display position / angle calculation function 27h, a colon core line correction function 27i, a colon CAD function 27j, and a line-of-sight tracking function 27k.

  The storage unit 26 can be realized by appropriately using a storage device such as a semiconductor memory or a hard disk device, or a removable recording medium such as a magnetic disk, a magneto-optical disk, or an optical disk. Each function of the storage unit 26 can be realized as a storage area physically or virtually set in the storage device or the recording medium. The image processing unit 27 can use, for example, a general-purpose processor as basic hardware. Each of the above functions can be realized by executing the above processor image processing program.

  The CT volume / raw data storage function 26a stores raw data and volume data.

  The display function / display state output function 27a includes volume rendering by parallel projection or PVR, display of the inner wall of the large intestine, MPR display on the clipping plane of the intestine, CPR (curved planar reconstruction) along the intestinal core line, The display unit 24 is controlled for display by various renderings such as SPR (stretched CPR). In addition, the display function / display state output function 27a controls the display unit 24 to perform overlay display of characters and markers on the rendered image. Furthermore, the display function / display state output function 27a also has a display state output function for evaluating the display state of the inner wall of the intestinal tract.

  The display condition storage function 26b manages and stores display conditions such as direction, position, opacity, color, magnification, or cross-sectional position.

  The large intestine extraction function 27b extracts the large intestine included in the collection range of the volume data based on the volume data. More specifically, the large intestine extraction function 27b includes the inner wall segment data only for the surface portion of the intestinal inner wall for endoscopic display, and the lumen segment data of the intestinal lumen area used for large intestine appearance display and core line extraction. Are extracted respectively. Note that when the volume data is collected, gas is injected into the large intestine to increase the contrast between the intestinal tract and the gas, so that the large intestine can be easily extracted by threshold processing.

  The extracted segment management / storage function 26c manages and stores the inner wall segment data and the lumen segment data.

  The large intestine core line extraction function 27c extracts the large intestine core line based on the lumen segment data, and generates large intestine core line information representing the form and position of the large intestine core line.

  The large intestine core wire management / storage function 26d manages and stores large intestine core wire information.

  The intestinal inner wall surface voxel extraction function 27d extracts voxels (hereinafter referred to as intestinal inner wall surface voxels) corresponding to the surface of the intestinal inner wall as an evaluation target of the observation state.

  The intestinal inner wall position information management / storage function 26e stores the position information of the intestinal inner wall surface voxels. For the storage of the position information, a method of holding spatial coordinates as bit data or a method of holding them in a coordinate table can be used.

  The evaluation / display state management / storage function 26f stores the result of evaluating each intestinal inner wall surface voxel by display or prior navigation.

  The display condition setting function 27e sets display conditions based on the evaluation result for each intestinal inner wall surface voxel.

  The display state evaluation function 27f evaluates each intestinal inner wall surface voxel by display or prior navigation.

  The three-dimensional clipping function 27g clips (does not display) a part of the three-dimensional object. The three-dimensional clipping function 27g is used to display in advance limited to the target organ or to change the clipping position according to the display position. In the present invention, in particular, in endoscopic display, it is used as front clipping for clipping at a plane at a certain distance from the viewpoint, and used for displaying a cross section of the intestinal tract.

  The display position / angle calculation function 27h calculates a display (viewpoint) position and a display angle direction based on the result of evaluating each intestinal inner wall surface voxel by display or prior navigation.

  The large intestine core line correction function 27i provides a core line parallel to the inner wall surface for observing a cross-section clipped at an appropriate angle with respect to a place where the clipping angle is evaluated to be in an appropriate state after observation of the large intestine or after prior evaluation. calculate.

  The colorectal CAD function 27j detects a colorectal cancer candidate by a well-known CAD (computer-aided detection) technique and presents its position. In this embodiment, when an evaluation point is added to the position of the candidate obtained by the large intestine CAD function 27j, or the degree of probability of the candidate is provided in addition to the CAD function, the evaluation point corresponding to the degree is provided. It is used to increase the coverage rate in screening tests by adding.

  The line-of-sight tracking function 27k tracks the line-of-sight position of the observer based on the line-of-sight position information obtained by the line-of-sight detection unit 28. The result of this tracking is used to evaluate whether or not the observer can observe the large intestine inner wall surface uniformly.

  Next, the operation of the CT apparatus 100 configured as described above will be described.

  FIG. 3 is a flowchart showing a processing procedure of the image processing unit 27 when performing a large intestine screening test.

  Prior to performing the large intestine screening test, the large intestine region of the subject is scanned with the scan gantry 1. This scan is usually performed in a state in which intestinal lavage is performed in advance using a laxative, an antispasmodic agent such as buscopan is administered to the patient to suppress built-in convulsions, and gas is injected into the large intestine. The raw data collected by this scan is stored by the CT volume / raw data storage function 26a. Further, the image reconstruction unit 23 generates volume data based on the collected raw data. This volume data is also stored by the CT volume / raw data storage function 26a.

  In step Sa1, the image processing unit 27 uses the large intestine extraction function 27b to collect a set of voxels contained in the large intestine area (the large intestine segment from the volume data stored by the CT volume / raw data storage function 26a as described above. ) Data is extracted. The large intestine region represents an air region in the intestinal tract. The large intestine segment data representing the large intestine segment extracted here is stored by the extracted segment management / storage function 26c.

  In step Sa2, the image processing unit 27 uses the large intestine core line extraction function 27c to extract the large intestine core line as the central line of the intestinal tract from the large intestine segment data generated in step Sa1, and the large intestine core line information representing the form and position of the large intestine core line. Is generated. As a process for obtaining the center line of the intestinal tract, a known thinning process can be used. The large intestine core wire information is stored by the large intestine core wire management / storage function 26d. Colon core information is used for navigation of an endoscopic display of the intestinal lumen.

  In step Sa3, the image processing unit 27 uses the intestinal inner wall surface voxel extraction function 27d to extract a set of voxels (intestinal inner wall segment) corresponding to the intestinal inner wall from the large intestine segment data generated in step Sa1. The position information of the intestinal inner wall surface voxels included in the intestinal inner wall segment is stored by the intestinal inner wall position information management / storage function 26e.

  FIG. 4 is a diagram showing a concept of processing for extracting an intestinal inner wall segment from large intestine segment data.

  The intestinal inner wall segment 41 can be extracted by subtracting the large intestine segment 43 from the large intestine segment 42 that has been subjected to the expansion process. However, the expansion process is a process of expanding the thickness of the designated number of voxels (expansion size) from the surface of the object. Since the large intestine segment is an air region extracted from the intestinal tract, by setting the expansion size to “1”, a set of voxels (intestinal inner wall surface voxels) corresponding to the inner wall surface is extracted. In order to obtain a more accurate intestinal inner wall surface voxel, the expansion size is set to “2” or more, and the inner wall is extracted without passing through, and the opacity is high in the extracted voxel group (the voxel value is 0. 0). 8 or more), and voxels containing air voxels in adjacent voxels (in contact with the air surface) may be intestinal inner wall surface voxels. The adjacent voxels are typically six voxels adjacent in the vertical and horizontal directions, or 26 voxels obtained by adding 20 adjacent voxels obliquely thereto. In addition, there are cases where the evaluation target voxels in step Sa5 described later are reduced in order to reduce the data size and increase the processing speed. As a reduction method, there is a method in which the coordinates or the position of the storage table is limited to a multiple of a certain constant.

  In step Sa4, the image processing unit 27 uses the display function / display state output function 27a to start screening in the intestinal tract by endoscopic display for the intestinal inner wall surface voxel extracted in step Sa3. Specifically, the display function / display state output function 27a is a surface or cross section of the inner wall of the large intestine when viewing a range within a predetermined visual field angle centered on a predetermined visual axis direction from an arbitrary viewpoint position on the large intestine core line. Is generated and displayed on the display unit 24. At this time, the viewpoint position may be automatically moved along the large intestine core line, or may be moved according to an instruction from the user. The center of the line-of-sight direction is basically the direction along the large intestine core line, but may be a direction inclined from the large intestine core line. At this time, the angle formed by the center of the line-of-sight direction and the direction along the large intestine core line may be determined according to a change instruction from the user, or a preset inclination angle is commonly applied to a plurality of visual field positions. You may decide. That is, the conventional PVR technique can be used as it is. It is also possible to apply various techniques conventionally used in CTC, such as endoscopic display by a fish-eye projection method and unfolded display of the intestinal lumen surface.

  In step Sa5, the image processing unit 27 uses the display state evaluation function 27f to start providing evaluation points for the intestinal inner wall surface voxels extracted in step Sa3. Here, the evaluation score is an index value indicating whether or not the observer has sufficiently observed each intestinal inner wall surface voxel based on screening. In this embodiment, one or a plurality of evaluation points for one intestinal inner wall surface voxel using the values of the items shown in the following (a) to (g) individually, a plurality, or a combination of a plurality of values. give.

  Although the calculation rule of the evaluation points may be arbitrary, in this embodiment, for example, an initial value “100” is given to each intestinal inner wall surface voxel, and the evaluation value of each intestinal inner wall surface voxel is appropriately set by the following evaluation. Decrease to such an extent that it can be evaluated as observed. Therefore, in this embodiment, the evaluation is such that a voxel having a higher evaluation score after screening is more likely to be overlooked. It should be noted that the evaluation value of the intestinal inner wall surface voxel that could not be properly observed can be increased.

(A) Number of times or cumulative time of entering endoscopic display field of view In the above screening, the display image is updated each time the field of view is changed in accordance with changes in the viewpoint position, the field of view direction, and the like. Therefore, every time the field of view is changed, the number of times of the intestinal inner wall surface voxels in the field of view after the change is counted up. For example, in response to the display image for the visual field shown in FIG. 5 being displayed, the number of times for voxels B3 to B9, B15, and B16 is counted up one by one, but voxels B1, B2, B10 to B14, B17 are counted. , B18 does not count up. The number of times counted by this corresponds to the number of times that each intestinal inner wall surface voxel has entered the field of view of the endoscopic display during screening, and the smaller the number, the higher the possibility of oversight. However, when the time period when the visual field is changed is indefinite, the time during which the image including the intestinal inner wall surface voxel is displayed is accumulated for each intestinal inner wall surface voxel. The time required in this way corresponds to the cumulative time that each intestinal inner wall surface voxel has entered the field of view of the endoscopic display during screening, and the smaller this is, the higher the possibility of oversight. Then, the evaluation score is subtracted by a value obtained by scaling with a certain fixed value with respect to the number of times or the accumulated time.

(B) Angle between viewing direction and normal direction in endoscopic display In order to detect a minute polyp, the viewing angle with respect to the polyp may be an important factor. Therefore, as shown in FIG. 6, the angle θ1 between the normal direction of the intestinal inner wall surface at the position of a certain intestinal inner wall surface voxel and the viewing direction to the intestinal inner wall surface voxel is calculated. Then, the smaller the direction vector of the angle θ1 or the inner product value of the unit vectors, the more difficult the detection is, and the evaluation value is subtracted by a value corrected with a bias value (for example, 0.5) and a scaling value.

(C) Distance from viewpoint position in endoscopic display to intestinal inner wall surface voxel In PVR and fisheye projection, the size of each intestinal inner wall surface voxel projected on the screen is from the viewpoint to each intestinal inner wall surface voxel Depending on the distance, the smaller the distance, the easier it is to detect a micropolyp. Therefore, the distance D1 shown in FIG. 6 is calculated, and the evaluation score is subtracted by a value scaled by a certain fixed value with respect to the value of D1.

(D) Distance on the screen from the projection screen center to the intestinal inner wall surface voxel in the endoscopic display In the endoscopic display, the surface and cross section of the large intestine wall are observed while moving the viewpoint position. Because the displayed image changes every moment, there is a possibility of overlooking cancer. In particular, a portion displayed at the edge position of the screen is likely to be overlooked. Therefore, as shown in FIG. 7, the distance D2 on the screen from the screen center to the screen projection position of the intestinal inner wall surface voxel is calculated, and the evaluation point is subtracted by a value obtained by scaling the value of D2.

(E) Clipping angle of the intestinal inner wall surface voxel at the clipping position with respect to the intestinal inner wall surface In the endoscopic display, the front side is clipped (erased) in order to detect cancer or to see the degree and range of cancer invasion, In some cases, the MPR is displayed on the section after clipping. This clipping is performed by the three-dimensional clipping function 27g, and the clipping position and angle information is sent from the three-dimensional clipping function 27g to the display state evaluation function 27f. As shown in FIG. 8, the clipping plane is preferably perpendicular to the inner wall surface of the intestinal tract. However, depending on the shape of the intestinal lining and the shape of the intestinal tract, an appropriate cross section cannot be obtained as shown in FIG. May interfere. Therefore, the angle θ2 formed by the normal direction of the inner wall of the intestinal tract and the clipping direction is calculated, and the evaluation score is subtracted by a value obtained by scaling the value of θ2.

(F) Distance on the screen from the observer's line-of-sight position to the projection screen in the endoscopic display to the intestinal inner wall surface voxel The above (d) is based on the premise that the center of the screen is often observed Based. However, depending on the observer, there may be a case where a portion other than the central portion of the screen is intensively observed. Therefore, in order to consider the actual observation position of the observer, a distance D3 on the screen from the observer's line-of-sight position detected by the line-of-sight detection unit 28 to the screen projection position of the intestinal inner wall surface voxel as shown in FIG. And the evaluation score is subtracted by a value obtained by scaling the value of D3.

  Since the observer's line-of-sight position moves frequently, the detection result of the line-of-sight detection unit 28 changes frequently. Therefore, the line-of-sight position used for calculating the distance D3 may be determined as the center position of the movement range including a plurality of detection results as shown in FIG.

(G) Certainty factor of colorectal cancer candidate obtained by large intestine CAD function The large intestine CAD function 27j extracts polyp-shaped ones mainly from the inner wall surface of the intestinal tract by well-known processing and makes it a colon cancer candidate. Further, the large intestine CAD function 27j can also calculate the degree of polyp (confidence) based on the shape and size of the extracted colon cancer candidate. In view of this, the presence or absence of a colorectal cancer candidate detected by the colorectal CAD function 27j or the certainty of the colorectal cancer candidate is converted into an evaluation score. Specifically, for example, in CAD, one point indicating a representative position of a polyp or the position of the entire polyp is calculated, and therefore, a constant value is subtracted from the evaluation point for the inner wall voxel corresponding to the calculated position. . Further, when the degree of polyp (confidence) is calculated for each of the positions calculated as described above in CAD, the value (scaling value) corresponding to the certainty can be subtracted from the evaluation score. good. The large intestine CAD is performed prior to the endoscopic display of the large intestine screening and the intestinal inner wall surface expansion display.

  The evaluation points obtained for each of the inner wall surface voxels of the large intestine are stored by the evaluation / display state management / storage function 26f.

  By the way, in the intestinal tract, there may be a residue in the intestinal tract as shown in FIG. This intestinal residue is desirably removed in order to observe the intestinal surface, but screening can also be performed without removing the intestinal residue. In this case, since the residue surface is extracted as the inner wall surface, the residue surface is excluded from the evaluation target. The extraction of the residue can be performed alone or in combination with threshold processing and shape recognition of the residue liquid surface (plane).

  In parallel with the above screening and evaluation point assignment, in step Sa6, the image processing unit 27 waits for the screening to end. If the screening is completed, the image processing unit 27 proceeds from step Sa6 to step Sa7.

  In step Sa7, the image processing unit 27 can use the display function / display state output function 27a to overlook any part of the intestinal tract by the above-described screening based on the evaluation points assigned to the inner wall surface voxels of each intestinal tract. An evaluation image representing whether the property is high is generated, and this evaluation image is displayed on the display unit 24.

  FIG. 12 is a diagram illustrating an example of an evaluation image. The table or image shown in FIG. 12 shows the marker M at one place where the possibility of oversight on the two-dimensional image representing the structure of the appearance surface of the intestinal tract is high.

  Thus, when a marker is shown only at one place, for example, each intestinal inner wall surface voxel may be sequentially selected according to a predetermined rule, and the marker may be indicated at the position of the selected intestinal inner wall surface voxel. Although the above rules may be arbitrary, for example, the evaluation score is in descending order. Alternatively, it is also possible to display a list showing the evaluation points of the intestinal inner wall surface voxels so that the evaluation points can be confirmed, and to make the order of user instructions based on this list.

  A plurality of markers may be simultaneously shown in the evaluation image. For example, if there are a plurality of intestinal inner wall surface voxels whose evaluation points are equal to or greater than a threshold value, a marker is indicated at each position of the intestinal inner wall surface voxels. Alternatively, a marker is shown at each position of the upper intestinal inner wall surface voxels in descending order of evaluation score. Furthermore, a list showing the evaluation points of each intestinal inner wall surface voxel is displayed, and the intestinal inner wall surface voxel designated by the user on this list or the list after editing according to the user instruction It is also conceivable to show a marker at each position of the intestinal inner wall surface voxel included in the predetermined range.

  The marker may change at least one of its size, color, and shape according to the evaluation point.

  The evaluation image may be an image obtained by coloring each intestinal inner wall voxel according to the evaluation point. In the coloring, one or more of hue, saturation, and brightness are made different depending on the value of the evaluation point.

  The evaluation image may be based on an image representing a form such as the intestinal inner wall surface or the intestinal core wire. When coloring an image representing the form of the intestinal tract core line, the maximum and average values of the intestinal inner wall surface voxel group located at the shortest position with respect to the point on the intestinal tract core line or the intestinal inner wall surface voxel group positioned in the vertical plane of the intestinal tract core line Or, set the color according to the type of evaluation.

  Evaluation images are grouped into blocks connecting intestinal inner wall surface voxels with evaluation points above a certain threshold value, and markers are displayed and colored according to the maximum value, average value, or evaluation type for each block. May be generated.

  Furthermore, the evaluation image may be displayed only on the near side after the hidden surface is erased, or may be the entire visualization including the object on the back side.

  Various conditions for the evaluation image as described above may be arbitrarily applied, and information indicating the display condition of the evaluation image is stored by the display condition storage function 26b. The information indicating the default display conditions may be fixedly stored by the display condition storage function 26b, or information indicating the display conditions set by the display condition setting function 27e according to the user's support, for example. You may store by the display condition storage function 26b.

  In step Sa8, the image processing unit 27 starts rescreening. This re-screening is a screening for additionally observing a portion having a high evaluation score and a high possibility of being overlooked, and is performed as follows, for example.

  The first example is a method of screening an additional observation position designated by the observer based on the evaluation image. When an observer designates a position on an image such as an intestinal appearance image, an endoscopic display of the intestinal tract, an intestinal inner wall surface development display, or the like, a point on the inner wall surface side, or a point on the core line closest to the voxel, or Endoscopic display is started with a viewpoint at a point on the core line perpendicular to the core line. Furthermore, there is a case where the line-of-sight direction is directed from the viewpoint position to the position designated by the observer or the voxel having the maximum evaluation point within a certain range. As this range, designation within a certain distance or a field of view is used.

  The second example is a method of performing automatic movement for additional observation of a position with a high evaluation score. The intestinal inner wall surface voxel or the above-mentioned block whose evaluation score is equal to or greater than a certain threshold is set as an additional observation position. Examples of the selection and order designation method when there are a plurality of additional observation positions include the following.

  (A) List all additional observation voxels or blocks and select them in that order. However, the list also displays the evaluation items applied during the process in step Sa5, their evaluation points, and the like. The list can also be reordered, sorted, and narrowed down by sorting and filtering.

  (B) Selection in the order from the anal side or vice versa (forward direction or reverse direction of the core wire).

  (C) Selection in order of highest evaluation score.

  When the observer finishes observing a certain location, the observer instructs movement to the next observation location. In response to this instruction, the image processing unit 27 changes the additional observation position to be subjected to endoscopic display.

  By the way, when the additional observation position is displayed endoscopically, if the viewpoint and the line-of-sight direction are the same as those in the screening at step Sa4, an image suitable for observation of the additional observation position may not be obtained. Therefore, for example, it is preferable to determine the viewpoint and the line-of-sight direction as follows.

  FIG. 13 is a diagram illustrating an example of determining the viewpoint and the line-of-sight direction for a block. First, a representative point of a block to be subjected to additional observation is obtained as an average position of coordinates of all voxels of the block, a center of gravity position, or a center of a circumscribed circle. The viewpoint is the point on the intestinal tract that is closest to the block representative point, and the direction from the viewpoint toward the block representative point is the line-of-sight direction. When the evaluation point of the distance from the viewpoint to the voxel is high, a certain position on the line from the viewpoint to the intestinal inner wall surface voxel on the line of sight line, for example, a position of 30% of the distance from the intestinal inner wall surface voxel to the line Processing such as moving the viewpoint may be performed.

  FIG. 14 is a diagram illustrating an example in which the viewpoint and the line-of-sight direction are determined when the evaluation point of the angle between the line-of-sight direction and the normal direction of the intestinal inner wall surface voxel is high. As shown in FIG. 14, in the endoscopic display of the viewpoint along the intestinal tract and the gaze direction, since the angle to the intestinal inner wall surface voxel is shallow, the point on the intestinal tract closest to the intestinal inner wall surface voxel The viewpoint is moved to and the line of sight is directed toward the voxel on the inner wall of the intestinal tract from there.

  15 and 16 are diagrams illustrating an example in which the viewpoint and the line-of-sight direction are determined when the evaluation score based on the clipping angle is high. As shown in FIG. 15, the combination of the intestinal tract shape and the intestinal tract core shape results in an intestinal inner wall surface voxel in which the clipping plane is not perpendicular to the intestinal tract. Here, these are the blocks to be additionally observed. In order to obtain an optimal clipping angle at the position of each intestinal inner wall surface voxel of the target block, an additional observation core wire is provided. The additional observation core wire is obtained by projecting the original core wire toward the inner wall surface as shown in FIG. For example, a single smooth curve is obtained by spline processing etc. based on a point group in which the points on the core line closest to each intestinal inner wall surface voxel included in the target block are moved in the respective voxel direction, This is the additional observation core wire. At this time, the additional observation core wire may be moved away from the inner wall surface by a certain distance by moving the voxel to a point at a certain distance. Then, endoscopic display and MPR display on the clipping plane are performed along the additional observation core line.

  In order to determine a proper viewpoint and line-of-sight direction by looking at a plurality of situations as in the above three specific examples, the evaluation result for each evaluation item in the display state evaluation function 27f may be referred to. For example, when the deduction amount accompanying the evaluation of the clipping angle with respect to the evaluation point of the intestinal inner wall surface voxel to be subjected to the additional observation is less than or equal to the prescribed amount, the above third example may be taken.

  The determination of the viewpoint and line-of-sight direction as described above is performed by the display position / angle calculation function 27h, and information indicating the determined viewpoint and line-of-sight direction is stored by the display condition storage function 26b. Further, the additional observation core wire is calculated by the large intestine core wire correction function 27i, and information indicating this is stored by the large intestine core wire management / storage function 26d. Based on the information stored by the display condition storage function 26b and the large intestine core management / storage function 26d as described above, the display function / display state output function 27a performs display processing for additional observation. Used for endoscopic display.

  In step Sa9, the image processing unit 27 starts recalculation of evaluation points in order to reflect the result of the rescreening in step Sa8. This recalculation of the evaluation points may be the same processing as that described with respect to step Sa5.

  As described above, according to the present embodiment, an evaluation point indicating the degree of risk of oversight in the screening that has already been performed is given to each of the intestinal inner wall surface voxels. Therefore, the user can recognize where the site having a high risk of being overlooked in the observations that have already been made by referring to the evaluation points. In this way, it is possible to efficiently re-observe a site that has a high risk of being overlooked by the observation that has already been performed, and it is possible to efficiently carry out a screening test for colon cancer by endoscopic observation of CTC.

  Furthermore, according to the present embodiment, since the evaluation image reflecting the evaluation point is displayed, it is possible to easily recognize where the portion having a high risk of oversight in the observation that has already been performed.

  Furthermore, according to the present embodiment, screening for additionally observing a portion having a high evaluation score and a high possibility of being overlooked is performed, so that the labor of the observer for the additional observation is saved, and the CTC endoscope Screening screening for colorectal cancer can be performed efficiently. A screening test for colon cancer by endoscopic observation of CTC can be performed more efficiently.

  Furthermore, according to the present embodiment, in the re-screening, the viewpoint and the line-of-sight direction are automatically determined according to the evaluation result of the intestinal inner wall surface voxel to be subjected to the additional observation. Additional observations can be performed, and a screening test for colorectal cancer by endoscopic observation of CTC can be performed more efficiently.

  This embodiment can be variously modified as follows.

  The above embodiment is an example in which evaluation points are assigned during screening involving observation by the observer, and the possibility of oversight in the observation is presented to the observer after the observation is completed. However, prior to observation by the observer, screening may be performed in a simulated manner to give an evaluation point, and the possibility of oversight in observation may be presented to the observer before observation. Specifically, the screening started in step Sa4 may be performed in a simulated manner without displaying on the display unit 24.

  The present invention can also be applied to observation of tubular organs other than the large intestine.

  The present invention can also be applied to other medical modality apparatuses such as an MRI image diagnostic apparatus. Alternatively, the present invention can be implemented in a medical image processing apparatus that processes volume data obtained by various medical modality apparatuses.

  The evaluation point may be assigned to each region including a plurality of intestinal inner wall surface voxels.

  The evaluation points may be given for only a part of all the intestinal inner wall surface voxels. For example, the evaluation point may be assigned only to some representative intestinal inner wall surface voxels of a plurality of intestinal inner wall surface voxels included in a certain region.

  Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

  DESCRIPTION OF SYMBOLS 1 ... Scan gantry, 2 ... Computer apparatus, 10 ... Bed, 11 ... X-ray tube apparatus, 12 ... X-ray detector, 13 ... Mount rotation drive part, 14 ... High voltage generation part, 15 ... Bed drive part, 16 ... Data collection unit, 21 ... Gantry control unit, 22 ... Pre-processing unit, 23 ... Image reconstruction unit, 24 ... Display unit, 25 ... Console, 26 ... Storage unit, 27 ... Image processing unit, 26a ... Raw data storage function 26b ... display condition storage function, 26c ... extracted segment management / storage function, 26d ... colon core line management / storage function, 26e ... gut inner wall position information management / storage function, 26f ... display state management / storage function, 27a ... display state Output function, 27b ... large intestine extraction function, 27c ... large intestine core line extraction function, 27d ... intestinal inner wall voxel extraction function, 27e ... display condition setting function, 27f ... display state evaluation function, 27g ... three-dimensional click Bridging functions, 27h ... display position and angle calculating function, 27i ... colon core correction function, 27j ... colon CAD functions, 27k ... eye tracking function, 28 ... line-of-sight detecting unit, 29 ... control bus, 100 ... CT apparatus.

Claims (23)

Morphological image generation means for generating a morphological image representing the morphology of the inner wall surface of the organ based on volume data representing the morphology of the tubular organ;
A medical image, comprising: an assigning unit that assigns an evaluation point representing difficulty of being seen in the morphological image for each of a plurality of evaluation regions set by dividing the inner wall surface of the organ into a plurality of parts. Processing equipment.   The medical image processing apparatus according to claim 1, wherein the assigning unit assigns an evaluation score corresponding to the number of times the evaluation region has entered the morphological image with respect to the evaluation region.   The medical image processing apparatus according to claim 1, wherein the assigning unit assigns an evaluation score corresponding to an accumulated time that the evaluation region has entered the morphological image with respect to the evaluation region.   The assigning means assigns, for the evaluation region, an evaluation point corresponding to an angle between a line-of-sight direction in the morphological image of the evaluation region and a normal direction of the inner wall surface of the organ in the evaluation region. The medical image processing apparatus according to claim 1, wherein: The morphological image generation means generates a morphological image in a state in which the inner wall surface of the organ is viewed from an arbitrary viewpoint position,
The medical image processing apparatus according to claim 1, wherein the assigning unit assigns an evaluation point corresponding to a distance from the viewpoint position to the evaluation region with respect to the evaluation region. The morphological image generation means generates a morphological image representing a projection screen that is obtained by projecting the inner wall surface of the organ,
The medical image processing apparatus according to claim 1, wherein the assigning unit assigns an evaluation point corresponding to a distance from the center of the projection screen to the evaluation region with respect to the evaluation region. The morphological image generating means generates the morphological image as an image representing the morphology of the inner wall surface on the far side in the line of sight with respect to the clipping position in the organ and the cross section of the organ at the clipping position,
2. The medical device according to claim 1, wherein the assigning unit assigns an evaluation point according to a clipping angle with respect to an inner wall surface of the organ when the evaluation region becomes the clipping position with respect to the evaluation region. Image processing device. Means for inputting line-of-sight information representing the line-of-sight position on the morphological image of the observer;
The medical image processing apparatus according to claim 1, wherein the assigning unit assigns an evaluation point corresponding to a distance from the evaluation region to a visual line position represented by the visual line information with respect to the evaluation region. Means for inputting intensity information indicating the intensity of a colorectal cancer candidate for each of the evaluation areas;
The medical image processing apparatus according to claim 1, wherein the assigning unit assigns an evaluation score corresponding to the strength indicated in the strength information regarding the evaluation region with respect to the evaluation region.   The medical image processing apparatus according to claim 1, further comprising an evaluation image generation unit configured to generate an evaluation image representing a relationship between the plurality of evaluation areas and evaluation points assigned to each of the evaluation areas. .   The medical image processing apparatus according to claim 10, wherein the evaluation image generation unit generates the evaluation image as an image obtained by coloring the morphological image according to the evaluation points.   The medical image processing apparatus according to claim 10, wherein the evaluation image generation unit generates the evaluation image as an image obtained by adding a marker corresponding to the evaluation point to the morphological image.   The medical image processing apparatus according to claim 12, wherein the evaluation image generation unit changes display / non-display of the marker in the evaluation image according to the evaluation point.   The medical image processing apparatus according to claim 12, wherein the evaluation image generation unit varies the size of the marker in the evaluation image according to the evaluation point.   The medical image processing apparatus according to claim 12, wherein the evaluation image generation unit changes a color of a marker in the evaluation image according to the evaluation point.   The medical image processing apparatus according to claim 12, wherein the evaluation image generation unit changes the shape of the marker in the evaluation image according to the evaluation point. Based on the evaluation point, a specifying unit that specifies an evaluation region that is difficult to see in the morphological image already generated by the morphological image generation unit among the plurality of evaluation regions;
The medical image processing apparatus according to claim 1, further comprising a control unit that controls the morphological image generation unit so as to generate a morphological image including the evaluation area specified by the specifying unit.   When a plurality of evaluation areas are specified by the specifying means, the control means sequentially selects the evaluation areas according to a predetermined rule, and generates a morphological image including the selected evaluation areas. The medical image processing apparatus according to claim 17, wherein the morphological image generation unit is controlled as described above.   The control means presents a list of a plurality of evaluation areas specified by the specifying means to the user, edits the list in accordance with the user's instruction, and then displays the evaluation areas indicated in the edited list. The medical image processing apparatus according to claim 18, wherein selection is performed in the order shown in the list.   19. The medical image processing apparatus according to claim 18, wherein the control unit selects a plurality of evaluation regions specified by the specifying unit in order along the core line of the organ according to the position.   The medical image processing apparatus according to claim 18, wherein the control unit selects a plurality of evaluation regions specified by the specifying unit in an order corresponding to the evaluation points. A collecting means for collecting volume data representing the morphology of a tubular organ in the subject;
Morphological image generation means for generating a morphological image representing the morphology of the inner wall surface of the organ based on the volume data;
A medical image, comprising: an assigning unit that assigns an evaluation point representing difficulty of being seen in the morphological image for each of a plurality of evaluation regions set by dividing the inner wall surface of the organ into a plurality of parts. Diagnostic device. A computer capable of accessing a storage medium storing volume data representing the form of a tubular organ;
Morphological image generation means for reading the volume data from the storage medium and generating a morphological image representing the morphology of the inner wall surface of the organ based on the volume data;
Medical image processing characterized by functioning as an assigning means for assigning an evaluation point representing difficulty of being seen in the morphological image for each of a plurality of evaluation areas set by dividing the inner wall surface of the organ into a plurality of parts program.

Images