JP5702943B2 - Pathological diagnosis support device, pathological diagnosis support method, control program for pathological diagnosis support, and recording medium recording the control program - Google Patents

Description

  The present invention relates to a technique for supporting pathological diagnosis by a computer system or the like.

  One pathological diagnosis is a biopsy tissue diagnosis in which a part of an organ is collected from a patient as a specimen, and this is made into a tissue section glass slide specimen (preparation) and observed under a microscope for diagnosis. Such a diagnosis itself must be performed by a doctor, but the amount of work for that is enormous and the burden on the doctor is heavy. Therefore, in order to efficiently carry out observation work for diagnosis and reduce the burden on the doctor, various diagnosis support apparatuses using image processing technology by a computer system have been proposed. For example, Patent Document 1 proposes a pathological diagnosis support apparatus that specifies the position of a cancerous cell by searching a region that satisfies a specific condition from an image obtained by imaging a stained specimen. Patent Document 2 proposes an image processing apparatus that estimates a dye amount from an image obtained by imaging a stained specimen, classifies each pixel in the image based on the dye amount, and displays a classification result. Yes.

JP2009-115598A (for example, paragraph 0033) Japanese Patent Laying-Open No. 2005-331394 (for example, FIG. 7)

  In order to make an efficient diagnosis using this type of device, a function of easily knowing the distribution of cells having certain characteristics in the specimen is important. However, the pathological diagnosis support apparatus of the prior art has not been able to sufficiently meet such a demand. The main reason is as follows. That is, in the prior art, the main focus is on extracting images having characteristics in common with known abnormal tissue sample images by individually evaluating images taken in the specimen. However, since there are large individual differences in cell shape and the like in an actual specimen, it is not always possible to obtain a satisfactory result simply by comparing and contrasting sample images collected from other individuals and partial images of the specimen. Thus, in the conventional diagnosis support technology, there is still room for further improvement in terms of efficiency of diagnosis.

  The present invention has been made in view of the above problems, and provides a technique capable of more efficiently performing pathological diagnosis (for example, stump judgment) in a technical field that supports a pathological diagnosis by a doctor using a computer system or the like. For the purpose.

In one aspect of the pathological diagnosis support apparatus according to the present invention, in order to achieve the above object, an image acquisition unit that acquires an original image obtained by imaging a sample tissue, and the captured original image is divided into a plurality of grid images, A processing unit that classifies the grid image into a plurality of classification categories based on the feature values of each grid image, and a mode in which the plurality of grid images are arranged based on the arrangement in the original image and are different for each classification category. And a display unit that displays the visual information, and the processing unit classifies the grid image into the plurality of classification categories based on the feature amount of the reference image and the feature amount of the grid image. A first classifier using an image prepared in advance as the reference image, and a grid image selected from the plurality of grid images as the reference image And a second classifier using the display means, characterized in that a plurality of classification results obtained by the different classification rules by said processing means, for displaying the same image in different colors with different colors It is said.

In the invention configured as described above, the grid image obtained by dividing the original image is classified into a plurality of classification categories according to the feature. The classification category here is, for example, “having a certain feature and not having it”. The processing means uses a prepared image as a reference image, a first classifier that classifies the grid image into a plurality of classification categories based on the feature amount and the feature amount of the grid image, and selects from the plurality of grid images And a second classifier that classifies the grid image into a plurality of classification categories based on the feature amount of the grid image and the feature amount of the grid image. By applying a classifier and classifying each grid image based on the reference image, grid images having various characteristics can be easily classified into a finite classification category. In particular, as described above, since an image obtained by imaging a specimen tissue has a large individual difference, a good result may not be obtained by classification using only a known image as a reference image. By using together with a classifier that uses a grid image in an original image obtained by imaging the same specimen tissue as a reference image, each grid image included in the original image can be classified with higher accuracy. Then, the grid image with the visual information based on the classification result is displayed in an arrangement corresponding to the arrangement in the original image. For this reason, the user of the apparatus can confirm at a glance to which position in the specimen the grid image classified into which classification category is distributed, thereby enabling the user to perform diagnosis efficiently. . As the display accompanied with “visual information” here, for example, attachment of a symbol indicating a classification category, color coding for each classification category, and highlighting of a grid image classified into a specific classification category are possible. In particular, if multiple classification results obtained by different classification rules are displayed in the same image with different colors, the multiple classification results are reflected individually in the same image, and the visibility of the results Can be further improved.

  In the above aspect, the display means uses the whole or a partial area of the original image as a display target area, and displays each grid image in the display target area in the same arrangement as the arrangement in the original image and based on the classification result. You may make it display. In this way, it is possible to display the same image as when observing the specimen with a microscope and to display visual information indicating the classification result at the same time, which can greatly improve user convenience and diagnosis. Can be made more efficient.

  In this case, the display means may be capable of switching and displaying at least one of the size of the display target area and the position in the original image. In this way, the display target area can be enlarged / reduced or the observation position can be changed, and the visual information based on the classification result is attached to each grid image, so the convenience is further improved.

  In addition, an input unit that receives an instruction input for determining a classification rule in the processing unit from a user may be provided, and the processing unit may classify the grid image with a classification rule corresponding to the instruction input to the input unit. By doing so, classification and display can be performed in various ways according to the user's request, and the functions of the apparatus can be expanded.

  In addition, for example, a selection unit for causing the user to select a reference image from a plurality of grid images may be provided, and the processing unit may classify other grid images using the selected grid image as a reference image. . In this way, other grid images are classified based on the characteristics of the grid image selected from the original image, so that, for example, a part having characteristics similar to a certain part in the specimen tissue is extracted, or a certain part Can be used to evaluate the degree of similarity with other parts. Further, since the user can select the reference image, the feature is not limited to the feature shown in the known sample image, and the feature of an arbitrary portion that the user wants to examine can be evaluated.

In order to achieve the above object, the pathological diagnosis support method according to the present invention acquires an original image obtained by imaging a sample tissue, divides the captured original image into a plurality of grid images, A processing step of classifying an image into a plurality of classification categories based on feature values for each grid image, and a plurality of the grid images arranged based on the arrangement in the original image, and in a different manner for each classification category A display step of displaying information on the display means, and in the processing step, using the image prepared in advance as a reference image, the grid image based on the feature amount of the reference image and the feature amount of the grid image A first classification process for classifying into the plurality of classification categories, and the reference image using a grid image selected from the plurality of grid images as a reference image; The grid image based on the feature amount of the feature amount and the grid image to perform a second classification processing for classifying the plurality of classification categories, in the display step, determined at different classification rules in the processing step The plurality of classification results obtained are color-coded by different colors in the same image and displayed on the display means.

  In the invention configured as described above, as in the above-described invention of the pathological diagnosis support apparatus, the user who uses the method can identify at which position in the specimen the grid image classified into which classification category is distributed. This makes it possible to make an efficient diagnosis.

  Here, in the processing step, the grid image other than the reference image is classified based on the feature amount of the reference image selected from the plurality of grid images by the user and the feature amount of the grid image other than the reference image. May be. As described above, since there are large individual differences in the cell shape and the like in the specimen tissue, there are cases where sufficient results cannot be obtained by comparison with existing samples. In the present invention, since the reference image is selected from the original image obtained by imaging the specimen tissue, it is possible to classify the image with high accuracy corresponding to such individual differences.

  Further, in the processing step, a plurality of classifications based on different classification rules may be performed for each grid image, and in the display step, an image reflecting a plurality of classification results may be displayed on the display means. . Specifically, a grid image group classified into a classification category as a result of one classification and a grid image group classified into a certain classification category as a result of another classification are displayed on the same screen, Set operation results such as union and intersection can be displayed. If it does in this way, comprehensive diagnosis based on the classification result by various classification rules becomes possible, and it can make a user perform more efficient and accurate diagnosis.

In order to achieve the above object, the control program according to the present invention divides an original image obtained by imaging a sample tissue into a plurality of grid images, and the grid images are classified into a plurality of classifications based on feature amounts of the respective grid images. A processing step for classifying into categories, and a plurality of grid images are arranged on the basis of the arrangement in the original image, and are displayed on the display means with visual information of a different aspect for each classification category, and the processing step A display step of displaying a plurality of classification results obtained by different classification rules in different colors in the same image and displaying them on the display means , and the processing step includes an image prepared in advance. The grid image is classified into the plurality of classification categories based on the feature amount of the reference image and the feature amount of the grid image with reference image as a reference image Classifying the grid image into the plurality of classification categories based on a class process and a grid image selected from the plurality of grid images as a reference image based on a feature amount of the reference image and a feature amount of the grid image It is characterized by causing a computer to execute a process including two classification processes . Furthermore, a recording medium according to the present invention is a computer-readable recording medium, and is characterized by recording a control program for supporting pathological diagnosis described above. According to these inventions, the pathological diagnosis support method according to the above invention can be executed by the computer, and the diagnosis by the user can be performed efficiently.

  According to the pathological diagnosis support technique according to the present invention, it is possible to efficiently perform pathological diagnosis (for example, stump judgment) by a user (diagnostic doctor).

It is a figure which shows one Embodiment of the pathological diagnosis assistance apparatus concerning this invention. It is a block diagram which shows the structure of this pathological diagnosis assistance apparatus. It is a flowchart which shows the diagnostic assistance process in this apparatus. It is a figure which shows the example of a target image. It is a figure which shows the example of a model of a classifier. It is a flowchart which shows the selection process of a reference | standard image. It is a figure which shows the example of the screen display in the selection process of a reference | standard image. It is a flowchart which shows the display process in this embodiment. It is a figure which shows the example of the image displayed on a screen. It is a flowchart which shows set operation processing. It is a figure which shows the concept of set operation processing.

  FIG. 1 is a diagram showing an embodiment of a pathological diagnosis support apparatus according to the present invention. The diagnosis support apparatus 1 includes a host computer 10 and an imaging unit 20. The host computer 10 has a configuration and functions equivalent to, for example, a known personal computer or workstation terminal, and includes a processor unit 100 that executes various control programs, an input unit 150 that receives operation input from a user, A display unit 160 that displays various types of information, a disk drive 170 that reads out data by accessing an optical disk OD that is an external recording medium on which various types of data such as a control program and image data are recorded, and an interface 180 are provided. . The host computer 10 can be connected to a telecommunication line 30 such as a local LAN or the Internet via an interface 180.

  As the display unit 160, for example, a liquid crystal display can be used. Further, as the input unit 150, for example, a touch panel integrated with the display unit 160 may be used in addition to a pointing device such as a mouse or a keyboard. Further, the recording medium on which the control program and the image data are recorded is not limited to the optical disk, and any hard disk, memory card, USB memory, or the like can be used.

  The imaging unit 20 includes a microscope 202 to which a CCD camera 201 is attached. The microscope 202 forms an enlarged optical image of a preparation prepared from a specimen tissue collected from a patient on a light receiving unit of the CCD camera 201, and the CCD camera 201 captures the image and converts it into digital data to create a so-called virtual slide. . The specimen tissue is stained by an appropriate staining method according to the purpose of observation (for example, hematoxylin / eosin (HE) staining).

  The host computer 10 performs various types of image processing on the image of the sample tissue imaged by the imaging unit 20 and converted into digital data in accordance with a control program read from the optical disk OD, and displays the image on the display unit 160, thereby allowing the user to Support pathological diagnosis by (diagnostician).

  FIG. 2 is a block diagram showing the configuration of this pathological diagnosis support apparatus. More specifically, each function block surrounded by a one-dot chain line in FIG. 2 is realized by software in the processor unit 100 by executing a control program read from the optical disk OD by the host computer 10. The contents of the diagnosis support processing realized by these functional blocks will be described in detail later. Here, the outline of each functional block will be briefly described.

  An image to be processed (target image) in the apparatus 1 is, for example, an entire virtual slide obtained by imaging a sample tissue and converting it into data, or an image obtained by dividing the virtual slide into a plurality of parts. Also, a plurality of images obtained by individually capturing a plurality of sample tissues collected from the same or a plurality of patients may be set as target images. When one virtual slide is divided into a plurality of target images, information that can specify the position of each target image in the original virtual slide so that the virtual slide can be restored in a later process (for example, coordinates Information) is desirably attached to each target image data.

  The grid image generation unit 101 divides the target image input from the imaging unit 20 into a plurality of grid images with a grid of an arbitrary size that is determined in advance or specified by the user. At this time, information (for example, coordinate information) indicating where each grid image is located in the original target image is given to the grid image data. The input receiving unit 102 functions as an interface for receiving an operation input from the user via the input unit 150, and receives an operation input such as a reference image and a processing parameter from the user as will be described later.

  Each of the first image feature amount calculation unit 104 and the second image feature amount calculation unit 105 has a function of calculating the feature amount of each created grid image. As the feature amount of the image, for example, various things such as the brightness of the image, the texture information, the size of the region satisfying a predetermined condition can be used. As to what physical quantity is adopted as the feature quantity, the first image feature quantity calculation unit 104 uses the predetermined sample image reference classifier parameter 103 and the second image feature quantity calculation unit 105. The target image reference classifier parameter 106 is determined in advance or determined according to an operation input from the user.

  The feature amount of the grid image calculated by the first image feature amount calculation unit 104 is input to the sample image reference classifier 107. The sample image reference classifier 107 uses a sample image stored in advance in the sample image storage unit 112, for example, an image in which a characteristic characteristic of cancerous cells appears as a reference image, and is extracted from the captured target image. Classify the grid image. The classification is performed using a known classification technique, for example, the Euclidean distance in the feature quantity space, and the feature quantity and the classification algorithm used for the classification are determined by the sample image reference classifier parameter 103. Similarly, the feature amount of the grid image calculated by the second image feature amount calculation unit 105 is input to the target image reference classifier 108. The target image reference classifier 108 classifies other grid images using a grid image selected by the user from among the captured target images as a reference image. The feature amount used for classification is determined by the target image reference classifier parameter 106 designated by the operation input from the user.

  The classification results processed by the sample image reference classifier 107 and the target image reference classifier 108 are accumulated in the classification result accumulation unit 109. The classification results are set by the set calculation unit 110 as necessary, and the result is processed by the display processing unit 111 to generate display data, which is sent to the display unit 160 to display the target image. Displayed on the part 160.

  FIG. 3 is a flowchart showing diagnosis support processing in this apparatus. FIG. 4 is a diagram showing an example of the target image. In the diagnosis support process, first, a target image to be diagnosed is acquired (step S101). As shown in FIG. 4A, the target image is a virtual slide image Iv obtained mainly by imaging the entire specimen tissue S. However, as shown by a broken line in FIG. The image Iv may be divided into a plurality of partial images, and each partial image may be a target image. Further, the virtual slide image Iv can be received from the imaging unit 20, or may be received from another imaging device or storage via the electric communication line 30 or the like.

  The target image thus acquired (here, the partial image Ip of the virtual slide image Iv) is divided into a plurality of grid images Ig by an arbitrarily sized grid as shown in FIG. 4B (step S102). And the feature-value of each grid image Ig produced in this way is calculated (step S103). Subsequently, the classifier classifies each grid image Ig based on the obtained feature amount.

  FIG. 5 is a diagram showing a model example of the classifier. In the classifiers 107 and 108 in the present embodiment, as in the classifier 5 shown in FIG. 5A, a certain reference image and an input image are input, and it is determined whether they are similar or dissimilar to each other. Judge and output. More specifically, the classifier 5 obtains both Euclidean distances in the feature amount space from the feature amounts of the reference image and the input image, and determines that the two are similar if the distance is smaller than a predetermined threshold. If the distance is larger than the threshold value, it is determined that they are dissimilar.

  By replacing each grid image in the target image and sequentially inputting it to such a classifier 5, each grid image can be classified into those that are similar to the reference image and those that are not. The classification result changes each time the classification rule such as how to select the reference image and the feature amount to be used is changed. Therefore, the classifier 5 in this embodiment can be expressed as shown in FIG. In other words, N grid images # 1, # 2,..., #N are input to the classifier 5 as input images, and the grid images are classified by giving classification rules thereto. For example, a classification result A is obtained for a certain classification rule, and classification results B, C, and D are obtained for another classification rule. Various models such as a neural network and discriminant analysis have been proposed as such a classifier algorithm, and the present invention is not limited to the above, and such a known classifier (for example, JP-A-2003-317082) is proposed. It is possible to appropriately select and use those described in the publication.

  As shown in FIG. 2, in this embodiment, a sample image reference classifier 107 and a target image reference classifier 108 are provided as classifiers. The sample image reference classifier 107 classifies each grid image using a pre-collected image as a reference image, that is, one input image of the classifier 5. Therefore, the sample image reference classifier 107 determines whether each grid image constituting the target image has a feature similar to a known sample image. On the other hand, the target image reference classifier 108 classifies other grid images using one grid image selected from the target images as a reference image. That is, it is determined whether or not there is a portion similar to the grid image selected as the reference image in the target image.

  Returning to FIG. 3, the description of the diagnosis support process will be continued. When the feature amount of each grid image is calculated in step S103, the grid image is subsequently classified by the classifier described above. Prior to that, two classifiers (sample image reference classifier 107, target image reference classifier) are classified. 108) is used to determine the classification (step S104). This determination may be performed based on the user's instruction, and the classification by the sample image reference classifier 107 can be performed using a predetermined parameter, so that it is performed regardless of the user's instruction. Also good.

  First, a case where classification by the sample image reference classifier 107 is performed will be described. At this time, a parameter representing the classification rule of the sample image reference classifier 107 is set to a predetermined parameter as the classifier parameter 103 (step S105). This parameter may be changeable by the user. Based on the parameters set in this way, the classification work by the sample image reference classifier 107 is performed (step S106), and the classification result is recorded in the classification result accumulation unit 109 (step S107). Further, the classification result is displayed on the display unit 160 as necessary (step S108).

  On the other hand, when performing classification by the target image reference classifier 108, first, the user selects a reference image (step S110).

  FIG. 6 is a flowchart showing a reference image selection process. FIG. 7 is a diagram showing an example of screen display in the reference image selection process. In this process, when there are a plurality of target images, these target images are displayed as thumbnails on the screen 161 of the display unit 160 (step S301). In the example of FIG. 7A, four target images (target images A, B, C, and D) are displayed. Then, the mouse pointer is displayed to allow the user to select and input one of the thumbnail images with the mouse, and the selection input is accepted (step S302).

  Here, it is assumed that the target image B is selected. Thus, the target image corresponding to the thumbnail image selected by the user is enlarged and displayed on the screen 161 of the display unit 160 as shown in FIG. 7B (step S303). Further, in order to clearly show the boundary of the grid image, the grid line G is displayed so as to overlap the target image (step S304). Then, the mouse pointer is displayed again, the user is allowed to select a target image from the displayed grid images, and the selection input is accepted (step S305, FIG. 7B).

  If there is a selection input from the user, highlighting such as surrounding the selected grid image with a frame or changing luminance or color is performed to clearly indicate the selection location (step S306, FIG. 7C), and further selection is performed. Accept. In other words, the user can select whether to specify a new reference image from the currently displayed target image or to specify a reference image from other target images displayed as thumbnails. If selected, the process returns to step S305 to accept a new selection input (step S307). If designation from another target image is selected, the process returns to the thumbnail image selection in step S302 (step S308). If no new selection is required, the process ends.

  Returning to FIG. 3, when the reference image is selected by the user in step S110, the feature amount designated by the target image reference classifier parameter 106 is calculated for the reference image (step S111). Then, based on the classification rule determined by the reference image and the target image reference classifier parameter 106 (step S112), the target image reference classifier 108 classifies each grid image (step S113). Subsequent operations (step S107 and subsequent steps) are the same as the classification work by the sample image reference classifier 107, and the classification result by the target image reference classifier 108 thus obtained is also recorded in the classification result accumulation unit 109.

  In this way, the target image reference classifier 108 classifies other grid images in the target image using the grid image selected from the target image as a reference image. As described above, since there are individual differences in biological tissue, similarity within the same tissue may be more important than similarity with samples obtained from other individuals. Since the target image reference classifier 108 classifies the grid image in the target image as a reference image, it is extremely effective in achieving such an object. In addition, since the user can immediately know the distribution of a portion similar to the reference image simply by selecting the reference image, the diagnosis can be performed efficiently and accurately.

  When the user desires further classification based on a new classification rule after completion of the classification work by the classifier, or when further classification is necessary to realize a predetermined function, the process returns to step S104 and the above processing is performed. Is repeatedly executed (step S109). Then, a process for displaying the classification result thus obtained in various modes is executed.

  In this embodiment, a plurality of classification results obtained by changing the classification rule as described above are accumulated and recorded in the classification result accumulation unit 109, and these classification results are singly or appropriately combined in the display unit 160. By displaying, the efficiency of diagnosis by the user is improved. The processing performed prior to the display processing (step S115) in order to display a plurality of classification results in combination is a set operation (step S114) of the classification results. Here, in order to facilitate understanding, A display process (step S115) when displaying the classification result alone will be described.

  FIG. 8 is a flowchart showing display processing in this embodiment. FIG. 9 is a diagram showing an example of an image displayed on the screen. In the display processing unit 111, each grid image is color-coded based on the classification result (step S501). This color classification can be performed, for example, such that the grid image determined to be similar to a certain reference image is red, and the other grid images have the same color as the original image. Thereby, visual information reflecting the classification result is attached to each grid image. Note that the visual information is not limited to such color coding, and as described above, for example, various information such as light and dark for each grid image and a frame attached to the image can be used.

Next, by arranging each grid image based on the coordinate information in the original target image attached to each grid image, the target image is reconstructed as a partial image Ip of the virtual slide image Iv (step). S502). In this embodiment, the user is allowed to select whether to display the entire virtual slide image or a part thereof. This selection content is determined (step S503), and if the entire display is not selected, the partial image Ip obtained as described above is displayed on the screen 161 of the display unit 160 (step S506).
As shown in FIG. 9A, the partial image Ip reconstructed and displayed in this manner is mapped by color-coding a part of the virtual slide image captured by the imaging unit 20 according to the classification result by the classifier. Is. Therefore, the user can grasp at a glance how the cells and tissues corresponding to the previously specified classification rule are distributed in the target image.

  As shown in FIG. 9B, the virtual slide image Iv is composed of a plurality of partial images Ip. Therefore, when the entire display is selected, the virtual slide image Iv is reproduced by arranging the partial images constituting the virtual slide image Iv based on the coordinate information indicating the position in the virtual slide image Iv. Configure (step S504). Then, reduced image data is created so that the entire reconstructed virtual slide image Iv fits on the screen 161 (step S505), and an image corresponding to the image data is displayed (step S506). By doing so, the entire virtual slide image Iv is displayed on the screen 161 of the display unit 160 as shown in FIG. 9C.

  At this time, since each grid image constituting the virtual slide image Iv reflects the classification result by color coding or the like, the user can see how the cells and tissues corresponding to the classification rule specified earlier are displayed in the virtual slide. You can see at a glance whether it is distributed. Further, the image can be enlarged or reduced by switching between the whole image and the partial image as necessary. Further, by switching the partial images to be enlarged and displayed, it is possible to perform observation while changing the position in the virtual slide.

  Next, a set operation of classification results will be described. By performing the screen display reflecting the classification result as described above, the user can grasp how cells having a certain characteristic are distributed in the specimen tissue, but more accurate diagnosis can be performed. In some cases, it may be desirable to display classification results under a plurality of conditions at the same time. For example, as a result of comparison with a known sample image, a grid that is clearly seen to have characteristics common to the sample image and a grid that is determined to be similar to the sample image, although the direct similarity is low This is a case where it is desired to view a portion where similarity is seen with an image at a time.

  As described in the section “Problems to be Solved by the Invention”, there is a large individual difference in the specimen tissue collected from the living body, so it is difficult to make a highly accurate diagnosis only by uniform comparison with existing sample images. . In this device 1, various classification results are combined and displayed on a single screen to support more accurate diagnosis. One of the processes for this is set operation processing described below.

  FIG. 10 is a flowchart showing the set operation processing. FIG. 11 is a diagram showing the concept of set operation processing. More specifically, FIG. 11A shows an example of a screen display during execution of the set operation process, and FIG. 11B shows an example of an image obtained by the set operation process. In the set operation processing, as shown in FIG. 11A, as a setting screen, display items of “calculation”, “operator input”, and “classification result input” are displayed on a screen 161, an “execute calculation” button, Is displayed (step S701). First, from the pull-down menu of “classification result input”, the user is made to select which classification result is reflected in the display among the plurality of classification results executed previously (step S702). The selected classification result is input to the “calculation” field (step S703). When an operator is selected from the “operator input” field (step S706), the selected operator is additionally input to the “calculation” field (step S707). Any logical operator can be used as the operator. An arithmetic expression is created by repeating this appropriately. During this time, when the “calculation execution” button is pressed (step S704), the set calculation unit 110 performs a set calculation between the classification results based on the generated calculation formula (step S705).

  This set operation corresponds to at least one of the two classification rules by taking the union of the first classification result obtained by a certain classification rule and the second classification result obtained by another classification rule, for example. This is an operation such as extracting a part to be performed. In this case, for example, as shown in FIG. 11B, the grid image corresponding to the first classification result and the grid image corresponding to the second classification result are color-coded differently from the other grid images, respectively. Images superimposed on each other are obtained as a calculation result. If different color classification is performed for each classification result, a plurality of classification results can be reflected individually, and the visibility of the results is further improved. The result of the set operation is sent to the display processing unit 111 and displayed on the screen 161 of the display unit 160 by display processing.

  By performing such a set operation in units of grid images for the entire virtual slide image Iv and displaying it, in this embodiment, a plurality of classification results can be displayed on the same screen. In addition, the entire virtual slide or part can be enlarged and reduced as necessary. For this reason, the user can grasp at a glance how the portions having various characteristics are distributed in the specimen tissue.

  As described above, in this embodiment, a virtual slide image obtained by imaging a specimen tissue is divided into a plurality of grid images, and classification by a classifier is performed based on the feature amount of each grid image. Then, visual information reflecting the classification result is attached to each grid image, and an image of the specimen tissue is reconstructed and displayed. In addition, as classification by the classifier, an existing sample image is used as a reference image, and a grid image selected from the target image is used as a reference image. In addition, it has a function of displaying a plurality of classification results as a set operation. With these functions, accurate diagnosis can be efficiently performed with a small amount of work, and the burden on the diagnostician is greatly reduced.

  As described above, in this embodiment, the imaging unit 20 functions as the “image acquisition unit” of the present invention, while the host computer 10, particularly the processor unit 100 functions as the “processing unit” of the present invention. . In this embodiment, the display unit 160 functions as the “display unit” of the present invention, and the input unit 150 functions as the “input unit” and the “selection unit” of the present invention. In this embodiment, the sample image reference classifier 107 and the target image reference classifier 108 function as the “first classifier” and the “second classifier” of the present invention, respectively. In this embodiment, the virtual slide image Iv corresponds to the “original image” of the present invention.

  In the pathological diagnosis support process (FIG. 3) of the above embodiment, step S101 corresponds to the “image acquisition process” of the present invention, and steps S102 to S114 correspond to the “processing process” of the present invention. . Further, step S115 corresponds to the “display process” of the present invention.

  The present invention is not limited to the above-described embodiment, and various modifications other than those described above can be made without departing from the spirit of the present invention. For example, in the above embodiment, the imaging unit 20 including the CCD camera 201 and the microscope 202 is provided as an “image acquisition unit”, but the imaging unit is not essential. As described above, it is also possible to receive an original image from the telecommunication line 30 via the interface 180 or from an external recording medium via an appropriate interface. In this case, these interfaces serve as “image acquisition means”. Will work. In this sense, the “step for acquiring an image” is not essential in the control program of the present invention.

  In the above embodiment, the sample image reference classifier 107 using the existing sample image as the reference image and the target image reference classifier 108 using the grid image selected from the target image as the reference image are provided. However, both of these are not always necessary, and it is possible to configure a pathological diagnosis support apparatus that is highly convenient with only one of them.

  In the above-described embodiment, the user can select an arbitrary grid image as a reference image from the target image, but the user can also select a reference image from a plurality of images for existing sample images. You may be able to do it. In the above embodiment, the target image reference classifier 108 performs the classification operation using the grid image selected by the user as the reference image. For example, the grid image determined to be similar to the sample image by the classification based on the sample image is automatically selected. In addition, by operating the target image reference classifier 108 as a reference image in the target image, the classification based on the sample image and the classification based on the grid image in the target image may be automatically performed as a series of processes. .

  In this embodiment, processing by each functional block is realized by the host computer 10 executing the control program recorded on the optical disk OD. However, the recording medium for recording the control program is limited to the optical disk. Any other recording medium can be used as described above. The control program may be distributed through the telecommunication line 30.

  Moreover, although the pathological diagnosis support apparatus 1 of the said embodiment is comprised combining the general purpose host computer 10 and the imaging part 20, it is made to comprise a pathological diagnosis support apparatus combining an imaging part and a dedicated terminal. Also good. Further, by incorporating the control program according to the present invention into an existing virtual slide creation device, the virtual slide creation device may function as a pathological diagnosis support device.

  Further, for example, the classification result and the set operation result can be distributed to the outside through the telecommunication line 30 or recorded on an external recording medium instead of or in addition to being displayed on the display unit 160. You may do it.

  The present invention can be suitably applied to a technical field in which various pathological diagnoses by a doctor are supported by a computer system or the like.

20 Imaging unit (image acquisition means)
100 processor unit (processing means)
107 Sample image reference classifier (first classifier)
108 Target image reference classifier (second classifier)
150 Input section (input means, selection means)
160 Display unit (display means)
Ig grid image Ip partial image (target image)
Iv Virtual slide image (original image)

Claims (9)

Image acquisition means for acquiring an original image obtained by imaging a sample tissue;
Processing means for dividing the captured original image into a plurality of grid images, and classifying the grid images into a plurality of classification categories based on feature amounts of the respective grid images;
Display means for displaying a plurality of the grid images with an arrangement based on the arrangement in the original image and with different visual information for each classification category;
The processing means uses a first image prepared as the reference image as a classifier that classifies the grid image into the plurality of classification categories based on the feature amount of the reference image and the feature amount of the grid image. And a second classifier that uses a grid image selected from the plurality of grid images as the reference image,
The pathological diagnosis support apparatus, wherein the display means displays a plurality of classification results obtained by the processing means according to different classification rules, in different colors, on the same image.   The display unit displays the entire or a partial area of the original image as a display target area, and displays each grid image in the display target area in the same arrangement as the arrangement in the original image and with the visual information. The pathological diagnosis support apparatus according to claim 1.   The pathological diagnosis support apparatus according to claim 2, wherein the display unit can switch and display at least one of a size of the display target region and a position in the original image.   An input means for receiving an instruction input for determining a classification rule in the processing means from a user, wherein the processing means classifies the grid image according to a classification rule according to an instruction input to the input means. 4. The pathological diagnosis support apparatus according to any one of 3 above. A selection means for selecting the reference image to the user from among the plurality of grid images, the processing means, the grid image selected in claim 1 for classifying other of said grid image as the reference image The pathological diagnosis support apparatus described. An image acquisition step of acquiring an original image obtained by imaging a sample tissue;
A process step of dividing the captured original image into a plurality of grid images, and classifying the grid images into a plurality of classification categories based on feature amounts of the respective grid images;
A display step of displaying the plurality of grid images on a display unit with visual information of a different form for each classification category in an arrangement based on the arrangement in the original image,
In the processing step, a first classification process for classifying the grid image into the plurality of classification categories based on a feature amount of the reference image and a feature amount of the grid image using an image prepared in advance as a reference image; A second classification process for classifying the grid image into the plurality of classification categories based on a feature amount of the reference image and a feature amount of the grid image using a grid image selected from the plurality of grid images as a reference image; Run
In the display step, a plurality of classification results obtained by different classification rules in the processing step are color-coded in different colors in the same image and displayed on the display means. In the processing step, the grid images other than the reference image are classified based on the feature amount of the reference image selected from the plurality of grid images by the user and the feature amount of the grid image other than the reference image. The pathological diagnosis support method according to claim 6 . A processing step of dividing an original image obtained by imaging a specimen tissue into a plurality of grid images, and classifying the grid images into a plurality of classification categories based on feature amounts of the respective grid images;
A plurality of grid images are arranged on the basis of the arrangement in the original image and displayed on the display means with different visual information for each classification category, and are obtained by different classification rules in the processing step. A display step of displaying the plurality of classification results in different colors in the same image and displaying them on the display means , and the processing step uses the image prepared in advance as a reference image of the reference image. A first classification process for classifying the grid image into the plurality of classification categories based on a feature amount and a feature amount of the grid image; and the reference image using the grid image selected from the plurality of grid images as a reference image And a second classification process for classifying the grid image into the plurality of classification categories based on the feature quantity of the image and the feature quantity of the grid image.
A control program for pathological diagnosis support, characterized by causing a computer to execute processing . A computer-readable recording medium on which the control program according to claim 8 is recorded.