Classification device, classification method, and program

Abstract

In order to improve the accuracy of classification of a pathology sample as benign or malignant, this classification device (1) comprises an imaging unit (11) for capturing an image that includes a partial range of a pathology sample as the subject, an acquisition unit (12) for acquiring the image captured by the imaging unit (11), and a classification unit (13) for classifying whether cells included as the subject in the image acquired by the acquisition unit (12) are benign or malignant.

Description

Technical Field

[0007] , , , , , solution , , , ,

[0006]

[0001] The present invention relates to a classification device, a classification method, and a program for classifying whether a cell is benign or malignant.

Background Art

[0002] There is disclosed a technique for photographing a sample using an imaging means such as a camera attached to a microscope and analyzing a substance included as a subject in the photographed microscope image.

[0003] For example, Patent Document 1 discloses a method for performing an analysis for classifying an image of a sample into a target region and a non-target region during a scanning operation of the sample using a microscope.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the technique of Patent Document 1, in order to rapidly generate a large amount of data, an analysis of an image of a sample is performed during a scanning operation of the sample. Therefore, in the technique of Patent Document 1 , solution there is room for improvement in terms of the accuracy of the analysis.

[0006] One aspect of the present invention has been made in view of the above problems, and an example of the object is to provide a technique for improving the accuracy of classifying a pathological sample as benign or malignant.

Means for Solving the Problems

[0007] The classification device according to one aspect of the present invention is an imaging device that images an image including a part of a pathological sample as a subject Department And the imaging Department The system includes an acquisition means for acquiring images captured by the system, and a classification means for classifying whether the cells included as subjects in the images acquired by the acquisition means are benign or malignant.

[0008] A classification method relating to one aspect of the present invention includes capturing an image that includes a portion of a pathological sample as the subject, acquiring the image captured in the capturing process, and classifying whether the cells included as the subject in the acquired image are benign or malignant.

[0009] A program relating to one aspect of the present invention is: It includes an imaging unit that captures an image that includes a portion of a pathological sample as the subject. A program for causing a computer to function as a classification device, the imaging Department The system functions as an acquisition means for acquiring images captured by the system, and a classification means for classifying whether the cells included as subjects in the images acquired by the acquisition means are benign or malignant. [Effects of the Invention]

[0010] According to one aspect of the present invention, the accuracy of classifying pathological samples as benign or malignant can be improved. [Brief explanation of the drawing]

[0011] [Figure 1] This is a block diagram showing the configuration of a classification device according to exemplary embodiment 1 of the present invention. [Figure 2] This is a flowchart showing the flow of a classification method according to exemplary embodiment 1 of the present invention. [Figure 3] This is a block diagram showing the configuration of a classification device according to an exemplary embodiment 2 of the present invention. [Figure 4] This figure shows an example of how the automatic control unit moves the imaging range of the imaging unit in exemplary embodiment 2 of the present invention. [Figure 5] This is a flowchart showing the flow of the classification method according to exemplary embodiment 2 of the present invention. [Figure 6]It is a block diagram showing the configuration of a classification device according to Exemplary Embodiment 3 of the present invention. [Figure 7] It is a flowchart showing the flow of a classification method according to Exemplary Embodiment 3 of the present invention. [Figure 8] It is an example of an image displayed on a display device in Exemplary Embodiment 3 of the present invention. [Figure 9] It is a block diagram showing the configuration of a classification device according to Exemplary Embodiment 4 of the present invention. [Figure 10] It is a flowchart showing the flow of a classification method according to Exemplary Embodiment 4 of the present invention. [Figure 11] It is an example of an image displayed on a display device in Exemplary Embodiment 4 of the present invention. [Figure 12] It is a flowchart showing the flow of a classification method according to a modified example of the present invention. [Figure 13] It is a block diagram showing an example of the hardware configuration of a classification device in each exemplary embodiment of the present invention.

Mode for Carrying Out the Invention

[0012] 〔Exemplary Embodiment **********1〕 The first exemplary embodiment of the present invention will be described in detail with reference to the drawings. This exemplary embodiment is a basic form for the exemplary embodiments described later.

[0013] [[ID=3**4]] (Configuration of Classification Device 1) The configuration of the classification device 1 according to this exemplary embodiment will be described with reference to FIG. 1. FIG. 1 is a block diagram showing the configuration of the classification device 1 according to this exemplary embodiment. [[ID=3**8]]

[0014] The classification device 1 is a device that acquires an image including a part of the range of a pathological sample as a subject, and classifies whether the cells included as the subject in the image are benign or malignant.

[0015] **Note**: There seems to be an issue with the tag `**26` in the original text where the number of asterisks is inconsistent. It is translated as is for now. If it is an error, please correct the original text for a more accurate translation.A pathological sample is a sample used in pathological diagnosis to determine whether it is benign or malignant. Samples used in cytological diagnosis, where the cells contained in the sample are classified as either benign or malignant, are also included in the category of pathological samples.

[0016] As shown in Figure 1, the classification device 1 comprises an imaging unit 11, an acquisition unit 12, and a classification unit 13. In this exemplary embodiment, the imaging unit 11, the acquisition unit 12, and the classification unit 13 are configured to implement imaging means, acquisition means, and classification means, respectively.

[0017] The imaging unit 11 captures an image that includes a portion of the pathological sample as the subject.

[0018] The acquisition unit 12 acquires the image captured by the imaging unit 11.

[0019] The classification unit 13 classifies whether the cells included as subjects in the image acquired by the acquisition unit 12 are benign or malignant. The method by which the classification unit 13 classifies whether the cells included as subjects in the image are benign or malignant can be a known method. For example, the classification unit 13 may take an image containing cells as input, input the image acquired by the acquisition unit 12 to a trained model that classifies whether the cells are benign or malignant, and obtain the classification result of the trained model.

[0020] As described above, the classification device 1 according to this exemplary embodiment employs a configuration comprising an imaging unit 11 that captures an image including a portion of a pathological sample as the subject, an acquisition unit 12 that acquires the image captured by the imaging unit 11, and a classification unit 13 that classifies whether the cells included as the subject in the image acquired by the acquisition unit 12 are benign or malignant. Therefore, the classification device 1 according to this exemplary embodiment can improve the accuracy of classifying pathological samples as benign or malignant.

[0021] (Process of classification method S1) The flow of the classification method S1 according to this exemplary embodiment will be explained with reference to Figure 2. Figure 2 is a flowchart showing the flow of the classification method S1 according to this exemplary embodiment.

[0022] (Step S11) In step S11, the imaging unit 11 captures an image that includes a portion of the pathological sample as the subject.

[0023] (Step S12) In step S12, the acquisition unit 12 acquires the image captured by the imaging unit 11 in step S11.

[0024] (Step S13) In step S13, the classification unit 13 classifies whether the cells included as subjects in the image acquired by the acquisition unit in step S12 are benign or malignant.

[0025] As described above, in the classification method S1 according to this exemplary embodiment, in step S11, the imaging unit 11 captures an image that includes a portion of the pathological sample as the subject; in step S12, the acquisition unit 12 acquires the image captured by the imaging unit 11 in step S11; and in step S13, the classification unit 13 classifies whether the cells included as the subject in the image acquired by the acquisition unit in step S12 are benign or malignant. Therefore, according to the classification method S1 according to this exemplary embodiment, the same effects as the classification device 1 can be obtained.

[0026] [Exemplary Embodiment 2] A second exemplary embodiment of the present invention will be described in detail with reference to the drawings. Components having the same function as those described in the above-described exemplary embodiment will be denoted by the same reference numerals, and their descriptions will be omitted as appropriate.

[0027] (Classifier 2) Classification device 2 is a device that classifies whether the cells contained in a pathological sample are benign or malignant. More specifically, classification device 2 acquires a first image, which is an image containing a portion of the pathological sample as the subject, and is an image with a first magnification. Next, if the number of cells contained in the first image is greater than a predetermined number (for example, 10 cells), classification device 2 acquires a second image with a second magnification, which is higher than the first magnification. Then, classification device 2 classifies whether the cells contained in the acquired second image as the subject are benign or malignant.

[0028] Here, the process from when the classification device 2 acquires the first image to when it classifies whether the cells included as subjects in the second image are benign or malignant is referred to as the automatic classification process.

[0029] On the other hand, if the number of cells classified as malignant is less than a predetermined number, the classification device 2 moves the imaging range and performs the automatic classification process again.

[0030] The first magnification is not particularly limited, but it is preferably a magnification sufficient to detect and count the number of cells included as subjects in the pathological sample, and in this exemplary embodiment, for example, it is 10x.

[0031] The second magnification is preferably higher than the first magnification and is sufficient to classify whether the cells included as subjects in the pathological sample are benign or malignant. For example, in this exemplary embodiment, it is 40x.

[0032] (Configuration of Classification Device 2) The configuration of the classification device 2 according to this exemplary embodiment will be described with reference to Figure 3. Figure 3 is a block diagram showing the configuration of the classification device 2 according to this exemplary embodiment.

[0033] As shown in Figure 3, the classification device 2 comprises a control unit 10, an imaging unit 11, a storage unit 20, and an output unit 22. The imaging unit 11 is configured to implement an imaging means in this exemplary embodiment.

[0034] The imaging unit 11 captures an image that includes a portion of the pathological sample as the subject. The imaging unit 11, for example, is equipped with an imaging sensor and captures an image of an object included in the field of view. Furthermore, the imaging magnification of the imaging unit 11 is adjusted by the adjustment unit 15, which will be described later. For example, the imaging magnification of the imaging unit is adjusted to a first magnification or a second magnification. The imaging unit 11 supplies the first image IP, captured at the first magnification, and the second image EP, captured at the second magnification, to the control unit 10, which will be described later.

[0035] The memory unit 20 stores data that the control unit 10, which will be described later, references. Examples of data stored in the memory unit 20 include the first image IP, the second image EP, coordinate information CI, cell number information NCI, and cell position information CLI. The coordinate information CI, cell number information NCI, and cell position information CLI will be described later.

[0036] The output unit 22 is an interface that outputs data supplied from the control unit 10 (described later) to other connected devices. Examples of other connected devices include display devices that display images and speakers that output sound.

[0037] (Control Unit 10) The control unit 10 controls each component of the classification device 2. For example, it stores data in the storage unit 20 and supplies data to the output unit 22.

[0038] As shown in Figure 3, the control unit 10 also functions as an acquisition unit 12, a classification unit 13, an adjustment unit 15, a drive unit 16, an estimation unit 17, and an automatic control unit 18. In this exemplary embodiment, the acquisition unit 12, classification unit 13, adjustment unit 15, drive unit 16, estimation unit 17, and automatic control unit 18 are configured to realize an acquisition means, a classification means, an adjustment means, a drive means, an estimation means, and a control means, respectively.

[0039] The acquisition unit 12 acquires images captured by the imaging unit 11. For example, the acquisition unit 12 acquires a first image IP and a second image EP captured by the imaging unit 11 in response to instructions from the automatic control unit 18, which will be described later. The acquisition unit 12 stores the acquired first image IP and second image EP in the storage unit 20.

[0040] The classification unit 13 acquires the second image EP stored in the memory unit 20 and classifies whether the cells included as subjects in the second image EP are benign or malignant. The classification unit 13 stores the classification result CR, which indicates the classification result, in the memory unit 20. As an example, the classification unit 13 performs the classification process in response to instructions from the automatic control unit 18, which will be described later. The method by which the classification unit 13 classifies whether the cells included as subjects in the second image EP are benign or malignant is as described above.

[0041] The adjustment unit 15 adjusts the imaging magnification of the imaging unit 11. For example, the adjustment unit 15 adjusts the imaging magnification of the imaging unit 11 to a first magnification or a second magnification in response to instructions from the automatic control unit 18, which will be described later.

[0042] The drive unit 16 moves the imaging range of the imaging unit 11. For example, the drive unit 16 moves the imaging range so as to comprehensively cover the entire pathological sample, starting from the center of the pathological sample, in response to instructions from the automatic control unit 18, which will be described later. With this configuration, the classification device 2 can start the classification process from the center of the pathological sample, where many cells are thought to be present, and thus the classification process can be performed quickly.

[0043] The estimation unit 17 estimates the number of cells included as subjects in the image. For example, the estimation unit 17 detects cells included as subjects in the first image IP acquired by the acquisition unit 12 in response to instructions from the automatic control unit 18, which will be described later, and estimates the number of detected cells. The estimation unit 17 may use known methods such as those described below to detect cells and estimate the number of cells. Step 1: Create a regressor (regression analyzer) that is trained to convert pathological images from RGB tones to grayscale tones corresponding to staining intensity (this converted image will be referred to as the "staining intensity image"). Step 2: Use the above regressor to convert the image to be analyzed into a staining intensity image. Step 3: Extract regions from the above staining intensity image that have a staining intensity above a certain level and a circularity above a certain level.

[0044] Another example is a configuration in which the estimation unit 17 takes an image containing cells as input, detects the cells contained in the image, and inputs the first image IP to a trained model that calculates the number of detected cells, and obtains the number of cells output by the trained model. The estimation unit 17 stores cell number information NCI indicating the estimated number of cells in the storage unit 20.

[0045] Furthermore, the estimation unit 17 stores cell position information CLI, which indicates the location of the detected cell, in the storage unit 20. As an example, the estimation unit 17 sets the center of the first image IP as the center of the two-dimensional coordinate system and generates cell position information CLI that indicates the coordinates of the detected cell's location.

[0046] (Processing performed by the automatic control unit 18) The automatic control unit 18 controls the acquisition unit 12, the classification unit 13, the adjustment unit 15, and the drive unit 16. As an example, the automatic control unit 18 instructs each unit to perform the following automatic classification process.

[0047] (1) The adjustment unit 15 is instructed to adjust the imaging magnification of the imaging unit 11 to the first magnification.

[0048] (2) The acquisition unit 12 is instructed to acquire the first image IP captured by the imaging unit 11.

[0049] (3) The estimation unit 17 is instructed to detect cells included as subjects in the first image IP and to estimate the number of detected cells.

[0050] (4) If the number of cells estimated by the estimation unit 17 is greater than or equal to a predetermined number, the adjustment unit 15 is instructed to adjust the imaging magnification of the imaging unit 11 to a second magnification, the acquisition unit 12 is instructed to acquire a second image EP, and the classification unit 13 is instructed to classify whether the cells included as subjects in the second image EP are benign or malignant.

[0051] Furthermore, if the number of cells classified as malignant by the classification unit 13 as a result of the automatic classification process is less than a predetermined number, the automatic control unit 18 instructs the drive unit 16 to move the imaging range and performs the automatic classification process again.

[0052] As an example, the automatic control unit 18 sets the center of the pathology sample as the center of the two-dimensional coordinate system and generates coordinate information CI indicating the coordinates of the destination of the drive unit 16. The automatic control unit 18 then supplies the generated coordinate information CI to the drive unit 16, instructing the drive unit 16 to move the imaging range of the imaging unit 11. The automatic control unit 18 also stores the generated coordinate information CI in the storage unit 20. An example of how the automatic control unit 18 moves the imaging range of the imaging unit 11 will be explained using Figure 4.

[0053] (Example of moving the imaging range) Figure 4 shows an example of how the automatic control unit 18 moves the imaging range of the imaging unit 11 in this exemplary embodiment.

[0054] The left side of Figure 4 shows a diagram of a pathology sample SA. As shown on the left side of Figure 4, the automatic control unit 18 sets the center of the pathology sample to the center CC of the two-dimensional coordinate system. In the example shown in Figure 4, the automatic control unit 18 moves the imaging unit 11 in a scanning sequence SO that comprehensively moves the entire pathology sample from the center CC of the pathology sample.

[0055] First, the automatic control unit 18 generates coordinate information CI indicating the coordinates of the central CC of the pathology sample SA in order to move the imaging unit 11 to the central CC. Next, the automatic control unit 18 supplies the generated coordinate information CI to the drive unit 16, instructing the drive unit 16 to make the center of the imaging range of the imaging unit 11 the central CC. The automatic control unit 18 also stores the generated coordinate information CI in the storage unit 20. Once the drive unit 16 has moved the imaging unit 11 to a position where the center of the imaging range is the central CC, the automatic control unit 18 executes the automatic classification process.

[0056] If the number of cells classified as malignant by the classification unit 13 as a result of the automatic classification process is less than a predetermined number, the automatic control unit 18 refers to the coordinate information CI stored in the memory unit 20 in order to move the imaging range. The automatic control unit 18 refers to the coordinate information CI and the scanning order SO and generates coordinate information CI indicating the coordinates of the next destination position MC1 for the imaging unit 11. The automatic control unit 18 supplies the generated coordinate information CI to the drive unit 16, instructing the drive unit 16 to set the center of the imaging range of the imaging unit 11 to position MC1. The automatic control unit 18 also stores the generated coordinate information CI in the memory unit 20. Once the drive unit 16 has moved the imaging unit 11 so that the center of the imaging range is at position MC1, the automatic control unit 18 executes the automatic classification process.

[0057] Thus, if the number of cells classified as malignant by the classification unit 13 as a result of the automatic classification process is less than a predetermined number, the automatic control unit 18 moves the imaging unit 11 in a scanning sequence SO that comprehensively moves the entire pathology sample from the center CC of the pathology sample. For example, when the automatic control unit 18 moves the imaging unit 11 so that the center of the imaging range becomes position MCN, it instructs the acquisition unit 12 to acquire the first image IPN of the imaging range REN, as shown on the right side of Figure 4, and starts the automatic classification process.

[0058] (Classification method S2 flow) The flow of the classification method S2 executed by the classification device 2 according to this exemplary embodiment will be explained with reference to Figure 5. Figure 5 is a flowchart showing the flow of the classification method S2 according to this exemplary embodiment.

[0059] (Step S21) In step S21, the automatic control unit 18 refers to the coordinate information CI stored in the memory unit 20 and determines whether or not the imaging unit 11 has been moved to the end of the scanning sequence SO.

[0060] If it is determined in step S21 that the entire sample has been imaged (step S21: Yes), the classification device 2 terminates the process shown in Figure 5.

[0061] (Step S22) If it is determined in step S21 that the entire sample has not been imaged (step S21: No), then in step S22, the automatic control unit 18 determines the coordinates of the destination of the imaging unit 11. The automatic control unit 18 generates coordinate information CI indicating the destination coordinates and supplies the generated coordinate information CI to the drive unit 16. The automatic control unit 18 also stores the generated coordinate information CI in the storage unit 20.

[0062] (Step S23) In step S23, the drive unit 16 moves the imaging unit 11 to the position indicated by the coordinate information CI supplied from the automatic control unit 18 in step S22.

[0063] (Step S24) In step S24, the automatic control unit 18 instructs the adjustment unit 15 to adjust the imaging magnification of the imaging unit 11 to a first magnification. The adjustment unit 15 adjusts the imaging magnification of the imaging unit 11 to the first magnification.

[0064] (Step S25) In step S25, the imaging unit 11 captures the first image IP.

[0065] (Step S26) In step S26, the automatic control unit 18 instructs the acquisition unit 12 to acquire the first image IP captured by the imaging unit 11. The acquisition unit 12 acquires the first image IP captured by the imaging unit 11. The acquisition unit 12 stores the acquired first image IP in the storage unit 20.

[0066] (Step S27) In step S27, the automatic control unit 18 instructs the estimation unit 17 to estimate the number of cells included as subjects in the first image IP. The estimation unit 17 acquires the image IP stored in the storage unit 20 and detects the cells included as subjects in the image IP. The estimation unit 17 then estimates the number of detected cells. The estimation unit 17 also stores cell number information NCI, which indicates the estimated number of cells, and cell position information CLI, which indicates the location of the detected cells, in the storage unit 20.

[0067] (Step S28) In step S28, the automatic control unit 18 acquires the cell count information NCI stored in the memory unit 20 and determines whether the number of cells indicated by the cell count information NCI is greater than or equal to a predetermined number.

[0068] If it is determined in step S28 that the number of cells is less than a predetermined number (step S28: No), the classification device 2 returns to the process in step S21.

[0069] (Step S29) If, in step S28, it is determined that the number of cells is greater than or equal to a predetermined number (step S28: Yes), then in step S29, the automatic control unit 18 instructs the adjustment unit 15 to adjust the imaging magnification of the imaging unit 11 to the second magnification. The adjustment unit 15 adjusts the imaging magnification of the imaging unit 11 to the second magnification.

[0070] (Step S30) In step S30, the imaging unit 11 captures a second image EP.

[0071] As an example of a method for capturing a second image EP, the automatic control unit 18 acquires cell position information CLI stored in the memory unit 20, and moves the imaging unit 11 via the drive unit 16 to the position indicated by the cell position information CLI so that the cell position indicated by the CLI is at the center of the imaging range. The imaging unit 11 then captures the second image EP.

[0072] For example, if the cell count information NCI stored in the memory unit 20 indicates that the estimated number of cells is N, it is preferable to acquire N second images EP in which the position of each of the N cells is at the center of the imaging range.

[0073] (Step S31) In step S31, the automatic control unit 18 instructs the acquisition unit 12 to acquire the second image EP captured by the imaging unit 11. The acquisition unit 12 acquires the second image EP captured by the imaging unit 11. The acquisition unit 12 stores the acquired second image EP in the storage unit 20.

[0074] (Step S32) In step S32, the automatic control unit 18 instructs the classification unit 13 to perform the classification process. The classification unit 13 retrieves the second image EP stored in the memory unit 20 and classifies whether the cells included as subjects in the second image EP are benign or malignant. The classification unit 13 stores the classification result CR, which indicates the classification result, in the memory unit 20.

[0075] (Step S33) In step S33, the automatic control unit 18 retrieves the classification result CR stored in the memory unit 20 and determines whether the number of cells classified as malignant by the classification unit 13 is greater than or equal to a predetermined number.

[0076] In step S33, if it is determined that the number of cells classified as malignant is greater than or equal to a predetermined number (step S33: Yes), the classification device 2 terminates the process shown in Figure 5.

[0077] On the other hand, if it is determined in step S33 that the number of cells classified as malignant is less than a predetermined number (step S33: No), the classification device 2 returns to the process in step S21.

[0078] As described above, the classification device 2 according to this exemplary embodiment acquires a second image EP, which is captured at a second imaging magnification higher than the first imaging magnification, if the number of cells included as subjects in the first image IP, which is captured at a first imaging magnification, is greater than or equal to a predetermined number. The classification device 2 according to this exemplary embodiment then employs a configuration that classifies whether the cells included as subjects in the second image EP are benign or malignant.

[0079] Therefore, according to the classification device 2 of this exemplary embodiment, the cells included as subjects in the second imaging image EP, which is an enlarged region containing a predetermined number of cells or more, are classified as benign or malignant, thereby improving the accuracy of the classification of pathological samples into benign and malignant.

[0080] Furthermore, according to the classification device 2 of this exemplary embodiment, the imaging range is automatically changed, and the automatic classification process is performed within the changed imaging range. Therefore, according to the classification device 2 of this exemplary embodiment, the same classification result can be obtained regardless of the skill level of the user using the classification device 2.

[0081] [Exemplary Embodiment 3] A third exemplary embodiment of the present invention will be described in detail with reference to the drawings. Components having the same function as those described in the above-described exemplary embodiments will be denoted by the same reference numerals, and their descriptions will be omitted as appropriate.

[0082] (Classifier 3) In this exemplary embodiment, the classification device 3, in step S27 of the flowchart shown in Figure 5 above, displays a second image EP, captured so that the cells included as subjects in the first image IP are at the center of the imaging range, in response to instructions from the user, if the number of cells estimated by the estimation unit 17 is 1 or more (in other words, if cells are detected as subjects in the first image IP). Instructions from the user that the classification device 3 receives will be described later.

[0083] (Configuration of Classification Device 3) The configuration of the classification device 3 according to this exemplary embodiment will be described with reference to Figure 6. Figure 6 is a block diagram showing the configuration of the classification device 3 according to this exemplary embodiment.

[0084] As shown in Figure 6, the classification device 3 comprises a control unit 10A, an imaging unit 11, a storage unit 20A, an output unit 22, and an input unit 23. The imaging unit 11 is configured to implement an imaging means in this exemplary embodiment.

[0085] The imaging unit 11 is as described above.

[0086] The memory unit 20A stores cell position information CLI, which indicates the location of the detected cells in the data stored in the memory unit 20 described above, in association with the classification result CR.

[0087] The output unit 22 is an interface that outputs data supplied from the control unit 10A (described later) to other connected devices. In this exemplary embodiment, the output unit 22 acquires image data from the control unit 10A and outputs the image data to a connected display device.

[0088] The input unit 23 is an interface that receives operations from the user. The input unit 23 supplies instruction information indicating the operation received from the user to the control unit 10A.

[0089] (Control Unit 10A) The control unit 10A controls each component of the classification device 3. For example, it stores data in the storage unit 20A and supplies data to the output unit 22.

[0090] As shown in Figure 6, the control unit 10A also functions as an acquisition unit 12, a classification unit 13, an adjustment unit 15, a drive unit 16, an estimation unit 17, an automatic control unit 18, and an input receiving unit 19. In this exemplary embodiment, the acquisition unit 12, classification unit 13, adjustment unit 15, drive unit 16, estimation unit 17, automatic control unit 18, and input receiving unit 19 also function as acquisition means, classification means, adjustment means, drive means, estimation means, control means, and receiving means.

[0091] The acquisition unit 12, adjustment unit 15, and drive unit 16 are as described above.

[0092] The estimation unit 17 estimates the number of cells included as subjects in the image. For example, the estimation unit 17 detects cells included as subjects in the first image IP acquired by the acquisition unit 12 in response to instructions from the automatic control unit 18, which will be described later, and estimates the number of detected cells. As described above, known methods can be used by the estimation unit 17 to estimate the number of cells. The estimation unit 17 stores cell number information NCI, which indicates the estimated number of cells, and cell position information CLI, which indicates the location of the detected cells, in the storage unit 20A. For example, the estimation unit 17 sets the center of the second image EP as the center of the two-dimensional coordinate system and generates cell position information CLI, which indicates the coordinates of the location of the detected cells.

[0093] The classification unit 13 acquires the second image EP and cell location information CLI stored in the memory unit 20A, and classifies whether the cells included as subjects in the second image EP and located at the positions indicated by the cell location information CLI are benign or malignant. The classification unit 13 performs the classification process in response to instructions from the automatic control unit 18, which will be described later, as an example. The method by which the classification unit 13 classifies whether the cells included as subjects in the second image EP are benign or malignant is as described above. The classification unit 13 associates the cell location information CLI, which indicates the position of the classified cells, with the classification result CR, which indicates the result of the classification, and stores them in the memory unit 20A.

[0094] The input receiving unit 19 receives instructions from the user. Specifically, the input receiving unit 19 acquires instruction information indicating the operation received from the user, which is supplied from the input unit 23. The input receiving unit 19 supplies the acquired instruction information to the automatic control unit 18.

[0095] (Processing performed by the automatic control unit 18 in this exemplary embodiment) In addition to the processing in the exemplary embodiment described above, the automatic control unit 18, when the number of cells estimated by the estimation unit 17 is 1 or more, performs an automatic display process instructing the drive unit 16 to move the imaging range so that one of the cells detected by the estimation unit 17 is at the center, and to display the image acquired by the acquisition unit 12 on the display device.

[0096] Here, the automatic control unit 18 may be configured to perform automatic display processing depending on user input. For example, the automatic control unit 18 displays a button (for example, a "Detailed Display" button) that accepts an instruction to display a second image EP in which any of the detected cells are included in the center of the imaging range as the subject. If the instruction information acquired by the input reception unit 19 indicates that the button has been pressed, the automatic control unit 18 executes the automatic display processing.

[0097] More specifically regarding the automatic display process, the automatic control unit 18 refers to the cell position information CLI stored in the memory unit 20A and instructs the drive unit 16 to move the imaging unit 11 so that the position indicated by the cell position information CLI becomes the center of the imaging range. The drive unit 16 moves the imaging unit 11 according to the instructions from the automatic control unit 18.

[0098] Furthermore, the automatic control unit 18 instructs the adjustment unit 15 to adjust the imaging magnification to the second magnification. The adjustment unit 15 adjusts the imaging magnification of the imaging unit 11 to the second magnification in accordance with the instruction from the automatic control unit 18.

[0099] The automatic control unit 18 then instructs the acquisition unit 12 to acquire the second image EP. The acquisition unit 12 acquires the second image EP captured by the imaging unit 11 after movement and after adjustment of imaging magnification, in accordance with the instructions from the automatic control unit 18. The automatic control unit 18 then displays the image data representing the second image EP on the display device via the output unit 22.

[0100] (Classification method S3 flow) The flow of the classification method S3 executed by the classification device 3 according to this exemplary embodiment will be explained with reference to Figure 7. Figure 7 is a flowchart showing the flow of the classification method S3 according to this exemplary embodiment. As an example, the flowchart shown in Figure 7 is executed after the processing of step S32 in the classification method S2 described above. Also as an example, the flowchart shown in Figure 7 is executed when the user requests to display a second image EP in which any of the detected cells are included in the center of the imaging range as the subject.

[0101] (Step S34) In step S34, the automatic control unit 18 obtains cell count information NCI from the storage unit 20A and determines whether or not cells are present in the imaging range.

[0102] If it is determined in step S34 that no cells are present in the imaging area (step S34: No), the classification device 3 terminates the process shown in Figure 7.

[0103] (Step S35) In step S34, if it is determined that cells are present in the imaging range (step S34: Yes), in step S35, the automatic control unit 18 refers to the cell position information CLI stored in the memory unit 20A and instructs the drive unit 16 to move the imaging range of the imaging unit 11 so that the cell position indicated by the cell position information CLI is at the center of the imaging range of the imaging unit 11.

[0104] (Step S36) In step S36, the automatic control unit 18 instructs the adjustment unit 15 to adjust the imaging magnification of the imaging unit 11 to the second magnification. The adjustment unit 15 adjusts the imaging magnification of the imaging unit 11 to the second magnification.

[0105] (Step S37) In step S37, the imaging unit 11 captures a second image EP in which the cell is included in the center of the imaging range as the subject.

[0106] (Step S38) In step S38, the automatic control unit 18 instructs the acquisition unit 12 to acquire the second image EP captured by the imaging unit 11. The acquisition unit 12 acquires the second image EP captured by the imaging unit 11. The acquisition unit 12 stores the acquired second image EP in the storage unit 20A.

[0107] (Step S39) In step S38, the automatic control unit 18 acquires a second image EP from the storage unit 20A and displays the acquired second image EP on the display device via the output unit 22.

[0108] (Step S40) In step S40, the automatic control unit 18 performs the automatic display processing shown in steps S35 to S39, and then executes processing according to the instructions received by the input receiving unit 19.

[0109] For example, if the input receiving unit 19 receives instruction information indicating that automatic display processing should be performed so that another cell is at the center of the imaging range, the automatic control unit 18 will perform the automatic display processing shown in steps S35 to S39, and then, in accordance with the instruction received by the input receiving unit 19, perform automatic display processing so that another cell is at the center of the imaging range.

[0110] For example, the automatic control unit 18 displays a button (e.g., a "Next" button) superimposed on the second image EP displayed in step S39, which allows the user to instruct the system to display a second image EP in which another cell is included as the subject in the center of the imaging range. When the input reception unit 19 receives instruction information indicating that the button has been pressed, the automatic control unit 18 refers to the cell position information CLI stored in the storage unit 20A and performs automatic display processing so that the other cell is displayed in the center of the imaging range.

[0111] As another example, if the user instruction received by the input reception unit 19 is an instruction to automatically perform display processing so that a different cell is at the center of the imaging range each time a predetermined period of time has elapsed, the automatic control unit 18 will perform automatic display processing so that a different cell is at the center of the imaging range each time a predetermined period of time has elapsed.

[0112] For example, the automatic control unit 18 displays a button (e.g., an "Automatic Display" button) superimposed on the second image EP displayed in step S39, which receives a command from the user to automatically display a second image EP in which another cell is included as the subject in the center of the imaging range every predetermined period of time (e.g., 10 seconds). When the input receiving unit 19 obtains instruction information indicating that the button has been pressed, the automatic control unit 18 refers to the cell position information CLI stored in the storage unit 20A, and after the predetermined period of time has elapsed, performs automatic display processing so that another cell is in the center of the imaging range.

[0113] As another example, if the user instruction received by the input reception unit 19 is to display only cells classified as malignant, the automatic control unit 18 will perform automatic display processing so that the cells classified as malignant are centered in the imaging range. In other words, the automatic control unit 18 will not perform automatic display processing so that the cells classified as benign are centered in the imaging range.

[0114] For example, the automatic control unit 18 overlays a button (for example, a "Show only malignant cells" button) on the second image EP displayed in step S39, which allows the user to instruct the system to display only cells classified as malignant. When the input reception unit 19 receives instruction information indicating that the button has been pressed, the automatic control unit 18 refers to the classification result CR stored in the memory unit 20A. If the classification result CR indicates that the cells have been classified as malignant, the automatic control unit 18 refers to the cell position information CLI associated with the classification result CR. The automatic control unit 18 then instructs the drive unit 16 to move the imaging range so that the cells classified as malignant by the classification unit 13 are at the center. Furthermore, the automatic control unit 18 performs an automatic display process to display the second image EP acquired by the acquisition unit 12 on the display device.

[0115] (Example of the displayed image 1) An example of an image displayed on the display device in this exemplary embodiment will be explained with reference to Figure 8. Figure 8 is an example of an image displayed on the display device in this exemplary embodiment.

[0116] The left-hand image in Figure 8 is a first image IP1 containing multiple cells as subjects, and is captured at a first magnification. The first image IP1 includes regions RE1 to RE6, each containing a different cell. When the input receiving unit 19 acquires instruction information indicating that a button (for example, the "Detailed Display" button) has been pressed to display a second image EP in which any of the detected cells are included in the center of the imaging range as subjects, the automatic control unit 18 displays the second image EP in which any of the cells are included in the center of the imaging range as subjects. The automatic control unit 18 displays the second image EP at a second magnification, which is a magnification higher than the first magnification.

[0117] The middle image in Figure 8 is the second image EP1, which was captured so that the cells included in region RE1 are at the center of the imaging range. When the input receiving unit 19 receives instruction information from the user indicating that a button (for example, the "Next" button) has been pressed to display the second image EP in which another cell is included as the subject at the center of the imaging range, the automatic control unit 18 displays the second image EP in which the other cell is included as the subject at the center of the imaging range.

[0118] The image on the right in Figure 8 is a second image, EP2, which was imaged with another cell within region RE2 at the center.

[0119] As described above, the classification device 3 according to this exemplary embodiment employs a configuration that automatically displays a second image EP, which is captured so that the cells included as subjects in the first image IP are at the center of the imaging range. Therefore, the classification device 3 according to this exemplary embodiment allows the user to confirm whether or not the classification result was correct. In addition, since the cells are automatically displayed at the center in the second image EP, the user's verification process can be made easier.

[0120] [Exemplary Embodiment 4] A fourth exemplary embodiment of the present invention will be described in detail with reference to the drawings. Components having the same function as those described in the above-described exemplary embodiments will be denoted by the same reference numerals, and their descriptions will be omitted as appropriate.

[0121] (Classifier 4) Classification device 4 is a device that classifies whether the cells contained in a pathology sample are benign or malignant. More specifically, the imaging range in classification device 4, in which the pathology sample is imaged at a first imaging magnification, is changed by the user. If there is no operation to move the imaging range in classification device 4 for a predetermined period of time, classification device 4 acquires a second image EP at a second imaging magnification that is higher than the first imaging magnification. Classification device 4 then classifies whether the cells contained as subjects in the acquired second image EP are benign or malignant.

[0122] (Configuration of Classification Device 4) The configuration of the classification device 4 according to this exemplary embodiment will be described with reference to Figure 9. Figure 9 is a block diagram showing the configuration of the classification device 4 according to this exemplary embodiment.

[0123] As shown in Figure 9, the classification device 4 comprises a control unit 10B, an imaging unit 11, a storage unit 20B, an output unit 22, and an input unit 23. The imaging unit 11 is configured to implement the imaging means in this exemplary embodiment. The imaging unit 11, the output unit 22, and the input unit 23 are as described above.

[0124] As an example, the memory unit 20B stores the second image EP and the classification result CR described above.

[0125] (Control Unit 10B) The control unit 10B controls each component of the classification device 4. For example, it stores data in the storage unit 20B and supplies data to the output unit 22.

[0126] As shown in Figure 9, the control unit 10B includes an acquisition unit 12, a classification unit 13, an adjustment unit 15, a drive unit 16, an automatic control unit 18, and an input receiving unit 19. In this exemplary embodiment, the acquisition unit 12, classification unit 13, adjustment unit 15, drive unit 16, automatic control unit 18, and input receiving unit 19 are configured to realize an acquisition means, a classification means, an adjustment means, a drive means, a control means, and an input receiving means, respectively. The acquisition unit 12 and the classification unit 13 are as described above.

[0127] The adjustment unit 15 adjusts the imaging magnification of the imaging unit 11. For example, the adjustment unit 15 adjusts the imaging magnification of the imaging unit 11 to a first magnification or a second magnification in response to instructions from the automatic control unit 18, which will be described later.

[0128] The drive unit 16 moves the imaging range of the imaging unit 11. For example, the drive unit 16 moves the imaging range of the imaging unit 11 in response to instructions from the automatic control unit 18, which will be described later.

[0129] The input receiving unit 19 receives instructions from the user. Specifically, when the input unit 23 receives instructions from the user to move the imaging range of the imaging unit 11 and to change the imaging magnification of the imaging unit 11, the input receiving unit 19 acquires instruction information indicating the operation received by the input unit 23. The input receiving unit 19 supplies the acquired instruction information to the automatic control unit 18.

[0130] In addition to the functions described above, the automatic control unit 18 acquires instruction information supplied by the input reception unit 19. The automatic control unit 18 moves the imaging range in accordance with the user instructions indicated by the acquired instruction information. The automatic control unit 18 also adjusts the imaging magnification in accordance with the user instructions indicated by the acquired instruction information.

[0131] For example, if the instruction information indicates that the imaging range of the imaging unit 11 should be changed, the automatic control unit 18 instructs the drive unit 16 to move the imaging range. Also, if the instruction information indicates that the imaging magnification should be adjusted from the first magnification to the second magnification, the automatic control unit 18 instructs the adjustment unit 15 to adjust the imaging magnification from the first magnification to the second magnification.

[0132] Furthermore, the automatic control unit 18 determines whether the input receiving unit 19 has acquired input information indicating user operation from the input unit 23 within a predetermined period (for example, 10 seconds). As an example, the automatic control unit 18 determines whether, after acquiring input information indicating a movement of the imaging range, it has acquired further input information indicating a movement of the imaging range from the input receiving unit 19 within a predetermined period.

[0133] As an example, if the input receiving unit 19 does not receive an instruction from the drive unit 16 to move the imaging range within a predetermined period, the automatic control unit 18 will cause the adjustment unit 15 to adjust the imaging magnification to a second magnification, the acquisition unit 12 to acquire a second image EP, and the classification unit 13 to classify whether the cells included as subjects in the second image EP are benign or malignant.

[0134] (Classification method S4 flow) The flow of the classification method S4 performed by the classification device 4 according to this exemplary embodiment will be explained with reference to Figure 10. Figure 10 is a flowchart showing the flow of the classification method S4 according to this exemplary embodiment.

[0135] (Step S51) In step S51, the automatic control unit 18 determines whether or not it has received an operation to move the imaging range within a predetermined period of time.

[0136] If it is determined in step S51 that an operation to move the imaging range has been received within a predetermined period (step S51: Yes), the classification device 4 returns to the process of step S51.

[0137] (Step S52) On the other hand, if it is determined in step S51 that no operation to move the imaging range has been received within a predetermined period (step S51: No), in step S52 the automatic control unit 18 instructs the adjustment unit 15 to adjust the imaging magnification from the first magnification to the second magnification. In other words, when the user moves the imaging range and stops searching for cells, the automatic control unit 18 starts the process of acquiring a second image EP in the imaging range at the time the movement stopped.

[0138] (Step S53) In step S53, the imaging unit 11 captures a second image EP.

[0139] (Step S54) In step S54, the automatic control unit 18 instructs the acquisition unit 12 to acquire the second image EP captured by the imaging unit 11. The acquisition unit 12 acquires the second image EP captured by the imaging unit 11. The acquisition unit 12 stores the acquired second image EP in the storage unit 20B.

[0140] (Step S55) In step S55, the automatic control unit 18 instructs the classification unit 13 to perform the classification process. The classification unit 13 acquires the second image EP stored in the memory unit 20B and classifies whether the cells included as subjects in the second image EP are benign or malignant.

[0141] (Example of the displayed image 2) An example of an image displayed on the display device in this exemplary embodiment will be explained with reference to Figure 11. Figure 11 is an example of an image displayed on the display device in this exemplary embodiment.

[0142] The left-hand image in Figure 11 is the first image IP3, showing the user performing an operation to change the imaging range. As shown in the left-hand image in Figure 11, when the user changes the imaging range, a wider imaging range is preferable, so the first image IP3 is captured at a first magnification, which is lower than the second magnification.

[0143] If the automatic control unit 18 determines that it has not received an operation to move the imaging range within a predetermined period while the first image IP3 is displayed, the automatic control unit 18 instructs the acquisition unit 12 to acquire the second image EP3 captured by the imaging unit 11, as shown in the right-hand diagram of Figure 11, and displays the second image EP3 on the display device.

[0144] As described above, in the classification device 4 according to this exemplary embodiment, the imaging range is changed by user operation, and if it is determined that the user has not performed an operation to change the imaging range for a predetermined period of time, a second image EP is acquired, and the cells included as subjects in the second image EP are classified as benign or malignant.

[0145] Therefore, the classification device 4 according to this exemplary embodiment follows the user's operation until cell detection, and then classifies whether the cells detected by the user are benign or malignant. Thus, the classification device 4 according to this exemplary embodiment can reduce the user's workload.

[0146] (modified version) A modification of the fourth exemplary embodiment of the present invention will be described in detail with reference to the drawings. Components having the same function as those described in the above-described exemplary embodiments will be denoted by the same reference numerals, and their descriptions will be omitted as appropriate.

[0147] In this modified example, in the classification device 4, when the input receiving unit 19 receives an instruction from the adjustment unit 15 to adjust the imaging magnification to a second magnification, the automatic control unit 18 causes the acquisition unit 12 to acquire a second image EP, and causes the classification unit 13 to classify whether the cells included as subjects in the second image EP are benign or malignant.

[0148] (Classification method S4A flow) The flow of the classification method S4A performed by the classification device 4 according to this modified example will be explained with reference to Figure 12. Figure 12 is a flowchart showing the flow of the classification method S4A according to this modified example.

[0149] (Step S61) In step S61, the automatic control unit 18 determines whether or not it has received an instruction from the input receiving unit 19 to increase the imaging magnification.

[0150] If it is determined in step S61 that the instruction to increase the imaging magnification has not been received (step S61: No), the classification device 4 returns to the process of step S61.

[0151] (Step S52) On the other hand, if it is determined in step S61 that an instruction to increase the imaging magnification has been received (step S61: Yes), step S 52 In this process, the automatic control unit 18 instructs the adjustment unit 15 to adjust the imaging magnification from the first magnification to the second magnification.

[0152] (Steps S53 to S55) The process by which the imaging unit 11 captures a second image EP and the classification unit 13 classifies whether the cells included as subjects in the second image EP are benign or malignant is as described above.

[0153] As described above, when the classification device 4 according to this modified example receives a user's operation to change the imaging magnification, it acquires a second image EP and classifies whether the cells included as subjects in the second image EP are benign or malignant.

[0154] Therefore, the classification device 4 according to this modified example follows the user's operation until cell detection occurs, and then classifies whether the cells detected by the user are benign or malignant. Thus, the classification device 4 according to this exemplary embodiment can reduce the user's workload.

[0155] [Examples of implementation using software] Some or all of the functions of classification devices 1 to 4 may be implemented by hardware such as integrated circuits (IC chips), or by software.

[0156] In the latter case, the classification devices 1 to 4 are implemented by a computer that executes instructions for a program, which is software that implements each function. An example of such a computer (hereinafter referred to as computer C) is shown in Figure 13. Computer C comprises at least one processor C1 and at least one memory C2. The memory C2 stores a program P that causes computer C to operate as classification devices 1 to 4. In computer C, the processor C1 reads program P from memory C2 and executes it, thereby implementing each function of the classification devices 1 to 4.

[0157] Processor C1 can include, for example, a CPU (Central Processing Unit), GPU (Graphic Processing Unit), DSP (Digital Signal Processor), MPU (Micro Processing Unit), FPU (Floating Point Number Processing Unit), PPU (Physics Processing Unit), microcontroller, or a combination thereof. Memory C2 can include, for example, flash memory, HDD (Hard Disk Drive), SSD (Solid State Drive), or a combination thereof.

[0158] Computer C may also be equipped with RAM (Random Access Memory) for loading program P at runtime and for temporarily storing various data. Furthermore, computer C may be equipped with communication interfaces for sending and receiving data with other devices. Additionally, computer C may be equipped with input / output interfaces for connecting input / output devices such as keyboards, mice, displays, and printers.

[0159] Furthermore, program P can be recorded on a non-temporary, tangible recording medium M that is readable by computer C. Such a recording medium M could be, for example, tape, disk, card, semiconductor memory, or programmable logic circuitry. Computer C can acquire program P via such a recording medium M. Program P can also be transmitted via a transmission medium. Such a transmission medium could be, for example, a communication network or broadcast waves. Computer C can also acquire program P via such a transmission medium.

[0160] [Additional Note 1] The present invention is not limited to the embodiments described above, and various modifications are possible within the scope of the claims. For example, embodiments obtained by appropriately combining the technical means disclosed in the embodiments described above are also included in the technical scope of the present invention.

[0161] [Additional Note 2] Some or all of the embodiments described above may also be described as follows. However, the present invention is not limited to the embodiments described below.

[0162] (Note 1) A classification device comprising: an imaging means for capturing an image that includes a portion of a pathological sample as the subject; an acquisition means for acquiring the image captured by the imaging means; and a classification means for classifying whether the cells included as the subject in the image acquired by the acquisition means are benign or malignant.

[0163] The above configuration can improve the accuracy of classifying pathological samples as benign or malignant.

[0164] (Note 2) The classification device according to Appendix 1, further comprising: an adjustment means for adjusting the imaging magnification of the imaging means; an estimation means for estimating the number of cells included as subjects in an image acquired by the acquisition means; and a control means for performing an automatic classification process, which includes (i) causing the adjustment means to adjust the imaging magnification to a first magnification; (ii) causing the acquisition means to acquire a first image at the first magnification; (iii) causing the estimation means to estimate the number of cells included as subjects in the first image; and (iv) if the number of cells estimated by the estimation means is greater than or equal to a predetermined number, causing the adjustment means to adjust the imaging magnification to a second magnification higher than the first magnification, causing the acquisition means to acquire a second image at the second magnification, and causing the classification means to classify whether the cells included as subjects in the second image are benign or malignant.

[0165] With the above configuration, the same classification results can be obtained regardless of the user's skill level when using the classification device.

[0166] (Note 3) The classification apparatus according to Appendix 2, further comprising a driving means for moving the imaging range of the imaging means, wherein the control means, if the number of cells classified as malignant by the classification means as a result of the automatic classification process is less than a predetermined number, moves the imaging range to the driving means and performs the automatic classification process again.

[0167] The above configuration can improve the accuracy of classifying pathological samples as benign or malignant.

[0168] (Note 4) The control means controls the drive means to move the entire pathology sample comprehensively from the center of the pathology sample. The imaging range The sorting device described in Appendix 3, which is to be moved.

[0169] According to the above configuration, the process of classifying pathological samples as benign or malignant can be performed rapidly.

[0170] (Note 5) The classification apparatus according to Appendix 3 or 4, wherein the control means further performs an automatic display process on the drive means to move the imaging range so that the cells detected by the estimation means are centered, and to display the image acquired by the acquisition means on a display device.

[0171] The above configuration makes it easier for users to perform verification tasks.

[0172] (Note 6) The classification device according to Appendix 5, further comprising a receiving means for receiving instructions from a user, wherein the control means, after performing the automatic display processing, performs the automatic display processing so that another cell is at the center of the imaging range in accordance with the instructions received by the receiving means.

[0173] The above configuration allows for flexible responses to user instructions.

[0174] (Note 7) If the instruction received by the receiving means from the user is an instruction to perform the automatic display processing so that a different cell is at the center of the imaging range each time a predetermined period of time has elapsed, the control means performs the automatic display processing so that a different cell is at the center of the imaging range each time a predetermined period of time has elapsed, the classification device as described in Appendix 6.

[0175] The above configuration allows for flexible responses to user instructions.

[0176] (Note 8) The classification device according to Appendix 6 or 7, wherein if the instruction from the user received by the receiving means is an instruction to display only cells classified as malignant, the control means performs the automatic display process so that the cells classified as malignant are at the center of the imaging range.

[0177] The above configuration allows for flexible responses to user instructions.

[0178] (Note 9) The classification device according to Appendix 1, further comprising: a driving means for moving the imaging range of the imaging means; an adjusting means for adjusting the imaging magnification of the imaging means; a receiving means for receiving instructions from a user; and a control means for causing the adjusting means to adjust the imaging magnification to a first magnification or a second magnification higher than the first magnification, and for causing the driving means to move the imaging range in response to instructions from a user received by the receiving means.

[0179] The above configuration allows for flexible responses to user instructions.

[0180] (Note 10) The classification apparatus according to Appendix 9, wherein the control means, if the receiving means does not receive an instruction from the driving means to move the imaging range for a predetermined period of time, causes the adjustment means to adjust the imaging magnification to a second magnification higher than the first magnification, causes the acquisition means to acquire a second image at the second magnification, and causes the classification means to perform an automatic classification process to classify whether the cells included as subjects in the second image are benign or malignant.

[0181] The above configuration can reduce the effort required from the user.

[0182] (Note 11) The classification apparatus as described in Appendix 9, wherein the control means, when the receiving means receives an instruction from the adjustment means to adjust the imaging magnification to a second magnification higher than the first magnification, causes the acquisition means to acquire a second image at the second magnification, and causes the classification means to classify whether the cells included as subjects in the second image are benign or malignant.

[0183] The above configuration can reduce the effort required from the user.

[0184] (Note 12) A classification method comprising: capturing an image that includes a portion of a pathological sample as the subject; acquiring the image captured in the capturing process; and classifying whether the cells included as the subject in the acquired image are benign or malignant.

[0185] The above method can improve the accuracy of classifying pathological samples as benign or malignant.

[0186] (Note 13) A program for causing a computer to function as a classification device, comprising: an imaging means for capturing an image that includes a portion of a pathological sample as the subject; an acquisition means for acquiring the image captured by the imaging means; and a classification means for classifying whether the cells included as the subject in the image acquired by the acquisition means are benign or malignant.

[0187] The above method can improve the accuracy of classifying pathological samples as benign or malignant.

[0188] [Additional Note 3] Some or all of the embodiments described above can also be expressed as follows:

[0189] A classification device comprising at least one processor, wherein the processor performs an imaging process for capturing an image that includes a portion of a pathological sample as the subject; an acquisition process for acquiring the image captured by the imaging process; and a classification process for classifying whether the cells included as the subject in the image acquired by the acquisition process are benign or malignant.

[0190] Furthermore, this classification device may also be equipped with memory, and this memory contains the aforementioned acquisition Processing and the above classification process and A program to be executed by the processor may be stored. Furthermore, this program may be recorded on a computer-readable, non-temporary, tangible recording medium. [Explanation of symbols]

[0191] 1, 2, 3, 4 Classifier 10, 10A, 10B Control Unit 11 Imaging Unit 12 Acquisition Department 13 Classification section 15 Adjustment part 16 Drive unit 17 Estimation part 18 Automatic Control Unit 19 Input Reception Section 20, 20A, 20B storage section 22 Output section 23 Input section

Claims

1. An imaging unit that captures an image that includes a portion of a pathological sample as the subject, An acquisition means for acquiring an image captured by the imaging unit, A classification means that classifies whether cells included as subjects in an image are benign or malignant, using a pre-trained model trained to classify whether cells included as subjects in an image are benign or malignant, An estimation means for estimating the number of cells included as subjects in the image acquired by the acquisition means, An adjustment means for adjusting the imaging magnification of the imaging unit, A driving means for moving the imaging range of the imaging unit, Control means for controlling the acquisition means, the classification means, the estimation means, the adjustment means, and the driving means, Equipped with, The control means is (i) The adjusting means adjusts the imaging magnification to the first magnification, (ii) The acquisition means is made to acquire a first image of the imaging range at the first magnification, (iii) The estimation means is made to estimate the number of cells included as subjects in the first image, (iv) If the number of cells estimated by the estimation means is greater than or equal to a predetermined number, the adjustment means adjusts the imaging magnification to a second magnification higher than the first magnification, the acquisition means acquires a second image of the imaging range at the second magnification, and the classification means classifies whether the cells included as subjects in the second image are benign or malignant; if the number of cells estimated by the estimation means is less than a predetermined number, the adjustment to the second magnification by the adjustment means and the classification by the classification means are not performed, and the driving means moves the imaging range in the pathology sample. Classification device.

2. The control means, when the number of cells classified as malignant by the classification means is less than a predetermined number, causes the drive means to move the imaging range in the pathology sample. The classification device according to claim 1.

3. The control means moves the driving means to move the imaging range from the center of the pathological sample to cover the entire pathological sample. The classification device according to claim 2.

4. The control means further performs an automatic display process in which the driving means moves the imaging range so that the cells detected by the estimation means are centered, and displays the image acquired by the acquisition means on a display device. The classification device according to claim 2 or 3.

5. Further mechanisms for receiving instructions from users will be added. After performing the automatic display processing, the control means performs the automatic display processing in accordance with the instruction received by the receiving means so that another cell is at the center of the imaging range. The classification device according to claim 4.

6. If the instruction received by the receiving means from the user is an instruction to perform the automatic display process so that a different cell is at the center of the imaging range each time a predetermined period of time has elapsed, The control means performs the automatic display process so that another cell is at the center of the imaging range each time the predetermined period has elapsed. The classification device according to claim 5.

7. If the instruction received by the aforementioned receiving means from the user is an instruction to display only cells classified as malignant, The control means performs the automatic display process so that the cells classified as malignant are centered within the imaging range. The classification device according to claim 5 or 6.

8. For each imaging range used to image the pathological sample (i) Adjust the imaging magnification to the first magnification, (ii) A first image of the imaging range is acquired at the first magnification, (iii) Estimate the number of cells included as subjects in the first image, (iv) If the estimated number of cells is greater than or equal to a predetermined number, the imaging magnification is adjusted to a second magnification higher than the first magnification, a second image of the imaging range is acquired at the second magnification, and the cells included as subjects in the second image are classified as benign or malignant using a trained model that has been trained to classify whether the cells included as subjects in the image are benign or malignant; if the estimated number of cells is less than the predetermined number, the adjustment to the second magnification and the classification are not performed, and the imaging range in the pathology sample is moved. Classification method.

9. A program for causing a computer to function as a classification device equipped with an imaging unit that captures an image that includes a portion of a pathological sample as the subject, An acquisition means for acquiring an image captured by the imaging unit, A classification means that classifies whether cells included as subjects in an image are benign or malignant, using a pre-trained model trained to classify whether cells included as subjects in an image are benign or malignant, An estimation means for estimating the number of cells included as subjects in the image acquired by the acquisition means, An adjustment means for adjusting the imaging magnification of the imaging unit, A driving means for moving the imaging range of the imaging unit, Control means for controlling the acquisition means, the classification means, the estimation means, the adjustment means, and the driving means, To make it function as, The control means is (i) The adjusting means adjusts the imaging magnification to the first magnification, (ii) The acquisition means is made to acquire a first image of the imaging range at the first magnification, (iii) The estimation means is made to estimate the number of cells included as subjects in the first image, (iv) If the number of cells estimated by the estimation means is greater than or equal to a predetermined number, the adjustment means adjusts the imaging magnification to a second magnification higher than the first magnification, the acquisition means acquires a second image of the imaging range at the second magnification, and the classification means classifies whether the cells included as subjects in the second image are benign or malignant; if the number of cells estimated by the estimation means is less than a predetermined number, the adjustment to the second magnification by the adjustment means and the classification by the classification means are not performed, and the driving means moves the imaging range in the pathology sample. program.

Images