Information processing device and program

Abstract

[Problem] To provide an information processing device and program capable of supporting a diagnosis performed by a doctor, by automatically identifying a lesioned portion from the result of a positron emission tomography (PET) examination. [Solution] One aspect of the present invention provides an information processing device for identifying a lesioned portion on the basis of a positron emission tomography examination. The information processing device is provided with an image acquiring unit, and a lesion identifying unit. The image acquiring unit is configured to be capable of acquiring an image including an anatomical region imaged by a positron emission tomography device. The lesion identifying unit is configured to be capable of identifying a lesioned portion from a part of the image in which positron-emitting radionuclides are accumulated, on the basis of trained data obtained by machine learning.

Description

[Technical Field]

[0001] The present invention relates to an information processing device and a program. [Background technology]

[0002] In recent years, a technology has been proposed that uses a computer to assist diagnosis based on images of test results (see, for example, Patent Document 1).

[0003] Recently, in the medical field as well, there has been a demand for using deep learning and other methods to recognize image features, and technologies to support this have been proposed (see, for example, Patent Document 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Special Publication No. 2009-516551 [Patent Document 2] Japanese Patent Application Publication No. 2019-010411 Summary of the Invention [Problem to be solved by the invention]

[0005] One imaging test is PET (positron emission tomography), which uses a drug called FDG (fluorodeoxyglucose), a glucose analog molecule labeled with the positron-emitting nuclide F-18. This PET scan accumulates in areas of disease, such as cancer and inflammation, making it easier to detect the lesions from the images.

[0006] However, because organs such as the brain and heart have active glucose metabolism, physiological accumulation of FDG is observed. FDG is also excreted in urine, so it accumulates in the kidneys, ureters, and bladder. Therefore, it is difficult to automatically extract lesions in PET scans. This makes it impossible to quantify the extent of lesions throughout the body.

[0007] In view of the above circumstances, the present invention provides an information processing device and a program that can automatically identify a lesion from the results of a PET examination and assist a doctor in making a diagnosis. [Means for solving the problem]

[0008] According to one aspect of the present invention, there is provided an information processing device that identifies a lesion based on a positron emission tomography examination. The information processing device includes an image acquisition unit and a lesion identification unit. The image acquisition unit is configured to acquire an image including an anatomical region captured by a positron emission tomography device. The lesion identification unit is configured to identify a lesion from a portion of the image where positron-emitting nuclides have accumulated, based on learned data obtained by machine learning.

[0009] According to one aspect of the present invention, it is possible to assist in diagnosis based on the results of a PET examination, thereby reducing the burden on doctors. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram showing an outline of the configuration of an information processing device 1 according to an embodiment of the present invention. [Figure 2] 1 is a block diagram showing a functional configuration of an information processing device 100. FIG. [Figure 3] FIG. 2 is a block diagram showing the functional configuration of an information processing device 110. [Figure 4] FIG. 2 is a block diagram showing the functional configuration of an information processing device 120. [Figure 5] 1 shows an example of an image captured by the positron emission tomography apparatus 2. [Figure 6]10 shows an example of an image of a lesion identified by the lesion identifying unit 102. [Figure 7] 1 shows an example of an anatomical region identified by the region identifying unit 104. [Figure 8] FIG. 2 is an activity diagram showing the flow of operations of the information processing device 100. [Figure 9] FIG. 10 is a diagram for explaining teacher data. [Figure 10] FIG. 10 is a diagram for explaining teacher data. [Figure 11] FIG. 10 is a diagram for explaining teacher data. [Figure 12] FIG. 10 is a diagram for explaining teacher data. [Figure 13] FIG. 10 is a diagram for explaining teacher data. [Figure 14] FIG. 10 is a diagram for explaining teacher data. [Figure 15] FIG. 10 is a diagram for explaining teacher data. [Figure 16] FIG. 10 is a diagram for explaining teacher data. [Figure 17] FIG. 10 is a diagram for explaining teacher data. [Figure 18] FIG. 10 is a diagram for explaining teacher data. [Figure 19] 10 is a diagram showing an example of a lesion identified by the lesion identifying unit 102. FIG. [Figure 20] 10 is a diagram showing an example of a lesion identified by the lesion identifying unit 102. FIG. [Figure 21] FIG. 1 is a diagram for explaining Case 1. [Figure 22] FIG. 1 is a diagram for explaining Case 1. [Figure 23] FIG. 1 is a diagram for explaining Case 1. [Figure 24] FIG. 1 is a diagram for explaining Case 1. [Figure 25] FIG. 1 is a diagram for explaining Case 1. [Figure 26] FIG. 1 is a diagram for explaining Case 1. [Figure 27] FIG. 10 is a diagram for explaining Case 2. [Figure 28] FIG. 10 is a diagram for explaining Case 2. [Figure 29] FIG. 10 is a diagram for explaining Case 2. [Figure 30] FIG. 10 is a diagram for explaining Case 2. [Figure 31] FIG. 10 is a diagram for explaining Case 3. [Figure 32] FIG. 10 is a diagram for explaining Case 3. [Figure 33] FIG. 10 is a diagram for explaining Case 3. [Figure 34] FIG. 10 is a diagram for explaining Case 3. DETAILED DESCRIPTION OF THE INVENTION

[0011] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the accompanying drawings. Various features shown in the following embodiments can be combined with each other.

[0012] Incidentally, the program for realizing the software appearing in this embodiment may be provided as a non-transitory computer-readable medium, or may be provided so that it can be downloaded from an external server, or may be provided so that the program is started on an external computer and its functions are realized on a client terminal (so-called cloud computing).

[0013] In this embodiment, the term "unit" may also include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In addition, various types of information are handled in this embodiment, and this information may be represented by, for example, physical values ​​of signal values ​​representing voltages and currents, high and low signal values ​​as a binary bit set consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculations may be performed on a circuit in the broad sense.

[0014] In addition, a circuit in the broad sense is a circuit realized by at least appropriately combining a circuit, circuitry, a processor, a memory, etc. That is, it includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)), etc.

[0015] 1. Configuration of information processing device 1 is a diagram showing an outline of the configuration of an information processing device 1 according to an embodiment of the present invention. As shown in the diagram, the information processing device 1 has a processing unit 11, a storage unit 12, a temporary storage unit 13, an external device connection unit 14, and a communication unit 15, and these components are electrically connected within the information processing device 1 via a communication bus 16.

[0016] The processing unit 11 is realized by, for example, a central processing unit (CPU), and operates according to a predetermined program stored in the storage unit 12 to realize various functions.

[0017] The storage unit 12 is a non-volatile storage medium that stores various information. This is realized by a storage device such as a hard disk drive (HDD) or a solid state drive (SSD). Note that the storage unit 12 can also be arranged in another device that can communicate with the information processing device 1.

[0018] The temporary storage unit 13 is a volatile storage medium, which is realized by a memory such as a random access memory (RAM), and stores information (arguments, arrays, etc.) that is temporarily required when the processing unit 11 operates.

[0019] The external device connection unit 14 is a connection unit that conforms to standards such as Universal Serial Bus (USB) and High-Definition Multimedia Interface (HDMI), and is capable of connecting input devices such as keyboards and display devices such as monitors.

[0020] The communication unit 15 is a communication means conforming to, for example, a local area network (LAN) standard, and realizes communication between the information processing device 1 and the local area network or a network such as the Internet via the local area network.

[0021] It should be noted that the information processing device 1 can be a general-purpose server computer, a personal computer, or the like, and the information processing device 1 can also be configured using a plurality of computers.

[0022] 2. Functions of information processing equipment Next, the functions of the information processing device 1 will be described. Here, it is assumed that the information processing device 1 operates in accordance with a program to realize the information processing device 100 or the information processing device 110. This program causes the information processing device 1, which is a computer, to function as the information processing device 100 or the information processing device 110. Both the information processing device 100 and the information processing device 110 are information processing devices that identify a lesion based on a positron emission tomography examination.

[0023] Fig. 2 is a block diagram showing the functional configuration of the information processing device 100. Fig. 3 is a block diagram showing the functional configuration of the information processing device 110, and Fig. 4 is a block diagram showing the functional configuration of the information processing device 120.

[0024] As shown in FIG. 2, information processing device 100 includes image acquisition unit 101, lesion identification unit 102, data storage unit 103, region identification unit 104, area calculation unit 105, index calculation unit 106, and output unit 107.

[0025] The image acquisition unit 101 is configured to be able to acquire images including an anatomical region captured by the positron emission tomography apparatus 2 from the image storage device 3. The image acquisition unit 101 can also be configured to acquire images directly from the positron emission tomography apparatus 2 without going through the image storage device 3. The images acquired by the image acquisition unit 101 are, for example, as shown in FIG. 5. FIG. 5 shows an example of an image captured by the positron emission tomography apparatus 2.

[0026] An anatomical region is a region that represents a part or the whole of an anatomical structure such as an organ or tissue of the body, and here, the anatomical region is assumed to include the whole body of the subject. Of course, the information processing device 100 can support diagnosis by treating only the chest, only the abdomen, or only the liver as an anatomical region.

[0027] Lesion identification unit 102 is configured to be able to identify a lesion from a portion where positron-emitting nuclides are accumulated in the image acquired by image acquisition unit 101, based on learned data obtained by machine learning. The result of lesion identification by lesion identification unit 102 is, for example, as shown in Figure 6. Figure 6 shows an example image of a lesion identified by lesion identification unit 102.

[0028] The data storage unit 103 stores the learned data. The learned data is generated by machine learning using, as training data, an image including an anatomical region captured by the positron emission tomography device 2 and instruction information indicating a lesion in the image. This instruction information is information indicating the lesion by a specialist. For the machine learning, for example, a deep learning algorithm called semantic segmentation is used. This semantic segmentation associates a label or category with every pixel in an image and is used to recognize a group of pixels that form a distinctive category. Details of the training data will be described later. The learned data stored in the data storage unit 103 may be periodically updated by obtaining the latest data from the learned data providing device 4.

[0029] The region identifying unit 104 is configured to be able to identify an anatomical region in an image. When the anatomical region is the whole body, the identified result is, for example, as shown in Fig. 7. Fig. 7 shows an example of an anatomical region identified by the region identifying unit 104.

[0030] Area calculation unit 105 is configured to be able to calculate the area of ​​the lesion identified by lesion identification unit 102 in the image acquired by image acquisition unit 101 and the area of ​​the anatomical region identified by region identification unit 104. The area of ​​the anatomical region is the projected area of ​​the subject in the image acquired by image acquisition unit 101, and preferably the area of ​​the anatomical region is the frontal projected area of ​​the subject in the image acquired by image acquisition unit 101. Area calculation unit 105 may calculate the area of ​​the lesion based on the contour of the lesion, and then calculate the area of ​​the anatomical region based on the contour of the anatomical region, or may calculate the area of ​​the lesion based on the number of pixels in the lesion, and then calculate the area of ​​the anatomical region based on the number of pixels in the anatomical region.

[0031] Index calculation unit 106 is configured to be able to calculate an index from the area of ​​the lesion identified by lesion identification unit 102 in the image acquired by image acquisition unit 101 and the area of ​​the anatomical region identified by region identification unit 104. The index is the ratio between the area of ​​the lesion and the area of ​​the anatomical region. Specifically, the index is the total area of ​​the lesion divided by the area of ​​the anatomical region; if the total area of ​​the lesion is 529.5 and the area of ​​the whole body, which is the anatomical region, is 32729.0, the index is 0.0162.

[0032] The output unit 107 displays the index calculated by the index calculation unit 106 on a display device (not shown), outputs data such as CSV (Comma Separated Value), or prints it out on paper via a printing device (not shown). At this time, the output unit 107 may output the image acquired by the image acquisition unit 101 together with the index.

[0033] 3, information processing device 110 includes image acquisition unit 111, lesion identification unit 112, data acquisition unit 113, region identification unit 114, area calculation unit 115, index calculation unit 116, and output unit 117. Note that image acquisition unit 111, lesion identification unit 112, region identification unit 114, area calculation unit 115, index calculation unit 116, and output unit 117 have the same functional units as image acquisition unit 101, lesion identification unit 102, region identification unit 104, area calculation unit 105, index calculation unit 106, and output unit 107 of information processing device 100, respectively, and therefore description thereof will be omitted here.

[0034] The data acquisition unit 113 is configured to be able to acquire learned data via a communication network from the learned data providing device 5. The learned data acquired from the learned data providing device 5 is of the same type as the learned data stored in the data storage unit 103 of the information processing device 100 and is the latest learned data.

[0035] As shown in FIG. 4, the information processing device 120 includes an image acquisition unit 121, a lesion identification unit 122, a data storage unit 123, a region identification unit 124, a volume calculation unit 125, an index calculation unit 126, and an output unit 127.

[0036] The image acquisition unit 121 is configured to be able to acquire images including an anatomical region captured by the positron emission tomography apparatus 2 from the image storage device 3. Note that the image acquisition unit 121 can also be configured to acquire images directly from the positron emission tomography apparatus 2 without going through the image storage device 3.

[0037] Lesion identification unit 122 is configured to be able to identify a lesion from a portion in the image acquired by image acquisition unit 121 where positron-emitting nuclides have accumulated, based on learned data obtained by machine learning.

[0038] The data storage unit 123 stores the learned data. The learned data is generated by machine learning using, as training data, an image including an anatomical region captured by the positron emission tomography apparatus 2 and instruction information indicating a lesion in the image. This instruction information indicates a lesion indicated by a specialist.

[0039] The region specifying unit 124 is configured to be able to specify an anatomical region in the image acquired by the image acquiring unit 121. This image is a three-dimensional image.

[0040] Furthermore, volume calculation unit 125 is configured to be able to calculate the volume of the lesion identified by lesion identification unit 122 in the image acquired by image acquisition unit 121 and the volume of the anatomical region identified by region identification unit 124. The volume of the anatomical region is the accumulated volume of the tomographic images of the subject in the image acquired by image acquisition unit 121. The volume calculation unit 125 may calculate the volume of the lesion based on the contour of the lesion, and may calculate the volume of the anatomical region based on the contour of the anatomical region, or may calculate the volume of the lesion based on the number of pixels in the tomographic image of the lesion, and may calculate the volume of the anatomical region based on the number of pixels in the anatomical region.

[0041] Index calculation unit 126 is configured to be able to calculate an index from the volume of the lesion identified by lesion identification unit 122 and the volume of the anatomical region identified by region identification unit 124. The index is the ratio between the volume of the lesion and the volume of the anatomical region.

[0042] The output unit 127 displays the indexes calculated by the index calculation unit 126 on a display device (not shown), outputs data in CSV format or the like, or prints out the data on paper via a printing device (not shown). Note that instead of the data storage unit 123, a data acquisition unit similar to the data acquisition unit 113 of the information processing device 110 may be provided.

[0043] 3. Operation of information processing device Next, a description will be given of the operation of the information processing device 100. Fig. 8 is an activity diagram showing the flow of the operation of the information processing device 100. Note that the operation of the information processing device 110 can be inferred from the operation of the information processing device 100, and therefore a description of the operation of the information processing device 110 will be omitted.

[0044] When the information processing device 100 starts its operation, first, an input unit (not shown) receives input of identification information (A101). The identification information is information for identifying a subject (examinee, patient), a reception number for imaging by the positron emission tomography imaging apparatus 2, etc., and although its operation differs depending on the facility, it is information that can identify an image captured by the positron emission tomography imaging apparatus 2.

[0045] When the information processing device 100 receives the identification information, the image acquisition unit 101 acquires the image captured by the positron emission tomography apparatus 2 from the image storage device 3 (A102).

[0046] Next, in the information processing device 100, the lesion identification unit 102 identifies a lesion from the image captured by the positron emission tomography apparatus 2 (A103), and the area calculation unit 105 calculates the area of ​​the lesion (A104). At this time, the region identification unit 104 identifies an anatomical region (A105), and the area calculation unit 105 calculates the area of ​​the region (A106). Note that these processes do not necessarily need to be performed in parallel, and the processes can also be performed in the order of A103, A104, A105, A106, or A103, A105, A104, A106, etc.

[0047] Then, the index calculation unit 106 calculates an index based on the area of ​​the lesion and the area of ​​the anatomical region (A107), the output unit 107 outputs the index (A108), and the information processing device 100 ends its operation.

[0048] 4. Training data Next, the training data will be described. Figures 9 to 18 are diagrams for explaining the training data. The training data is composed of a set of an image captured by a positron emission tomography apparatus and data in which a specialist or the like has identified an FDG accumulation portion contained in the image as a lesion, for example, marking data in which the lesion portion is marked. Furthermore, in order to increase the accuracy of recognizing FDG accumulation in normal areas such as the brain, heart, kidneys, ureters, and bladder as not being a lesion, blank data without markings is used as training data in place of data without lesions. Note that marking a lesion corresponds to labeling the FDG accumulation portion (setting a label indicating whether or not it is a lesion).

[0049] For example, an image captured by a positron emission tomography apparatus shown in Fig. 9 is paired with the marking data shown in Fig. 10. The marking data corresponds to the areas in the image captured by the positron emission tomography apparatus where FDG has accumulated, excluding mainly the brain BR, kidney KI, and bladder BL.

[0050] Furthermore, the image captured by the positron emission tomography device shown in Fig. 11 is paired with the marking data shown in Fig. 12. The marking data corresponds to the areas where FDG has accumulated in the image captured by the positron emission tomography device, excluding mainly the brain BR, heart HE, kidney KI, and bladder BL. Since the liver LI is basically a place where FDG accumulates weakly, areas with strong FDG accumulation occurring in the liver LI are marked as lesions.

[0051] Furthermore, the image captured by the positron emission tomography device shown in Figure 13 is paired with the marking data shown in Figure 14. The marking data corresponds to the FDG-accumulated areas in the image captured by the positron emission tomography device, mainly excluding the brain BR, kidneys KI, and bladder BL. Furthermore, the non-physiological accumulation area X is not the brain or kidneys where FDG generally accumulates, but is a slight leakage at the FDG injection site, and is not considered by a specialist or other medical professional to be a lesion, so no marking is performed.

[0052] Moreover, the image captured by the positron emission tomography apparatus shown in Fig. 15 is paired with the marking data shown in Fig. 16. The marking data corresponds to the areas where FDG has accumulated in the image captured by the positron emission tomography apparatus, mainly excluding the brain BR, heart HE, kidney KI, and bladder BL.

[0053] Furthermore, the image taken by the positron emission tomography device shown in Fig. 17 is paired with the marking data shown in Fig. 18. The marking data corresponds to the areas where FDG has accumulated in the image taken by the positron emission tomography device, mainly excluding the brain BR, heart HE, kidney KI, and bladder BL, but since it is determined that no lesions exist in the image taken by the positron emission tomography device shown in Fig. 12A, the marking data is blank.

[0054] As is clear from Figures 9 to 18, accumulation of FDG in areas other than the lesion does not necessarily show the same tendency. For example, in the examples shown in Figures 15 and 16, accumulation of FDG in cardiac HE is observed, but in the examples shown in Figures 13 and 14, there are cases where accumulation of FDG in the heart is not observed. Therefore, there are cases where using machine learning is very meaningful.

[0055] In addition, when generating training data, an information processing device is used that has the function of extracting FDG accumulation areas from images taken by a positron emission tomography device and the function of identifying the contours of the extracted accumulation areas, and specialists can create training data simply by pointing within the identified contours using a pointing device such as a mouse or a touch pen.

[0056] For reference, examples of lesions identified by lesion identification unit 102 are shown in Figures 19 and 20. Figures 19 and 20 are diagrams showing examples of lesions identified by lesion identification unit 102. The lesion shown in Figure 19 is the lesion identified by lesion identification unit 102 from the image shown in Figure 9, and the lesion shown in Figure 20 is the lesion identified by lesion identification unit 102 from the image shown in Figure 11.

[0057] 5. Examples of lesion identification Next, three examples of the results of identifying the lesion area by the information processing device 1 will be described. Figs. 21 to 26 are diagrams for explaining Case 1, Figs. 27 to 30 are diagrams for explaining Case 2, and Figs. 31 to 34 are diagrams for explaining Case 3. In all of Cases 1 to 3, FDG-PET was taken during treatment (Interim PET) for the purpose of assessing the therapeutic effect, and LDH (serum lactate dehydrogenase) and IL-2R (interleukin-2 receptor) were also measured.

[0058] 5-1. Case 1 Fig. 21 is an FDG-PET image of Case 1, and Fig. 22 shows the values ​​of LDH and IL-2R. Fig. 23 shows an image in which the lesions were identified by the information processing device 1, and Fig. 24 shows the results of calculating the total area of ​​the lesions, the trunk area, and the TLI from that image. TLI (Total Lesion Index) is the value obtained by dividing the total area of ​​the lesions (see Fig. 25) by the trunk area (see Fig. 26).

[0059] 5-2. Case 2 Fig. 27 shows an FDG-PET image of Case 2, and Fig. 28 shows the values ​​of LDH and IL-2R. Fig. 29 shows an image in which the lesions were identified by the information processing device 1, and Fig. 30 shows the results of calculating the total area of ​​the lesions, the trunk area, and the TLI from the image.

[0060] Case 3 Fig. 31 shows an FDG-PET image of Case 3, and Fig. 32 shows the values ​​of LDH and IL-2R. Fig. 33 shows an image in which the lesions were identified by the information processing device 1, and Fig. 34 shows the results of calculating the total area of ​​the lesions, the trunk area, and the TLI from the image.

[0061] 6.Other The present invention may be provided in the following aspects. The information processing device comprises a region identification unit, a volume calculation unit, and an index calculation unit, wherein the region identification unit is configured to be able to identify an anatomical region in the image, the image is a three-dimensional image, the volume calculation unit is configured to be able to calculate the volume of the lesion and the volume of the anatomical region, and the index calculation unit is configured to be able to calculate an index, wherein the index is a ratio between the volume of the lesion and the volume of the anatomical region. In the information processing device, the volume of the anatomical region is an integrated volume of tomographic images of the subject in the image. In the information processing device, the volume calculation unit calculates the volume of the lesion based on the contour of the lesion, and calculates the volume of the anatomical region based on the contour of the anatomical region. In the information processing device, the volume calculation unit calculates the volume of the lesion based on the number of pixels of a tomographic image of the lesion, and calculates the volume of the anatomical region based on the number of pixels of the anatomical region. The information processing device comprises a region identification unit, an area calculation unit, and an index calculation unit, wherein the region identification unit is configured to be able to identify an anatomical region in the image, the area calculation unit is configured to be able to calculate the area of ​​the lesion and the area of ​​the anatomical region, the area of ​​the anatomical region being a projected area of ​​the subject in the image, and the index calculation unit is configured to be able to calculate an index, wherein the index is a ratio between the area of ​​the lesion and the area of ​​the anatomical region. In the information processing device, the area of ​​the anatomical region is a frontal projection area of ​​the subject in the image. In the information processing device, the area calculation unit calculates the area of ​​the lesion based on the contour of the lesion, and calculates the area of ​​the anatomical region based on the contour of the anatomical region. In the information processing device, the area calculation unit calculates the area of ​​the lesion based on the number of pixels in the lesion, and calculates the area of ​​the anatomical region based on the number of pixels in the anatomical region. In the information processing device, the anatomical region includes the entire body of the subject. The information processing device further comprises a data storage unit, wherein the data storage unit stores the learned data. The information processing device includes a data acquisition unit, wherein the data acquisition unit is configured to be able to acquire the learned data from a learned data providing device via a communication network. In the information processing device, the learned data is generated by machine learning using as training data an image including an anatomical region captured by a positron emission tomography device and instruction information indicating a lesion in the image. In the information processing device, the instruction information is an indication of a lesion area by a specialist. A program that causes a computer to operate as an information processing device, the program causing the computer to function as the information processing device. Of course, this is not the case.

[0062] The program may also be provided as a computer-readable non-transitory recording medium that stores the program.

[0063] As described above, according to the present invention, since the lesion can be automatically identified from the test results of the positron emission tomography apparatus, even non-specialists can easily understand the test results, and even specialists can reduce the effort required to understand the test results. Furthermore, when an index is used, the lesion can be quantitatively evaluated, so that the state transition of cancer or inflammation can be understood in the same way as understanding the results of a blood test. [Explanation of symbols]

[0064] 1: Information processing equipment 2: Positron emission tomography 3: Image storage device 4: Learned data providing device 5: Learned data providing device 6: Learned data providing device 11: Processing section 12: Storage section 13:Temporary storage 14: External device connection 15: Communications Department 16: Communication bus 100: Information processing device 101: Image acquisition unit 102: Lesion identification section 103: Data storage unit 104: Area identification part 105: Area calculation part 106:Indicator calculation section 107: Output section 110: Information processing device 111: Image acquisition unit 112: Lesion identification section 113: Data acquisition section 114: Area identification part 115: Area calculation part 116: Indicator calculation section 117: Output section 120: Information processing device 121: Image acquisition unit 122: Lesion identification section 123: Data acquisition section 124: Area identification part 125: Volume calculation unit 126: Indicator calculation section 127: Output section BL: Bladder BR: Brain HE: Heart KI: Kidney LI: Liver X: Non-physiological accumulation area

Claims

1. An information processing device that identifies a lesion based on a positron emission tomography examination, An image acquisition unit and a lesion identification unit are provided, the image acquisition unit is configured to be able to acquire an image including an anatomical region captured by a positron emission tomography apparatus; The lesion identification unit is configured to be able to identify a lesion area, excluding a portion where the positron-emitting nuclide has physiologically accumulated, from a portion where the positron-emitting nuclide has accumulated in the image, based on learned data obtained by machine learning. Information processing device.

2. 2. The information processing device according to claim 1, The apparatus includes a region specifying unit, a volume calculating unit, and an index calculating unit, the region specifying unit is configured to be able to specify an anatomical region in the image; the image is a three-dimensional image, the volume calculation unit is configured to be able to calculate a volume of the lesion and a volume of the anatomical region; the index calculation unit is configured to be able to calculate an index, The index is the ratio of the volume of the lesion to the volume of the anatomical region. Information processing device.

3. 3. The information processing device according to claim 2, The volume of the anatomical region is the integrated volume of the tomographic images of the subject in the image. Information processing device.

4. 4. The information processing device according to claim 2, The volume calculation unit calculates the volume of the lesion based on the contour of the lesion, and calculates the volume of the anatomical region based on the contour of the anatomical region. Information processing device.

5. 4. The information processing device according to claim 2, The volume calculation unit calculates the volume of the lesion based on the number of pixels of the tomographic image of the lesion, and calculates the volume of the anatomical region based on the number of pixels of the anatomical region. Information processing device.

6. 2. The information processing device according to claim 1, The apparatus includes a region specifying unit, an area calculating unit, and an index calculating unit, the region specifying unit is configured to be able to specify an anatomical region in the image; the area calculation unit is configured to be able to calculate an area of ​​the lesion and an area of ​​the anatomical region; the area of ​​the anatomical region is the projected area of ​​the subject in the image; the index calculation unit is configured to be able to calculate an index, The index is the ratio of the area of ​​the lesion to the area of ​​the anatomical region. Information processing device.

7. 7. The information processing device according to claim 6, The area of ​​the anatomical region is the frontal projection area of ​​the subject in the image. Information processing device.

8. 8. The information processing device according to claim 6, The area calculation unit calculates the area of ​​the lesion based on the contour of the lesion, and calculates the area of ​​the anatomical region based on the contour of the anatomical region. Information processing device.

9. 8. The information processing device according to claim 6, The area calculation unit calculates the area of ​​the lesion based on the number of pixels of the lesion, and calculates the area of ​​the anatomical region based on the number of pixels of the anatomical region. Information processing device.

10. 10. The information processing device according to claim 1, The anatomical region includes the entire body of the subject. Information processing device.

11. 11. The information processing device according to claim 1, A data storage unit is provided, The data storage unit stores the learned data. Information processing device.

12. 11. The information processing device according to claim 1, A data acquisition unit is provided, The data acquisition unit is configured to be able to acquire the learned data from a learned data providing device via a communication network. Information processing device.

13. 13. The information processing device according to claim 1, The trained data is generated by machine learning using, as training data, an image including an anatomical region captured by a positron emission tomography device and instruction information indicating a lesion in the image. Information processing device.

14. 14. The information processing device according to claim 13, The instruction information is information indicating the lesion area indicated by a specialist. Information processing device.

15. A program that causes a computer to operate as an information processing device, A computer is caused to function as the information processing device according to any one of claims 1 to 14. program.

Images