Information processing apparatus, information processing method, and program

Abstract

To select an inference unit suitable for medical image data of an inference object among a plurality of inference units while reducing time and effort of a user.SOLUTION: According to one aspect of the present invention, an information processing device includes an inference unit information acquisition part for acquiring information on respective inference units constituting a plurality of inference units, an incidental information acquisition part for acquiring incidental information attached to medical image data of an inference object obtained by imaging a subject, and an inference unit selection part for selecting an inference unit to be applied to the medical image data of the inference object among the plurality of inference units on the basis of the information on the inference units and the incidental information.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program that select an appropriate inference device from a plurality of inference devices based on the attached information attached to the medical image data to be inferred.

Background Art

[0002] With the development of artificial intelligence technologies such as machine learning, a plurality of inference devices for making inferences about diseases, for example, are provided on platforms such as medical cloud services and workstations for medical image data processing. A user can input medical image data to be inferred to a desired inference device and receive an inference result. The medical image data used for learning the inference device often differs for each inference device depending on the disease to be inferred (for example, pneumothorax, lung nodule, cerebral infarction) and the type of medical image data (type of imaging device, imaging site).

[0003] In many cases, a user selects an inference device to be applied to medical image data to be inferred, and inference processing is executed by inputting the medical image data to the selected inference device. Of course, it is also possible to apply a plurality of inference devices without selecting an inference device. However, since the plurality of inference devices are inference devices learned under different conditions, it is necessary to check the medical image data to be input according to the inference device, or it is necessary for the user to check an appropriate result from a plurality of inference results obtained by the plurality of inference devices, which is inefficient. Therefore, it is required that an appropriate inference device for medical image data to be inferred can be selected while reducing the user's effort.

[0004] Here, a technique for determining whether medical image data input to an image data processing apparatus is a processing target (suitable for processing) is known. For example, Patent Document 1 discloses a technique for acquiring attached information of medical image data and determining whether the medical image data is a processing target (suitable for processing) of time difference processing.

Prior Art Documents

Patent Documents

[0005] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2019-0129989 [Summary of the Invention] [Problems to be Solved by the Invention]

[0006] However, in the technology disclosed in Patent Document 1, it is not possible to select an inference device suitable for the medical image data to be inferred from among a plurality of inference devices.

[0007] Therefore, an object of the present invention is to select, from among a plurality of inference devices, an inference device suitable for the medical image data to be inferred while reducing the user's effort. [Means for Solving the Problems]

[0008] The information processing apparatus according to the present invention includes an inference device information acquisition unit that acquires information about each inference device constituting a plurality of inference devices, an attached information acquisition unit that acquires attached information attached to medical image data to be inferred obtained by imaging a subject, and an inference device selection unit that selects, based on the information about the inference device and the attached information, an inference device to be applied to the medical image data to be inferred from among the plurality of inference devices. [Effects of the Invention]

[0009] According to the present invention, it is possible to select, from among a plurality of inference devices, an inference device suitable for the medical image data to be inferred while reducing the user's effort. [Brief Description of the Drawings]

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Best Mode for Carrying Out the Invention

[0011] Hereinafter, preferred embodiments for carrying out the present invention will be described with reference to the drawings. Note that the present invention is not limited to the illustrated examples.

[0012] <Embodiment 1> In the present embodiment, items constituting the attached information attached to the medical image data to be inferred are compared with items constituting information regarding each inference device constituting a plurality of inference devices to determine whether each item matches, and in response to the determination result, an inference device to be applied to the medical image data to be inferred is selected from among the plurality of inference devices. This will be described in detail below with reference to the drawings.

[0013] First, the configuration of the information processing apparatus in the present invention will be described with reference to FIG. 1.

[0014] As shown in FIG. 1, the information processing apparatus 100 includes an attached information acquisition unit 101 that acquires attached information attached to medical image data to be inferred, an inference device information acquisition unit 102 that acquires information about each inference device that constitutes a plurality of inference devices for performing inferences related to a predetermined disease, and an inference device selection unit 103 that selects, based on the attached information and the information about the inference devices, an inference device to be applied to the medical image data to be inferred from among the plurality of inference devices. The inference device selection unit 103 further includes an inference device determination unit 106 that determines whether the attached information matches the information about the inference device. The information processing apparatus 100 is also connected to a database 105 via a communication network 104. Hereinafter, each unit will be described in detail.

[0015] The attached information acquisition unit 101 acquires various information (hereinafter referred to as attached information) attached to the medical image data to be inferred that has imaged a subject. Here, the attached information is information such as, for example, the type of imaging device, the imaging site, and the imaging conditions. Hereinafter, various attached information will be described.

[0016] Among the attached information acquired by the attached information acquisition unit 101, the item of the imaging device type is information indicating the type of the medical image data imaging device that has generated the medical image data to be inferred. Examples of the types of medical image data imaging devices include the following. · CR (Computed Radiography) device, · CT (Computed Tomography) device, · MRI (Magnetic Resonance Imaging) device, · PET (Positron Emission Tomography) device, · SPECT image data (Single Photon Emission Computed Tomography), · Ultrasonic device (US; Ultrasound System)

[0017] Among the additional information acquired by the additional information acquisition unit 101, the item of the imaging site is information indicating the site where the subject was imaged by the imaging device. Specifically, it is information indicating specific sites such as the head, neck, chest, abdomen, pelvis, and limbs, or information indicating ranges such as the whole body and the chest and abdomen.

[0018] Furthermore, among the additional information acquired by the additional information acquisition unit 101, the imaging conditions are information including a plurality of items related to the conditions at the time of imaging. For example, it is information such as the presence or absence of contrast agent use and the slice thickness. Note that the imaging conditions may include information related to the conditions specific to the imaging device. For example, if the imaging device that captured the medical image data is a CT device, it may include information indicating the window level and window width at the time of displaying the medical image data, such as the longitudinal profile condition and the lung field condition. If the imaging device is an MRI device, it may include information indicating the imaging sequences such as T1, T2, and FLAIR.

[0019] The inference device information acquisition unit 102 acquires information related to the inference device from the database 105 connected via the communication network 104. Here, the information related to the inference device is information related to the teacher data when the inference device is being learned, and is information related to the disease name associated with the correct label constituting the teacher data and / or information related to the correct image data used for learning the inference device. In the database 105, the model information and the above-mentioned information related to each inference device of a plurality of inference devices are stored in association with each other, and these information can be acquired at a desired timing. Here, the model information of the inference device refers to the information necessary for the inference device to execute inference. For example, in the case of an inference device using a neural network as the inference method, it is the information of the network structure and weights. Note that in FIG. 1, the database 105 is connected via a communication network, but it is not limited to this, and any medium through which the information processing device 100 can acquire information related to the inference device may be used. For example, it may be a storage area of a hard disk physically linked to the information processing device 100.

[0020] The inference device selection unit 103 selects one inference device from among the plurality of inference devices acquired by the inference device information acquisition unit 102 in response to the determination result by the inference device determination unit 106.

[0021] The inference device determination unit 106 determines, based on the attached information acquired by the attached information acquisition unit 101 of the medical image data and the information regarding the inference device acquired by the inference device selection unit 103, whether each of the inference devices is suitable for performing inference on the medical image data to be inferred, and transmits the determination result to the inference device selection unit 103.

[0022] Figure 2 is a diagram showing an example of the hardware configuration in the information processing apparatus 100. The CPU (Central Processing Unit) 201 mainly controls the operations of each component. The main memory 202 stores the control program executed by the CPU 201 and provides a working area during program execution by the CPU 201. The magnetic disk 203 stores various application software including programs for realizing an OS (Operating System), device drivers for peripheral devices, and programs for performing processes described later. By the CPU 201 executing the programs stored in the main memory 202, the magnetic disk 203, etc., the functions (software) of the information processing apparatus 100 shown in FIG. 1 and the processes in the flowchart described later are realized.

[0023] The display memory 204 temporarily stores display data. The monitor 205 is, for example, a CRT monitor or a liquid crystal monitor, etc., and performs display of image data, text, etc. based on the data from the display memory 204. The mouse 206 and the keyboard 207 respectively perform pointing input and input of characters, etc. by the user. The above components are connected to be communicable with each other via a common bus 208.

[0024] The CPU 201 corresponds to an example of a processor. The image data processing apparatus 100 may have at least one of a GPU (Graphics Processing Unit) and an FPGA (Field-Programmable Gate Array) in addition to the CPU 201. Further, instead of the CPU 201, it may have at least one of a GPU and an FPGA. The main memory 202 and the magnetic disk 203 correspond to an example of a memory.

[0025] Next, with reference to FIG. 3, the processing flow of the information processing apparatus 100 in the present embodiment will be described.

[0026] In step S301, the attached information acquisition unit 101 acquires the attached information in the medical image data to be inferred. The attached information corresponding to the medical image data to be inferred by the attached information acquisition 101 is acquired using, for example, a DICOM header. It is stipulated that in the DICOM header, information corresponding to each item is described with a pair of four-digit numbers as a tag. Therefore, the attached information acquisition unit 101 can acquire the above-described attached information by referring to the information of a predetermined tag. Specifically, the type of imaging device is recorded in the (0008, 0060) tag, and information on the examination site is recorded in (0018, 0015). Further, the contrast agent flow rate is described in the (0018, 1046) tag, and information on the contrast agent flow period is described in (0018, 1047), and it is possible to determine whether or not a contrast agent has been administered.

[0027] Using FIG. 4(A), as an example of the attached information attached to the medical image data to be inferred acquired by the attached information acquisition unit 101, an example of the attached information corresponding to the medical image data obtained by photographing the chest of a subject without administering a contrast agent using a CT device is shown. Note that the attached information acquired by the attached information acquisition unit 101 is not limited to being acquired from DICOM header information, and it may be acquired from a text file associated with the medical image data.

[0028] In step S302, the inference device information acquisition unit 102 acquires information about the inference device from the database 105. An example of the information about the inference device acquired by the inference device information acquisition unit 102 is shown in FIG. 4(B). Each piece of information about the plurality of inference devices in FIG. 4(B) indicates information about the teacher data when learning each of the plurality of inference devices. The target disease name of each piece of information is, for example, information related to the correct label among the information constituting the teacher data, and the type of imaging device, the imaging site, the imaging conditions, the presence or absence of a contrast agent, the slice thickness, etc. are information related to the correct image data. When learning the inference device, for example, teacher data in which information related to the correct label and the correct image data are paired is used.

[0029] In step S303, the inference device selection unit 103 selects one from the information about the inference device in FIG. 4(B) acquired in S302 and transmits it to the inference device determination unit 106.

[0030] In steps S304 to S306, the inference device determination unit 106 determines whether the inference device indicated by the information about the inference device is suitable for inferring the medical image data to be inferred based on the additional information in the medical image data to be inferred acquired in S301 and the inference device information transmitted by the inference device selection unit 103 in S303.

[0031] First, in step S304, the inference device determination unit 106 determines whether the imaging devices match among the items constituting the information about the inference device and the additional information.

[0032] Next, in step S305, the inference device determination unit 106 determines whether the imaging sites match among the items constituting the information about the inference device and the additional information.

[0033] Next, in step S306, the inference device determination unit 106 determines whether the presence or absence of contrast agent use matches among the items constituting the information about the inference device and the additional information.

[0034] As described above, the inference device determination unit 106 sequentially determines whether the information input to each item of the attached information of the medical image data to be inferred and the information regarding the inference device is the same. If all match, it proceeds to step S307. If any item does not match, it proceeds to step S308. Note that, as the item to be determined by the inference device determination unit 106, any one of the imaging device, imaging site, presence or absence of a contrast agent, etc. may be used for determination, or the user may select the item to be determined, and determination may be made for the selected item.

[0035] In step S307, the inference device determination unit 106 stores the information regarding the inference device transmitted by the inference device selection unit 103 in S303 as an output candidate to the inference device selection unit 103.

[0036] In step S308, the inference device selection unit 103 determines whether there is information regarding an inference device for which the determination from steps S304 to S306 has not been performed. If it exists, it returns to step S303. If it does not exist, it proceeds to step S309.

[0037] In step S309, the inference device selection unit 103 acquires the information regarding the inference device determined by the inference device determination unit 106 that the items match in step S307, and selects an inference device for inferring the medical image data to be inferred from a plurality of inference devices.

[0038] Fig. 5 shows an example of information of an inference device in which the attached information selected by the inference device selection unit 103 completely matches based on the attached information of the medical image data to be inferred shown in Fig. 4(A) and the attached information regarding the inference device shown in Fig. 4(B).

[0039] As described above, according to the present embodiment, when a plurality of inference devices can be selected, an inference device suitable for the medical image data to be inferred can be selected while reducing the burden on the user.

[0040] (Modification Example 1 of Embodiment 1) In this embodiment, the inference device determination unit 106 determines whether the items of the imaging device, the imaging site, and the presence or absence of contrast agent use among the attached information of the medical image data match. However, the information to be compared for the determination is not limited to these. For example, other attached information such as slice thickness, window level, and window width may be added as items for the inference device determination unit 106 to determine, or it may be possible to set in advance which information among the items constituting the acquired attached information is to be compared.

[0041] (Modification Example 2 of Embodiment 1) In this embodiment, the inference device determination unit 106 determines whether each item constituting the attached information matches from step S304 to S306. However, in the determination of whether they match, it is not limited to the method in which the information between the items completely matches.

[0042] For example, when the information input for an item such as the imaging site is string information, the inference device determination unit 106 may prepare in advance a map that considers a specific string to match, and determine whether they match by referring to the map.

[0043] Specifically, FIG. 6 shows an example of a map to be referred to when the item to be compared by the inference device determination unit 106 is the information of the imaging site. The map in FIG. 6 shows the correspondence of information for determining that they match by marking with circles when comparing the information in the item of the imaging site, which is one of the attached information of the medical image data to be inferred, with the information in the item of the imaging site, which is one of the information related to the inference device.

[0044] For example, when the inference device determination unit 106 refers to the map in FIG. 6, even if the information in the item of the imaging site regarding the medical image data to be inferred is the chest and abdomen, and the information in the item of the imaging site, which is one of the items constituting the information regarding the inference device, is the chest or abdomen, it is determined that the attached information matches. Thus, when it is desired to infer medical image data obtained by imaging the chest and abdomen, the inference device determination unit 106 can determine that an inference device targeting the chest or abdomen matches. In response to the determination result by the inference device determination unit 106, the inference device selection unit 103 can select an appropriate inference device from a plurality of inference devices while reducing the user's effort. The method of using a map showing the relationship between character strings is particularly effective when comparing limited words such as the information of the chest and abdomen in the imaging site.

[0045] For example, when the item for which the inference device determination unit 106 makes a determination is, for example, the slice thickness or the window level, and the information input for the item is numerical information, a range of numerical values to be determined as matching by the inference device determination unit 106 may be determined in advance. For example, when the information input for the item of the slice thickness, which is one of the attached information of the medical image data to be inferred, is 3 mm by the inference device determination unit 106, if the information input for the item of the slice thickness, which is one of the information regarding the inference device, is in the range of 1 mm before and after from 2 mm to 4 mm, it is determined that the information input between the items of the slice thickness matches.

[0046] Also, the method for the inference device determination unit 106 to determine whether the attached information matches may differ for each item or may be set according to the form of the information input for the item. Furthermore, a map, threshold values, numerical value ranges, and other conditions set in advance to determine whether the information between items matches may be set for each inference device.

[0047] <Embodiment 2> In Embodiment 1, the inference device determination unit 106 in the inference device selection unit 103 compares the attached information of the medical image data to be inferred acquired by the attached information acquisition unit 101 with the information on the inference device acquired by the inference device information acquisition unit 102 for each item of the input information. When each item matches, the inference device selection unit 103 selects the matching inference device from a plurality of inference devices and selects it as the inference device for inferring the medical image data. The method of selection was described.

[0048] In this embodiment, the inference device determination unit 106 calculates the degree of match between the items constituting the attached information of the medical image data to be inferred and the items constituting the information on the inference device. Based on the degree of match, the inference device selection unit 103 selects the inference device for inferring the medical image data to be inferred. The inference device determination unit 106 applies a threshold value to the calculated degree of match to determine whether the inference device is appropriate for the medical image to be inferred.

[0049] Using FIG. 7, the processing flow from determining whether the inference device is suitable in the information processing apparatus 100 to receiving the determination result and selecting the inference device to be applied to the medical image to be inferred from a plurality of inference devices will be described.

[0050] Steps S701, S702, and S703 are the same as steps S301, S302, and S303 in the flowchart of FIG. 3 in Embodiment 1, so the description will be omitted.

[0051] In step S704, the inference device determination unit 106 calculates the degree of match rij (j = 1,..., n) for each item constituting the additional information and information regarding the inference device, such as the imaging device and the imaging site. Here, j is the index number assigned to each item, and n is the number of items. The degree of match rij of each item calculated by the inference device determination unit 106 is, for example, rij = 1 when the information between items completely matches, and rij = 0 when they do not match. Here, when the information input to item j is string information, the inference device determination unit 106 may use the similarity described in Embodiment 1 as the degree of match. Also, when the information input to item j is numerical information, the inference device determination unit 106 may calculate the absolute value of the difference between the numerical values between the items and use it as the degree of match. When calculating the degree of match from the absolute value of the difference between the numerical information between the items, the degree of match rij between the items calculated by the inference device determination unit 106 is calculated by Equation 1, where xj is the numerical value of item j of the additional information of the medical image data to be inferred, and yij is the numerical value of item j of the information regarding the inference device i.

[0052]

Number

[0053] As an example, regarding the item of slice thickness, a method for calculating the degree of match of the item of slice thickness by the inference device determination unit 106 when the numerical information input to the item of slice thickness in the additional information of the medical image data to be inferred is 1.5 mm is shown.

[0054] For example, when a plurality of inference devices are composed of three inference devices (i = 1, 2, 3), and the information input to the item of slice thickness constituting the information regarding each inference device is 1.0 mm, 1.5 mm, and 3.0 mm, respectively, the degrees of match in this item calculated by the inference device determination unit 106 are r1j = 0.75, r2j = 0.00, and r3j = 0.25, respectively.

[0055] In step S705, the inference device determination unit 106 calculates the degree of match Ri (i = 1, ..., N) between the attached information of the medical image to be inferred and the information regarding the inference device. Here, i is the index number assigned to the inference device, and N is the number of input inference devices. Ri is calculated by taking the sum of the degrees of match rij (j = 1, ..., n) for each item j by the inference device determination unit 106 calculated in step S704, as shown in Equation 2.

[0056]

Number

[0057] In step S706, the inference device determination unit 106 sets a threshold for the calculated degree of match as shown in FIG. 8(D), and determines whether the calculated degree of match is greater than or equal to the set threshold, thereby determining whether the inference device is suitable for the input medical image data. Then, in response to the result determined by the inference device determination unit 106, the inference device selection unit 103 selects an inference device for inferring the medical image to be inferred from among the plurality of inference devices.

[0058] Note that the threshold used by the inference device determination unit 106 can be arbitrarily set by the user. For example, when the threshold is set to 0.7 in the calculation result of FIG. 8(D), the inference device determination unit 106 determines that inference devices 2 and 4 are suitable for the medical image data to be inferred.

[0059] Steps S707, S708, and S709 are the same as steps S307, S308, and S309 in Embodiment 1, and thus the description thereof is omitted.

[0060] As described above, according to the present embodiment, the inference device determination unit 106 calculates the degree of coincidence between the attached information of the medical image to be inferred and the information regarding the inference device, and applies a threshold value to the calculated degree of coincidence to determine whether the inference device indicated by the information regarding the inference device is suitable for the medical image data to be inferred. By doing so, among the plurality of inference devices for inferring diseases by the inference device selection unit 103, it becomes possible to flexibly select an inference device.

[0061] <Embodiment 3> In Embodiment 2, a method was described in which the inference device determination unit 106 calculates the degree of coincidence of the attached information and applies a threshold value to the calculated degree of coincidence to determine whether the inference device indicated by the information regarding the inference device is suitable for the medical image data to be inferred.

[0062] In the present embodiment, the inference device determination unit 106 calculates the degree of coincidence in consideration of the priority of each item of the attached information, and a method will be described in which the inference device selection unit 103 selects an inference device for inferring a medical image to be inferred from a plurality of inference devices based on the calculated degree of coincidence. In the present embodiment, the inference device selection unit 103 sets weights for the items and selects an inference device based on the degree of coincidence of each item and the weights for the items.

[0063] The processing flow in the present embodiment is the same as that in Embodiment 2. However, the method for calculating the degree of coincidence Ri between the attached information of the medical image to be inferred and the information regarding the inference device in step S705 is different.

[0064] In the present embodiment, in calculating the degree of coincidence Ri between the information by the inference device determination unit 106, weights are set for each item j constituting both pieces of information, and a weighted sum is calculated. The degree of coincidence Ri between the information calculated by the inference device determination unit 106 is calculated by Expression (3) with the weight for item j being wj (j = 1,..., n). Note that the weights wj are set such that their sum is 1.

[0065]

Number

[0066] Similar to step S706 of Embodiment 2, a threshold is set for the degree of coincidence of the information calculated as shown in FIG. 9(E) by the inference device determination unit 106, and it is determined whether the calculated degree of coincidence is equal to or greater than the threshold, thereby determining whether the inference device indicated by the information regarding the inference device is suitable for inferring the medical image data to be inferred. In the calculation result of FIG. 9(E), when the threshold applied by the inference device determination unit 106 is 0.7, it is determined that the inference devices 2, 4, and 5 are suitable for the inference device that performs inference on the medical image data to be inferred.

[0067] As described above, according to the present embodiment, by calculating the degree of coincidence in consideration of the priority of each item by the inference device determination unit 106, the inference device selection unit 103 can select an inference device suitable for the medical image data of the inference injury from among a plurality of inference devices that infer diseases according to the importance of each item.

[0068] <Embodiment 4> In Embodiments 1, 2, and 3, a method of selecting an appropriate inference device using the attached information of the medical image data to be inferred and the information regarding the inference device has been described.

[0069] In this embodiment, the additional information acquisition unit 1001 further acquires information about the subject, and transmits the acquired information about the subject to the disease prediction unit. Then, the disease prediction unit predicts disease candidates based on the acquired information about the subject, and the inference device selection unit 1003 further selects an appropriate inference device from a plurality of inference devices based on the prediction result by the disease prediction unit.

[0070] First, the configuration of the present invention in this embodiment will be described with reference to FIG. 10. The information processing apparatus 1000 has a disease prediction unit 1005 that predicts disease candidates from information about the subject in addition to the configuration described in Embodiment 1. Further, the information processing apparatus 1000 is connected to a database 1007 and a medical information system 1008 via a communication network 1006. Note that systems used in hospitals, such as a hospital information system (HIS) and an electronic medical record system, are collectively referred to as the medical information system 1008.

[0071] Since the inference device information acquisition unit 1002 is the same as that in Embodiment 1, the description thereof will be omitted.

[0072] The additional information acquisition unit 1001 for the medical image data to be inferred further acquires subject information from the medical information system 1008 connected via the communication network 1006. Here, the subject information acquired by the additional information acquisition unit 1001 refers to, for example, interview information or information described in an electronic medical record. Note that, as systems for acquiring subject information in the additional information acquisition unit 1001, a hospital information system and an electronic medical record system are exemplified, but the information acquisition destination is not limited to those systems as long as subject information can be acquired.

[0073] The disease prediction unit 1005 predicts disease candidates based on the subject information further acquired by the additional information acquisition unit 1001.

[0074] Based on the supplementary information acquired by the supplementary information acquisition unit 1001 of the medical image data, the information regarding the inference device acquired by the inference device selection information acquisition unit 1002, and the disease candidates that are the prediction results of the disease prediction unit 1005, the inference device determination unit 1004 determines whether the inference device indicated by the information regarding the inference device is suitable as an inference device for performing inference on the medical image data to be inferred.

[0075] The above is the configuration of the information processing apparatus 1000 of the present invention in Embodiment 4.

[0076] Next, with reference to FIG. 11, the processing flow of the information processing apparatus 1000 in the present embodiment will be described.

[0077] In step S1101, the supplementary information acquisition unit 1001 acquires the supplementary information attached to the medical image to be inferred.

[0078] Until the inference device information acquisition unit 1002 acquires information regarding the inference device in step S1102, it is the same as in Embodiments 1, 2, and 3.

[0079] In step S1103, the supplementary information acquisition unit 1001 further acquires subject information from the medical information system 1008. As the subject information acquired by the supplementary information acquisition unit 1001, the subject's interview information or electronic medical record or both are acquired from the medical information system 1008. Specifically, items such as the subject name and subject ID are described in the DICOM header, and by querying the databases of each system such as an electronic medical record system that constitutes the medical information system 1008 using these, the information of the corresponding subject can be acquired.

[0080] In step S1104, the disease prediction unit 1105 predicts disease candidates based on the interview information, the electronic medical record information, or both, which are the subject information obtained by the attached information acquisition unit 1001 in step S1103. Here, in step S1104, as long as possible disease names are listed as disease candidates in the prediction result by the disease prediction unit 1105, the means of listing them may be any means described above. For example, when the subject information obtained by the attached information acquisition unit 1001 is interview information, disease candidates may be determined by comparing with rules registered in advance for the answer content of the interview, or may be estimated by known techniques such as machine learning based on the answer content of the interview. When listing disease candidates based on rules, for example, set it such that when chest pain is included in the symptoms, pneumothorax and aortic dissection are added as disease candidates. Also, in the case of an electronic medical record, information on the past medical history is extracted and added to the disease candidates. Further, diseases with a possibility of other complications may be predicted from the past medical history by a rule-based like a dictionary or machine learning and added as disease candidates. As described above, as shown in FIG. 12(A), information on the disease candidates predicted in step S1104 is added to the attached information of the medical image data to be inferred for the target disease name.

[0081] In step S1105, the inference device selection unit 1003 selects one from the information on the inference device obtained in S1102.

[0082] In step S1106, the inference device determination unit 1004 determines whether the target disease name of the inference device information selected in S1105 is included in the disease candidates predicted in S1104. If it is determined by the inference device determination unit 1004 that it is included, the process proceeds to step S1107. If it is determined by the inference device determination unit 1004 that it is not included, the process proceeds to step S1111.

[0083] Steps S1107 to S1112 are the same as steps S304 to S309 in Embodiment 1, so the description thereof will be omitted. Note that the processing flow from step S1107 to step S1112 may be replaced by steps S705 to S709 of Embodiment 2.

[0084] As described above, according to the present embodiment, the attached information acquisition unit 1001 further acquires information about the subject, and the disease prediction unit 1005 predicts disease candidates based on the acquired information about the subject. Thus, the inference device determination unit 1004 can determine whether the inference device for inferring the medical image to be inferred is appropriate. Further, in response to the determination result by the inference device determination unit 1004, the inference device selection unit 1003 can select an appropriate inference device for inferring the medical image to be inferred from among a plurality of inference devices.

[0085] (Other embodiments) Further, the present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or device via a network or various storage media, and a computer (or CPU, MPU, etc.) of the system or device reads and executes the program.

Claims

1. An inference device information acquisition unit that acquires information on each inference device that constitutes a plurality of inference devices, including a disease name related to the correct label used for learning the first inference device and a disease name related to the correct label used for learning the second inference device; An incidental information acquisition unit that acquires information on a subject for which medical image data to be inferred was taken; A disease prediction unit that predicts a disease candidate of the subject from the information on the subject; An inference device selection unit that selects an inference device to be applied to the medical image data to be inferred among the plurality of inference devices based on the disease name related to the correct label and the disease candidate of the subject predicted by the disease prediction unit; An information processing apparatus, characterized by comprising the above.

2. The information on the subject is information including at least one of interview information and electronic medical record information, The disease prediction unit predicts the disease candidate from at least one of the interview information and the electronic medical record information. The information processing apparatus according to claim 1.

3. The inference device selection unit selects, among the plurality of inference devices, an inference device in which the disease name related to the correct label is included in the disease candidate predicted by the disease prediction unit. The information processing apparatus according to claim 1 or 2.

4. The information on the inference device includes information related to the correct image data used for learning the plurality of inference devices, The incidental information acquisition unit further acquires incidental information attached to the medical image data to be inferred, When the disease name related to the correct label among the plurality of inference devices is included in the disease candidate predicted by the disease prediction unit, the inference device selection unit calculates the degree of coincidence between the content of the items constituting the information related to the correct image data and the content of the items constituting the incidental information, and selects an inference device to be applied to the medical image data to be inferred based on the calculated degree of coincidence. The information processing apparatus according to claim 3.

5. The information on the incidental information and the correct image data is information including at least one item of the type of imaging device, imaging site, and imaging conditions. The information processing apparatus according to claim 4.

6. An inference device information acquisition step of acquiring information on each inference device constituting a plurality of inference devices including a disease name related to the correct label used for learning of the first inference device and a disease name related to the correct label used for learning of the second inference device; An incidental information acquisition step of acquiring information on a subject for which medical image data to be inferred has been taken; A disease prediction step of predicting a disease candidate of the subject from the information on the subject; An inference device selection step of selecting an inference device to be applied to the medical image data to be inferred among the plurality of inference devices based on the disease name related to the correct label and the disease candidate of the subject predicted in the disease prediction step; An information processing method, characterized by comprising:

7. A program for causing a computer to execute the information processing method according to Claim 6.

Images