Medical information processing apparatus and program

Abstract

To improve the accuracy of determining a lesion.SOLUTION: A medical information processing apparatus according to an embodiment includes an image acquisition unit, a biomarker value acquisition unit, a region of interest determination unit, an image analysis unit, a biomarker analysis unit, and a determination unit. The image acquisition unit acquires a medical image obtained by picking up an image of a subject. The biomarker value acquisition unit acquires a value of a biomarker contained in a sample collected from the subject. The region of interest determination unit determines a region of interest on the acquired medical image. The image analysis unit analyzes the image of the determined region of interest and calculates the property of a drawn object. The biomarker analysis unit selects a value of a biomarker associated with the calculated property of the drawn object. The determination unit determines the consistency between the selected value of the biomarker and the calculated property of the drawn object.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The embodiments disclosed in this specification and the drawings relate to a medical information processing device and a program. [Background technology]

[0002] Liquid biopsy, which uses body fluids such as blood to diagnose lesions, has been known as a test that places less of a burden on subjects such as patients. It is also possible to make a diagnosis by taking into account image information and test information from the liquid biopsy. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-181289 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, according to the research of the inventors of the present invention, image information is morphological information, whereas liquid biopsy test information is characteristic information, so there are cases where the two do not correspond, making it difficult to improve the accuracy of lesion determination. One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to improve the accuracy of lesion determination. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described later can also be positioned as other problems. [Means for solving the problem]

[0005] A medical image processing device according to an embodiment includes an image acquisition unit, a biomarker value acquisition unit, a region of interest determination unit, an image analysis unit, a biomarker analysis unit, and a determination unit. The image acquisition unit acquires a medical image of a subject. The biomarker value acquisition unit acquires values ​​of biomarkers contained in a sample collected from the subject. The region of interest determination unit determines a region of interest on the medical image acquired by the image acquisition unit. The image analysis unit analyzes an image of the region of interest determined by the region of interest determination unit to calculate the properties of the depicted object. The biomarker analysis unit selects biomarker values ​​related to the properties of the depicted object calculated by the image analysis unit. The determination unit determines the consistency between the biomarker values ​​selected by the biomarker analysis unit and the properties of the depicted object calculated by the image analysis unit. [Brief description of the drawings]

[0006] [Figure 1] FIG. 1 is a block diagram showing an example of the arrangement of a medical information processing system according to a first embodiment. [Diagram 2] FIG. 2 is a block diagram showing an example of the arrangement of a medical image processing apparatus according to the first embodiment. [Diagram 3] FIG. 3 is a flowchart showing an example of the operation of the medical image processing apparatus according to the first embodiment. [Figure 4] FIG. 4 is a diagram showing an example of biomarker values ​​in an example of the operation of the medical information processing apparatus according to the first embodiment. [Diagram 5] FIG. 5 is a diagram showing the correspondence between cancer types and mammography image findings in an example of the operation of the medical information processing apparatus according to the first embodiment. [Figure 6] FIG. 6 is a diagram showing an example of a determination table used for determining consistency in an example of the operation of the medical image processing apparatus according to the first embodiment. [Figure 7] FIG. 7 is a flowchart showing a detailed operation example of the medical image processing apparatus according to the first embodiment. [Figure 8] FIG. 8 is a diagram showing a mammography image and a region of interest in an example of the operation of the medical image processing apparatus according to the first embodiment. [Figure 9]FIG. 9 is a flowchart showing the operation of the medical image processing apparatus according to the first modified example of the first embodiment. [Figure 10] FIG. 10 is a flowchart showing the operation of the medical image processing apparatus according to the second modified example of the first embodiment. [Figure 11] FIG. 11 is a block diagram showing the configuration of a medical image processing apparatus according to a third modified example of the first embodiment. [Figure 12] FIG. 12 is a flowchart showing the operation of the medical image processing apparatus according to the third modified example of the first embodiment. [Figure 13] FIG. 13 is a flowchart showing an example of the operation of the medical image processing apparatus according to the second embodiment. [Figure 14] FIG. 14 is a diagram showing a mammographic image before and a mammographic image after a biopsy in an example of the operation of the medical information processing apparatus according to the second embodiment. [Figure 15] FIG. 15 is a flowchart showing the operation of the medical image processing apparatus according to a modified example of the second embodiment. [Figure 16] FIG. 16 is a schematic diagram showing the operation of the medical image processing apparatus according to a modified example of the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] Hereinafter, an embodiment of a medical image processing device will be described with reference to the drawings. In the following description, components having substantially the same functions and configurations are denoted by the same reference numerals, and duplicated descriptions will be given only when necessary.

[0008] (First embodiment) Fig. 1 is a block diagram showing an example of the configuration of a medical information processing system 100 according to the first embodiment. Fig. 2 is a block diagram showing an example of the configuration of a medical information processing apparatus 1 according to the first embodiment.

[0009] As shown in FIG. 1, the medical information processing system 100 according to the first embodiment includes a medical image diagnostic device 1, an inspection device 2, and a medical information processing device 3. The medical image diagnostic device 1, the inspection device 2, and the medical information processing device 3 are communicably connected via a network 200. That is, the medical information processing device 3 is provided on the medical information processing system 100 connected via the network 200. Note that the medical image processing device 3 may be an offline device. In that case, the medical image processing device 3 may acquire information from the medical image diagnostic device 1 and the inspection device 2 via a portable recording medium.

[0010] The devices included in the medical information processing system 100 may be in a state in which they can communicate with each other directly or indirectly, for example, via an in-hospital LAN (Local Area Network) installed in a hospital. In addition, the medical information processing system 100 shown in FIG. 1 may be communicably connected to devices other than those shown in the figure (for example, a server that stores medical information, etc.).

[0011] For example, the medical information processing system 100 may include various systems such as an Electronic Health Records (EHR) system, a Picture Archiving and Communication System (PACS), a Hospital Information System (HIS), a Radiology Information System (RIS), a diagnostic report system, and a Laboratory Information System (LIS).

[0012] The medical image diagnostic device 1 captures an image of a subject and collects medical images. The medical image diagnostic device 1 then transmits the collected medical images to the medical information processing device 3. For example, the medical image diagnostic device 1 may be an X-ray diagnostic device such as a mammography device, an X-ray CT (Computed Tomography) device, an MRI (Magnetic Resonance Imaging) device, an ultrasound diagnostic device, a SPECT (Single Photon Emission Computed Tomography) device, and a PET (Positron Emission Computed Tomography) device. Although not limited to such a medical image, it is typically a morphological image that visualizes the shape characteristics inside the body of the subject, and in this sense, the medical image diagnostic device 1 is typically an X-ray diagnostic device, an X-ray CT device, an MRI device, an ultrasound diagnostic device, or the like.

[0013] The testing device 2 is operated by a clinical laboratory technician or the like, and performs specimen testing to analyze specimens (i.e., samples) such as blood and urine obtained from patients. The specimen testing includes, for example, pathological testing, blood and biochemical testing, liquid biopsy testing, general testing of urine and stool, immune serum testing, genetic testing, microbiological testing, blood transfusion and organ transplant-related testing, etc. The testing device 2 then transmits the test results of the specimen testing (measurement data, etc.) to the medical information processing device 3.

[0014] For example, the testing device 2 may be an automatic biochemical analyzer, an automatic immunoanalyzer, a flow cytometer, a genetic analyzer, a protein analyzer, an extracellular vesicle analyzer, a blood circulating tumor cell detection device, or the like. The flow cytometer is an instrument that performs flow cytometry. Flow cytometry is an analytical method in which a suspension of a measurement object is treated as a high-speed fluid, and the fluid is irradiated with laser light, mercury light, or the like to generate scattered light and fluorescence, thereby measuring the size, amount, and the like of the measurement object.

[0015] A gene analyzer is a device that analyzes gene sequences. For example, a gene analyzer is a device that amplifies nucleic acid molecules extracted from a biological sample using a PCR (Polymerase Chain Reaction) method or the like, and determines the presence or absence of a specific gene sequence or quantifies the amount of the sequence, or a device that analyzes sequence information of a nucleic acid molecule. A protein analyzer is, for example, an automatic immunoassay analyzer or a highly sensitive protein detector (e.g., Quanterix's Single Molecule Assay device SIMOA (registered trademark)).

[0016] The extracellular vesicle analysis device is, for example, a single extracellular vesicle analysis device (e.g., EXOVIEW (registered trademark) from NanoView Biosciences, Inc., etc.), and the blood circulating tumor cell detection device is, for example, a blood circulating tumor cell detection device using a nanomagnetic bead method (e.g., CELLSEARCH (registered trademark) system from Veridex, Inc., etc.), etc.

[0017] The medical information processing device 3 acquires various information from the medical image diagnostic device 1 and the inspection device 2, and performs various information processing using the acquired information. For example, the medical information processing device 3 is realized by computer equipment such as a server, a workstation, a personal computer, or a tablet terminal.

[0018] 2, the medical information processing device 3 includes an input interface 31, an output interface 32, a communication interface 33, a storage circuit 34, and a processing circuit 35. The input interface 31, the output interface 32, the communication interface 33, and the storage circuit 34 are connected to the processing circuit 35.

[0019] The input interface 31 accepts various instructions and input operations of information from an operator. Specifically, the input interface 31 converts the input operations accepted from the operator into electrical signals and outputs them to the processing circuit 35. For example, the input interface 31 is realized by a trackball, a switch button, a mouse, a keyboard, a touch pad that performs input operations by touching the operation surface, a touch screen in which a display screen and a touch pad are integrated, a non-contact input circuit using an optical sensor, and a voice input circuit. Note that the input interface 31 is not limited to only those that have physical operation parts such as a mouse and a keyboard. For example, an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the device and outputs the electrical signal to the processing circuit 35 is also included as an example of the input interface 31.

[0020] The output interface 32 outputs various information. For example, the output interface 32 includes a display. The display outputs medical images generated by the processing circuitry 35 and a GUI (Graphical User Interface) for receiving various operations from an operator. For example, as the display, a liquid crystal display (LCD), a cathode ray tube (CRT) display, an organic electro luminescence display (OELD), a plasma display, or any other arbitrary display can be used as appropriate. The output interface 32 may include a speaker.

[0021] The communication interface 33 controls communication between each device in the medical information processing system 100. Specifically, the communication interface 33 receives various information from each device and outputs the received information to the processing circuit 35. For example, the communication interface 33 is realized by a network card, a network adapter, a NIC (Network Interface Controller), or the like.

[0022] The memory circuitry 34 is a non-transient storage device that stores various information, and is, for example, a hard disk drive (HDD), an optical disk, a solid state drive (SSD), an integrated circuit storage device, etc. The memory circuitry 34 stores, for example, a control program that controls the medical information processing device 3 and various data used in executing the control program. The memory circuitry 34 may be a drive device that reads and writes various information between a portable storage medium such as a compact disc (CD), a digital versatile disc (DVD), and a flash memory, or a semiconductor memory element such as a random access memory (RAM), in addition to an HDD or an SSD.

[0023] The processing circuit 35 is a circuit that controls the operation of the entire medical information processing device 3 in response to an electrical signal of an input operation input from the input interface 31. For example, the processing circuit 35 includes an image acquisition function 351, a biomarker value acquisition function 352, a region of interest determination function 353, an image analysis function 354, a biomarker analysis function 355, a judgment function 356, a score calculation function 357, and a region of interest identification function 358. The image acquisition function 351 is an example of an image acquisition unit. The biomarker value acquisition function 352 is an example of a biomarker value acquisition unit. The region of interest determination function 353 is an example of a region of interest determination unit. The image analysis function 354 is an example of an image analysis unit. The biomarker analysis function 355 is an example of a biomarker analysis unit. The judgment function 356 is an example of a judgment unit. The score calculation function 357 is an example of a score calculation unit. The region of interest identification function 358 is an example of a region of interest identification unit.

[0024] Here, for example, each processing function executed by an image acquisition function 351, a biomarker value acquisition function 352, a region of interest determination function 353, an image analysis function 354, a biomarker analysis function 355, a judgment function 356, a score calculation function 357, and a region of interest identification function 358, which are components of the processing circuit 35 shown in Fig. 2, is recorded in the storage circuit 34 in the form of a program executable by a computer. The processing circuit 35 is, for example, a processor. The processor constituting the processing circuit 35 realizes a function corresponding to each program read out by reading out each program from the storage circuit 34 and executing it. In other words, the processing circuit 35 in a state in which each program has been read out has each function shown in the processing circuit 35 in Fig. 2.

[0025] 2 shows a case where each of the processing functions, image acquisition function 351, biomarker value acquisition function 352, region of interest determination function 353, image analysis function 354, biomarker analysis function 355, judgment function 356, score calculation function 357, and region of interest identification function 358, is realized by a single processing circuit 35, but the embodiment is not limited to this. For example, the processing circuit 35 may be configured by combining multiple independent processors, and each processor may execute each program to realize each processing function. Furthermore, each processing function of the processing circuit 35 may be realized by being appropriately distributed or integrated into a single or multiple processing circuits.

[0026] The image acquiring function 351 acquires a medical image obtained by capturing an image of a subject. For example, the image acquiring function 351 acquires a medical image acquired by the medical image diagnostic apparatus 1 capturing an image of a subject from the medical image diagnostic apparatus 1 via the network 200. The medical image is received from the medical image diagnostic apparatus 1 by the communication interface 33 and stored in the storage circuitry 34, and then read out from the storage circuitry 34 and acquired by the image acquiring function 351.

[0027] The biomarker value acquisition function 352 acquires the values ​​of biomarkers contained in a sample collected from a subject. For example, the biomarker value acquisition function 352 acquires the values ​​of biomarkers acquired by the testing device 2 testing the subject via the network 200 from the testing device 2. The sample collected from the subject is, for example, a body fluid of the subject, such as blood or urine. The biomarker value acquisition function 352 acquires the values ​​of biomarkers associated with a lesion, such as cancer, that may be present in the subject.

[0028] The region of interest determination function 353 determines a region of interest (ROI) on the medical image acquired by the image acquisition function 351. The region of interest is determined, for example, as a region on the medical image where the presence of a lesion is suspected. For example, the region of interest determination function 353 determines the region of interest based on machine learning. That is, the region of interest determination function 353 determines the region of interest based on a trained model that has learned a lesion on the medical image. Note that, when information indicating the region of interest set by the medical image diagnostic device 1 is acquired by the medical information processing device 3, the region of interest determination function 353 may determine the region of interest based on the information indicating the region of interest. The region of interest determination function 353 can determine multiple regions of interest on the medical image acquired by the image acquisition function 351. The region of interest determination function 353 may determine the region of interest when the value of the biomarker acquired by the biomarker value acquisition function 352 is a value indicating that a lesion may exist in the subject.

[0029] The image analysis function 354 analyzes the image of the region of interest determined by the region of interest determination function 353, and calculates (i.e., detects) the properties of the depicted object in the region of interest. The image analysis function 354 calculates the properties of the depicted object according to the type of lesion. For example, the image analysis function 354 may calculate, as a result of analyzing the image of the region of interest, that the edges of the depicted object (i.e., the mass) are smooth, as a property of the depicted object corresponding to triple-negative breast cancer. Also, for example, the image analysis function 354 may calculate, as a result of analyzing the image of the region of interest, that the depicted object is irregular or includes spicules, as a property of the depicted object corresponding to luminal A breast cancer. Also, for example, the image analysis function 354 may calculate, as a result of analyzing the image of the region of interest, that the depicted object includes pleomorphic, fine linear, or branched calcification, as one of the properties of the depicted object corresponding to HER2-positive breast cancer. The image analysis function 354 may calculate the properties of the depicted object using, for example, image recognition or a trained model that has learned the properties of the depicted object.

[0030] The biomarker analysis function 355 selects a biomarker value related to the property of the depicted object calculated by the image analysis function 354. For example, the biomarker analysis function 355 selects a biomarker value of a type related to the property of the depicted object calculated by the image analysis function 354 from among the multiple types of biomarker values ​​acquired by the biomarker value acquisition function 352. The biomarker value selected by the biomarker analysis function 355 is not limited to one type, and may be multiple types.

[0031] For example, when the property of the depicted object calculated by the image analysis function 354 is that the edge of the depicted object is smooth, the biomarker analysis function 355 selects the value of a biomarker associated with triple-negative breast cancer from the values ​​of the biomarkers acquired by the biomarker value acquisition function 352. The values ​​of the biomarkers associated with triple-negative breast cancer are, for example, the values ​​of epidermal growth factor receptor (EGFR), estrogen receptor (ER), progesterone receptor (PR), and HER2.

[0032] More specifically, when the value of epidermal growth factor receptor is higher than the normal range (positive case), the subject may have triple-negative breast cancer. Therefore, the biomarker analysis function 355 may select the value of epidermal growth factor receptor showing positive as the value of the biomarker related to triple-negative breast cancer. Also, when the values ​​of estrogen receptor, progesterone receptor, and HER2 are within the normal range (negative case), the subject may have triple-negative breast cancer. Therefore, the biomarker analysis function 355 may select the values ​​of estrogen receptor, progesterone receptor, and HER2 showing negative as the value of the biomarker related to triple-negative breast cancer.

[0033] Furthermore, for example, when the nature of the depicted object calculated by the image analysis function 354 is that the depicted object is irregular or that it is accompanied by spicules, the biomarker analysis function 355 selects a biomarker value associated with luminal A breast cancer from the biomarker values ​​acquired by the biomarker value acquisition function 352. The biomarker values ​​associated with luminal A breast cancer are, for example, values ​​of estrogen receptor, progesterone receptor, HER2, etc.

[0034] More specifically, when either or both of the estrogen receptor and the progesterone receptor are higher than the normal range (positive case) and the HER2 value is within the normal range (negative case), the subject may have luminal A breast cancer. Therefore, biomarker analysis function 355 may select the values ​​of either or both of the estrogen receptor and the progesterone receptor that are positive, and the value of HER2 that is negative, as the values ​​of the biomarkers related to luminal A breast cancer.

[0035] Also, for example, when the nature of the image calculated by the image analysis function 354 is that the image involves calcification, the biomarker analysis function 355 selects a HER2 value associated with HER2-positive breast cancer from the biomarker values ​​acquired by the biomarker value acquisition function 352.

[0036] More specifically, when the HER2 value is higher than the normal range (positive), the subject may have HER2-positive breast cancer. Therefore, the biomarker analysis function 355 may select a positive HER2 value as a biomarker value associated with HER2-positive breast cancer.

[0037] There are multiple characteristics of an object corresponding to one cancer subtype (see FIG. 6). For example, there are multiple characteristics of an object corresponding to triple-negative breast cancer, including smooth edges of the object (see FIG. 6). There are also multiple characteristics of an object corresponding to luminal A breast cancer, including irregular shapes of the object (see FIG. 6). There are also multiple characteristics of an object corresponding to HER2-positive breast cancer, depending on the form of calcification (see FIG. 6). If at least one of the multiple characteristics of an object for each subtype is calculated by the image analysis function 354, the biomarker analysis function 355 selects a biomarker value associated with the corresponding subtype. For example, if the biomarker analysis function 355 calculates that the edge of the object is smooth, it selects a biomarker value associated with triple-negative breast cancer, even if other characteristics of the object corresponding to triple-negative breast cancer (e.g., circular, etc.) are not calculated.

[0038] Among the biomarker values ​​associated with one subtype, there are also values ​​of the type of biomarker common to other subtypes. For example, as described above, the values ​​of estrogen receptor, progesterone receptor, and HER2 associated with triple-negative breast cancer are common to luminal A breast cancer and partially common to HER2-positive breast cancer. In addition, the positive or negative values ​​of biomarkers may be partially common between multiple subtypes. In addition, there are also properties of the depicted object (e.g., calcification) common to multiple subtypes. In addition, when selecting the value of the biomarker associated with the properties of the depicted object, the occurrence frequency of the properties of the depicted object is not taken into consideration as in the case of the determination of consistency described below, so it is not known how effective the selected biomarker value is in suspecting the existence of the corresponding subtype. Therefore, when the image analysis function 354 simultaneously calculates the properties of the depicted object corresponding to each of multiple subtypes, it is difficult to determine which subtype the selected biomarker value is strongly influenced by at the time when the biomarker analysis function 355 selects the value of the biomarker associated with the properties of the depicted object. This determination can be made through a consistency determination by a determination function 356 and a score calculation by a score calculation function 357, which will be described later.

[0039] Furthermore, the biomarker analysis function 355 is not limited to the biomarker values ​​acquired by the biomarker value acquisition function 352, and may further analyze the subject's past medical information to select a biomarker value related to the nature of the image calculated by the image analysis function 354. For example, the biomarker analysis function 355 may select a biomarker value related to the nature of the image by analyzing the subject's past medical information acquired from an electronic medical record (EHR) system and a medical image processing system (PACS), etc. The subject's past medical information may be, for example, past test values, needle biopsy treatment results, past images, or differences between those and the medical information acquired by the image acquisition function 351. The subject's past medical information is received from the electronic medical record system and the medical image processing system, etc. by the communication interface 33 and stored in the memory circuitry 34, and then read from the memory circuitry 34 by the biomarker analysis function 355 and used for analysis. Additionally, current medical information may be analyzed instead of or in addition to the historical medical information to select biomarker values ​​related to the characteristics of the features calculated by the image analysis function 354.

[0040] The determination function 356 determines the consistency between the values ​​of the biomarkers selected by the biomarker analysis function 355 and the properties of the depicted object calculated by the image analysis function 354.

[0041] For example, the determination function 356 determines the consistency for a subtype corresponding to the selected biomarker value based on whether the calculated property of the representation matches each of a plurality of known properties that the representation corresponding to the subtype should have, and the occurrence frequency of the matching property of the representation. For example, it is assumed that the image analysis function 354 calculates that the representation includes spicules, the corresponding subtype is luminal A breast cancer, and the occurrence frequency of spicules in luminal A breast cancer is A. In this case, the determination function 356 determines that the consistency (i.e., the degree of consistency) between the selected biomarker value (estrogen receptor: negative, progesterone receptor: positive, HER2: negative) related to luminal A breast cancer and spicules is A. On the other hand, it is assumed that the property that the representation corresponding to luminal A breast cancer should have includes calcification with an occurrence frequency of B, but calcification was not calculated by the image analysis function 354. In this case, the determination function 356 determines that there is no consistency between the value of the selected biomarker associated with luminal A breast cancer and calcification. The determination function 356 may determine the consistency using a determination table indicating a correspondence relationship between the type of subtype, a plurality of properties of the depiction that the subtype of the type should have, and the occurrence frequency of the properties of the depiction.

[0042] The score calculation function 357 calculates a score indicating the possibility that the depicted object is a lesion based on the consistency determined by the determination function 356. For example, the score calculation function 357 calculates a high score value when the determination function 356 determines that the consistency is high. On the other hand, the score calculation function 357 calculates a low score value when the determination function 356 determines that the consistency is low.

[0043] The region of interest identifying function 358 identifies a region of interest having high consistency from among the multiple regions of interest determined by the region of interest determining function 353 .

[0044] Next, an example of the operation of the medical information processing apparatus 3 according to the first embodiment configured as above will be described. Fig. 3 is a flowchart showing an example of the operation of the medical information processing apparatus 3 according to the first embodiment.

[0045] First, as shown in FIG. 3, the biomarker value acquisition function 352 acquires the values ​​of the biomarkers related to the lesion (step S1). For example, when the input operation of selecting a subject by the operator is accepted by the input interface 31, the biomarker value acquisition function 352 reads and acquires the biomarker values ​​of the selected subject from the storage circuitry 34. FIG. 4 is a diagram showing an example of the biomarker values. The biomarker value acquisition function 352 may display the acquired biomarker values ​​by the output interface 32. In FIG. 4, the biomarker values ​​of each of the multiple types of biomarkers BM1 to BM11 in time series are shown as rectangular images. The density of the rectangular image is higher as the biomarker value is larger. The symbol A1 in FIG. 4 is an indicator showing the density of the rectangular image (i.e., the value of the biomarker). The symbol A2 is the current value of the biomarker. The symbol A3 is an image showing whether or not the current biomarker value shown by the symbol "x" is within the normal range shown in the dashed frame A4.

[0046] After the values ​​of the biomarkers related to the lesion are acquired, as shown in FIG. 3, the image acquisition function 351 determines whether the values ​​of the biomarkers indicate that the subject may have a lesion (step S2). For example, if a positive epidermal growth factor receptor value or negative estrogen receptor, progesterone receptor, and HER2 values ​​are acquired, the image acquisition function 351 determines that the subject may have triple-negative breast cancer. For example, if one or both of the estrogen receptor and progesterone receptor values ​​are positive and a negative HER2 value is acquired, the image acquisition function 351 determines that the subject may have luminal A breast cancer. For example, if a positive HER2 value is acquired, the image acquisition function 351 determines that the subject may have HER2-positive breast cancer.

[0047] If the biomarker value indicates that the subject may have a lesion (step S2: YES), the image acquisition function 351 reads out and acquires a medical image of the subject from the storage circuitry 34. Then, the region of interest determination function 353 determines an ROI (region of interest) on the medical image acquired by the image acquisition function 351 (step S3). On the other hand, if the biomarker value does not indicate that the subject may have a lesion (step S2: NO), the process ends.

[0048] After the ROI is determined, the image analysis function 354 analyzes the image of the ROI to calculate the properties of the depicted object (step S4). FIG. 5 is a diagram showing the correspondence between the type of cancer and the mammography image findings in an example of the operation of the medical information processing device 3 according to the first embodiment. There is a correspondence between the type of cancer and the properties of the depicted object as shown in FIG. 5. For example, the depicted object indicating triple-negative breast cancer has multiple properties such as circular, single, finely lobulated margin (smooth for breast cancer), and low-frequency calcification. Also, for example, the depicted object indicating luminal A breast cancer has multiple properties such as irregular shape, spicules, and medium-frequency calcification. Also, for example, the depicted object indicating HER2-positive breast cancer has multiple properties such as pleomorphism, fine linear or branching calcification, and calcified mass.

[0049] After the characteristics of the imaging object are calculated, the biomarker analysis function 355 selects values ​​of biomarkers related to the characteristics of the imaging object (step S5), as shown in Figure 3. For example, the biomarker analysis function 355 calculates values ​​of biomarkers corresponding to lesions having the characteristics of the imaging object calculated by the image analysis function 354.

[0050] Next, the determination function 356 determines the consistency between the values ​​of the selected biomarkers and the properties of the depiction (step S6).

[0051] Fig. 6 is a diagram showing an example of a judgment table used for judging consistency in an operation example of the medical image processing device 3 according to the first embodiment. In the example shown in Fig. 6, the judgment function 356 judges consistency using a judgment table indicating the correspondence between the type of subtype corresponding to the value of the biomarker selected by the biomarker analysis function 355 and the properties of multiple depictions of the subtype of that type.

[0052] More specifically, the judgment table shown in FIG. 6 describes luminal A breast cancer, HER2-positive breast cancer, and triple-negative breast cancer as types of subtypes. In addition, the judgment table shown in FIG. 6 shows the occurrence frequency of each of multiple characteristics of the image with values ​​A to C. A indicates that the characteristic of the image corresponding to the subtype occurs at a high frequency. B indicates that the characteristic of the image corresponding to the subtype occurs at a medium frequency. C indicates that the characteristic of the image corresponding to the subtype occurs at a low frequency. An image characteristic that is not described as A, B, or C means that it does not occur in the corresponding subtype. Note that, for convenience of explaining the calculation of the score described later, some occurrence frequencies A to C are accompanied by numerical values ​​(such as α%) indicating specific occurrence frequencies. HER2-positive calcification: 76% and triple-negative breast cancer calcification: 12 to 36% shown in FIG. 6 correspond to the occurrence frequency of calcification in the mammography image findings shown in FIG. 5.

[0053] For example, assume that biomarker values ​​related to luminal A breast cancer (estrogen receptor: negative, progesterone receptor: positive, HER2: negative) and biomarker values ​​related to triple-negative breast cancer (epidermal growth factor receptor: positive, estrogen receptor: negative, HER2: negative) have been selected. Assume also that the properties of the depicted object are calculated to be accompanied by spicules, irregular shape, and smooth edges. In this case, according to the judgment table of FIG. 6, for luminal A breast cancer, two properties (spicules and irregular shape) are calculated after the three properties that a depicted object should have (spicules, calcification, and irregular shape). That is, for luminal A breast cancer, two items of consistency are satisfied. On the other hand, for triple-negative breast cancer, only one property (smooth edges) is calculated out of the four properties that a depicted object should have (calcification, round, single, smooth edges). That is, for triple-negative breast cancer, only one property (smooth edges) is calculated. In this case, the determination function 356 determines that the consistency for luminal A breast cancer is higher than the consistency for triple negative breast cancer.

[0054] After the consistency is determined, as shown in Fig. 3, the score calculation function 357 calculates a score indicating the possibility that the imaged object is a lesion based on the determined consistency (step S7). The score calculation function 357 outputs the calculated score to the output interface 32. For example, for luminal A breast cancer, when spicules and irregularities are calculated by the image analysis function 354 but calcification is not calculated, the score calculation function 357 calculates the score for luminal A breast cancer according to the following formula. Score = α × (1 - β) × γ (1) α is the specific occurrence frequency of spicules calculated by the image analysis function 354 (i.e., consistency). γ is the specific occurrence frequency of irregularities calculated by the image analysis function 354 (i.e., consistency). β is the specific occurrence frequency of calcification. However, since calcification is not calculated by the image analysis function 354 and is not consistent, (1-β) is used to calculate the score. 1-β is a value smaller than β and functions to reduce the score. Note that the score may be calculated according to the formula (1) and then converted into a simpler numerical value and output. Here, the parameters α, β, and γ indicating the occurrence frequency are set to values ​​of 0.5 or more to calculate the score. For example, when the frequency of "calcification" is 0.4, the frequency of "absence of calcification" is set to 0.6 as a parameter to calculate the score.

[0055] For subtypes other than luminal A breast cancer, the score is also calculated by multiplying the occurrence frequency of the property of the image calculated by image analysis function 354 by the value obtained by subtracting from 1 the occurrence frequency of the property of the image not calculated by image analysis function 354. This makes it possible to calculate a score of a magnitude according to the consistency.

[0056] After the score is calculated, the region of interest identifying function 358 identifies an ROI having a high score from among the ROIs when multiple ROIs have been determined by the region of interest determining function 353 (step S8). The region of interest identifying function 358 outputs the identified ROI having a high score to the output interface 32.

[0057] Fig. 7 is a flowchart showing a detailed operation example of the medical information processing device 3 according to the first embodiment. Fig. 8 is a diagram showing a mammography image and a region of interest in the operation example of the medical information processing device 3 according to the first embodiment. Figs. 7 and 8 show more detailed examples of some steps in the flowchart in Fig. 3.

[0058] 7 and 8, the region of interest determining function 353 determines ROI1 and ROI2 on the mammography image I (step S31). In the example shown in Fig. 8, ROI1 and ROI2 are determined on the left and right breasts, respectively.

[0059] 7 and 8, the image analysis function 354 calculates that the edge of the object in ROI1 is smooth as a property of the object (step S41), and calculates that the object in ROI2 is accompanied by calcification as a property of the object (step S42).

[0060] 7 and 8, the biomarker analysis function 355 selects, for ROI1, biomarker values ​​associated with triple-negative breast cancer, which is common when the image has a smooth edge (e.g., epidermal growth factor receptor: positive, estrogen receptor: negative, HER2: negative) (step S51). Also, for ROI2, the biomarker analysis function 355 selects, for ROI2, biomarker values ​​associated with luminal A breast cancer, which is common when the image has calcification (e.g., estrogen receptor: negative, progesterone receptor: positive, HER2: negative) (step S52).

[0061] 7 and 8, the determination function 356 determines that the consistency for ROI1 is low based on the small number of features of the depiction corresponding to triple-negative breast cancer calculated by the image analysis function 354 (step S61). On the other hand, the determination function 356 determines that the consistency for ROI2 is high based on the large number of features of the depiction corresponding to luminal A breast cancer calculated by the image analysis function 354 (step S62).

[0062] 7 and 8, the score calculation function 357 calculates a low score for ROI1 based on the determination that the consistency for ROI1 is low (step S71). On the other hand, the score calculation function 357 calculates a high score for ROI2 based on the determination that the consistency for ROI2 is high (step S72).

[0063] 7 and 8, the region of interest identifying function 358 identifies the region of interest with a high score as ROI2 (step S81). In this case, the determining function 356 may estimate information about the subject based at least on the region of interest with a high score (ROI2) and output the estimation result to the output interface 32. This estimation may be performed based on the nature of the depiction calculated in step S4 for the identified region of interest, or may be performed by inputting the values ​​of the biomarkers and an image of the identified region of interest into a machine learning model.

[0064] As described above, in the first embodiment, the image acquisition function 351 acquires a medical image of an object captured by imaging the object. The biomarker value acquisition function 352 acquires values ​​of biomarkers contained in a sample taken from the object. The region of interest determination function 353 determines a region of interest on the medical image acquired by the image acquisition function 351. The image analysis function 354 analyzes an image of the region of interest determined by the region of interest determination function 353 to calculate the nature of the depicted object. The biomarker analysis function 355 selects a biomarker value related to the nature of the depicted object calculated by the image analysis function 354. The determination function 356 determines the consistency between the biomarker value selected by the biomarker analysis function 355 and the nature of the depicted object calculated by the image analysis function 354.

[0065] This allows for efficient determination of consistency by selecting biomarker values ​​related to the properties of the depicted object within the region of interest. In addition, the presence or absence and location of a lesion can be determined with high accuracy based on the consistency between the selected biomarker values ​​and the properties of the depicted object. Therefore, according to the first embodiment, it is possible to improve the accuracy of lesion determination.

[0066] In addition, in the first embodiment, a biomarker value acquisition function 352 acquires the values ​​of biomarkers associated with lesions that may be present in the subject, and an image analysis function 354 calculates the properties of the image according to the type of lesion.

[0067] This makes it possible to determine the presence and location of a lesion with higher accuracy based on the consistency between the values ​​of the biomarkers related to the lesion and the properties of the depicted object according to the type of lesion, thereby further improving the accuracy of lesion determination.

[0068] Furthermore, in the first embodiment, the region of interest determining function 353 can determine a plurality of regions of interest on the medical image acquired by the image acquiring function 351. The region of interest identifying function 358 identifies a region of interest having high consistency from among the plurality of regions of interest determined by the region of interest determining function 353.

[0069] This makes it possible to easily determine, when multiple regions of interest have been determined, which region of interest should be the subject of needle biopsy (cytological or histological diagnosis).

[0070] In addition, in the first embodiment, the biomarker analysis function 355 further analyzes the past medical information to select biomarker values ​​related to the characteristics of the imaging object calculated by the image analysis function 354.

[0071] This allows the consistency to be determined with higher accuracy based additionally on the subject's past medical information, thereby further improving the accuracy of lesion determination.

[0072] Moreover, in the first embodiment, the score calculation function 357 calculates a score indicating the possibility that the imaged object is a lesion, based on the consistency determined by the determination function 356 .

[0073] This makes it easy to determine the likelihood that a corresponding lesion exists based on the score.

[0074] In addition, in the first embodiment, when the value of the biomarker acquired by the biomarker value acquisition function 352 indicates the possibility that a lesion is present in the subject, the region of interest determination function 353 determines the region of interest, and the image analysis function 354 calculates the properties of the depicted object.

[0075] This makes it possible to determine the location of a lesion based on the consistency between the biomarker values ​​and the properties of the extract when the location of the lesion cannot be determined from the biomarker values ​​alone.

[0076] (First Modification of the First Embodiment) Next, a first modification of the first embodiment in which a biomarker value is acquired after determining an ROI will be described, focusing on the differences from the above-mentioned embodiment. Fig. 9 is a flowchart showing the operation of the medical information processing device 3 according to the first modification of the first embodiment.

[0077] 9, first, the region of interest determining function 353 determines an ROI on a medical image (step S3). For example, when an input operation by an operator to select a subject is accepted by the input interface 31, the image acquiring function 351 reads and acquires the biomarker values ​​for the selected subject from the storage circuitry 34. Then, the region of interest determining function 353 determines an ROI on the medical image acquired by the image acquiring function 351 (step S3).

[0078] After the ROI is determined, the image analysis function 354 analyzes the image of the ROI to calculate the nature of the object (step S4).

[0079] After the properties of the imaging object are calculated, the biomarker value acquisition function 352 acquires the values ​​of the biomarkers associated with the lesion (step S1). For example, the biomarker value acquisition function 352 reads and acquires the values ​​of the biomarkers for the selected subject from the storage circuitry 34.

[0080] After the values ​​of the biomarkers associated with the lesion are obtained, the biomarker analysis function 355 determines whether or not the obtained biomarker values ​​indicate the possibility that a lesion is present in the subject (step S2).

[0081] If the biomarker value indicates the possibility that a lesion may exist in the subject (step S2: YES), the biomarker analysis function 355 selects the biomarker value related to the property of the image (step S5). Thereafter, the process from step S6 onwards is carried out as in Fig. 3. On the other hand, if the biomarker value does not indicate the possibility that a lesion may exist in the subject (step S2: NO), the process ends.

[0082] In the example shown in FIG. 9, as in the above-described embodiment, the presence and location of a lesion can be determined with high accuracy based on the consistency between the biomarker values ​​and the properties of the depicted object, thereby improving the accuracy of lesion determination.

[0083] (Second Modification of the First Embodiment) Next, a second modification of the first embodiment in which the ROI is modified in response to an input operation will be described, focusing on the differences from the above-mentioned embodiment. Fig. 10 is a flowchart showing the operation of the medical information processing device 3 according to the second modification of the first embodiment.

[0084] In the example shown in FIG. 10, after determining an ROI (step S3), the region of interest determining function 353 determines whether an input operation to correct the ROI has been accepted by the input interface 31 (step S9). For example, when an operator performs an input operation to select a subject on a screen for selecting a subject displayed by the output interface 32, the region of interest determining function 353 reads out a mammography image of the selected subject from the storage circuitry 34 and displays it on the screen. Then, the region of interest determining function 353 automatically determines an ROI suspected of being a lesion for the mammography image displayed on the screen using machine learning. The determined ROI is displayed as an enlarged image. The determined ROI is also displayed in a manner that allows manual correction.

[0085] When an input operation to modify the ROI is accepted (step S9: YES), the region of interest determination function 353 modifies the ROI according to the input operation (step S10). After the ROI is modified, or when an input operation to modify the ROI is not accepted (step S9: NO), the process from step S4 onwards is carried out. In step S9, in addition to or instead of the input operation to modify the ROI, a new ROI setting may be accepted from the operator.

[0086] According to the example shown in FIG. 10, the ROI can be modified, thereby improving convenience.

[0087] (Third Modification of the First Embodiment) Next, a third modification of the first embodiment, which outputs an alert when the score is low, will be described, focusing on the differences from the above-mentioned embodiments. Fig. 11 is a block diagram showing the configuration of a medical information processing device 3 according to the third modification of the first embodiment.

[0088] In the example shown in Fig. 11, the processing circuitry 35 of the medical information processing device 3 further includes an alert output function 359 in addition to the configuration shown in Fig. 2. The alert output function 359 is an example of an alert output unit. When the score calculated by the score calculation function 357 is lower than a threshold, the alert output function 359 outputs an alert notifying the possibility of a lesion being present in a site other than the region of interest via the output interface 32.

[0089] Fig. 12 is a flowchart showing an example of the operation of the medical image processing device 3 according to the third modified example of the first embodiment. In the example shown in Fig. 12, after the region of interest specifying function 358 specifies an ROI with a high score, the alert output function 359 determines whether the score of the specified ROI is equal to or higher than a threshold (step S9).

[0090] If the score of the identified ROI is not equal to or greater than the threshold (step S9: NO), the alert output function 359 outputs an alert (step S10). On the other hand, if the score of the identified ROI is equal to or greater than the threshold (step S9: YES), the process ends.

[0091] According to the examples shown in FIGS. 11 and 12, by outputting an alert notifying that a lesion may exist in a site other than the region of interest, it is possible to further improve the accuracy of lesion determination.

[0092] Second Embodiment Next, a second embodiment for improving the accuracy of determining an object not related to a lesion will be described, focusing on the difference from the above-mentioned embodiment.

[0093] In the second embodiment, the biomarker analysis function 355 selects a value of a biomarker not associated with a lesion that may be present in the subject when the value of the biomarker acquired by the biomarker value acquisition function 352 is not a value indicating that a lesion may be present in the subject. The determination function 356 determines the consistency between the value of the biomarker and the nature of the depicted object based on the value of the biomarker not associated with the lesion selected by the biomarker analysis function 355.

[0094] Fig. 13 is a flowchart showing an example of the operation of the medical image processing apparatus 3 according to the second embodiment. In the example shown in Fig. 13, the region of interest determining function 353 determines an ROI on a medical image (step S3).

[0095] After the ROI is determined, the image analysis function 354 analyzes the image of the ROI to calculate the nature of the object (step S4).

[0096] After the characteristics of the imaging object are calculated, the image acquisition function 351 acquires the values ​​of the biomarkers associated with the lesion (step S1). For example, the values ​​of the biomarkers for the selected subject are read from the storage circuitry 34 and acquired.

[0097] After the values ​​of the biomarkers associated with the lesion are acquired, the biomarker analysis function 355 determines whether the acquired biomarker values ​​indicate that a lesion may be present in the subject (step S2).

[0098] If the biomarker value indicates that the subject may have a lesion (step S2: YES), the biomarker analysis function 355 selects the biomarker value related to the property of the image (step S5). After selecting the biomarker value related to the property of the image, the process from step S6 onward is carried out as in FIG. 3.

[0099] On the other hand, if the biomarker value does not indicate that a lesion may be present in the subject (step S2: NO), the biomarker analysis function 355 selects the value of a biomarker related to fibrosis as a biomarker value not related to a lesion that may be present in the subject (step S11).

[0100] Fig. 14 is a diagram showing a mammography image before biopsy and a mammography image after biopsy in an operation example of the medical information processing device 3 according to the second embodiment. For example, as shown in Fig. 14, a mammography image I1 before needle biopsy shows suspicious microcalcifications, but after needle biopsy reveals ductal epithelial hyperplasia without dysplasia, formation of a noticeable scar S may be confirmed in a mammography image I2 obtained by a follow-up mammography 12 months later. The value of the biomarker related to fibrosis is, for example, the value of a biomarker related to such a scar S.

[0101] 13, the determination function 356 checks the subject's past needle biopsy results (step S12). For example, the determination function 356 reads out from the storage circuitry 34 the subject's past medical information that has been acquired from an electronic medical record system, a medical image processing system, or the like and stored in the storage circuitry 34. Then, the determination function 356 confirms the presence or absence of a needle biopsy and the site thereof based on the read out past needle biopsy results of the subject.

[0102] After checking the subject's past needle biopsy results, the determination function 356 determines the consistency between the biomarker value selected by the biomarker analysis function 355 and the nature of the imaging object based on the value of the biomarker related to fibrosis and the subject's past needle biopsy results. For example, if a biomarker value related to fibrosis is selected and the imaging object in the ROI is determined to be a scar from a past needle biopsy, the determination function 356 determines that the consistency for fibrosis is high. In the calculation of the score by the score calculation function 357, fibrosis is used as a parameter with a low value. Therefore, the score for fibrosis is lower than the score for the subtype.

[0103] Even if a breast cancer is judged to be benign after a needle biopsy, if new findings appear in a follow-up mammogram, it may be difficult to determine whether it is breast cancer or a scar caused by the needle biopsy. According to the second embodiment, false positives can be suppressed by highly accurately determining that there is no lesion based on the values ​​of biomarkers that are not related to the lesion. This makes it easier to determine whether there is a lesion.

[0104] (Modification of the second embodiment) Next, a modified example of the second embodiment will be described, focusing on the differences from the above-mentioned embodiment. Fig. 15 is a flowchart showing an example of the operation of the medical information processing device 3 according to the modified example of the second embodiment. Fig. 16 is a schematic diagram showing an example of the operation of the medical information processing device 3 according to the modified example of the second embodiment.

[0105] In the example shown in FIGS. 15 and 16, a check of the subject's past needle biopsy results (step S12) is performed as follows.

[0106] First, the determination function 356 reads and acquires from the storage circuitry 34 the biopsy implementation information of the past needle biopsy that has been acquired from the electronic medical record system, the medical image processing system, and the like and stored in the storage circuitry 34 (step S121).

[0107] After acquiring the past biopsy implementation information, the determination function 356 confirms the past biopsy position from the previous image, which is one of the past medical information of the subject. Then, the determination function 356 assigns an ROI to the confirmed biopsy position. biopsy is set (step S122).

[0108] ROI biopsy After setting, the determination function 356 performs warping processing on the previous image taking into consideration the degree of deformation of the current image with respect to the previous image, thereby generating a warped image (step S123).

[0109] After generating the warped image, the determination function 356 determines the ROI on the warped image. biopsy and the ROI on the current image determined in step S3 current In the example shown in FIG. 16, the determination function 356 overlays the current image and the warped image to determine the ROI. biopsy and ROI current Compare with.

[0110] ROI on warped image biopsy and the current ROI on the image current After comparing the ROIs, the determination function 356 determines the ROIs on the warped image based on the result of the determination. biopsy and the current ROI on the image current The similarity between the image and the scar is calculated (step S125). The similarity may be, for example, a Dice coefficient. When the calculated similarity is equal to or greater than a threshold, the determination function 356 determines that the image is a scar caused by needle biopsy.

[0111] According to a modification of the second embodiment, the accuracy of scar determination can be improved by warping the current image taking into account the degree of deformation of the current image relative to the previous image.

[0112] The term "processor" used in the above description means a circuit such as a central processing unit (CPU), a graphics processing unit (GPU), 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)). The processor realizes its function by reading and executing a program stored in a memory circuit. Instead of storing a program in a memory circuit, the processor may be configured to directly incorporate the program in the circuit. In this case, the processor realizes its function by reading and executing the program incorporated in the circuit. The processor is not limited to being configured as a single processor circuit, and may be configured as a single processor by combining multiple independent circuits to realize its function. Furthermore, the multiple components in FIG. 2 may be integrated into a single processor to realize its function.

[0113] In the first and second embodiments and their modified examples described above, the biomarker value related to the nature of the depicted object calculated by the image analysis function is selected by the biomarker analysis function, and the consistency between the selected biomarker value and the nature of the depicted object calculated by the image analysis function is judged by the judgment function. However, an embodiment in which the biomarker analysis function is omitted is also possible. In this case, the consistency between all of the biomarker values ​​or a part that does not depend on the nature of the depicted object and the nature of the depicted object is judged. For example, the consistency can be judged by selecting a function for calculating the consistency of the biomarker value according to the nature of the depicted object and inputting the biomarker value into the function.

[0114] According to this embodiment, the first and second embodiments, and their modified examples, a medical information processing device is disclosed that includes an image processing unit, a biomarker value acquisition unit, a region of interest determination unit, an image analysis unit, and a determination unit as described below. The image acquisition unit acquires a medical image of a subject. The biomarker value acquisition unit acquires values ​​of biomarkers contained in a sample collected from the subject. The region of interest determination unit determines a region of interest on the medical image acquired by the image acquisition unit. The image analysis unit analyzes an image of the region of interest determined by the region of interest determination unit to calculate the properties of the depicted object. The determination unit determines the consistency between all or a part of the values ​​of the biomarkers acquired by the biomarker value acquisition unit and the properties of the depicted object calculated by the image analysis unit.

[0115] According to at least one of the embodiments described above, it is possible to improve the accuracy of lesion determination.

[0116] Although several embodiments have been described above, these embodiments are presented only as examples and are not intended to limit the scope of the invention. The novel device and method described in this specification can be implemented in various other forms. In addition, various omissions, substitutions, and modifications can be made to the forms of the device and method described in this specification without departing from the spirit of the invention. The appended claims and their equivalents are intended to include such forms and modifications that fall within the scope and spirit of the invention. [Explanation of symbols]

[0117] 3 Medical information processing equipment 31 Input Interface 351 Image Acquisition Function 352 Biomarker value acquisition function 353 Region of interest determination function 354 Image analysis function 355 Biomarker analysis function 356 Judgment Function 357 Score Calculation Function 358 Region of interest identification function 359 Alert output function

Claims

1. an image acquisition unit for acquiring a medical image of a subject; a biomarker value acquiring unit that acquires values ​​of biomarkers contained in a sample collected from the subject; a region of interest determining unit that determines a region of interest on the medical image acquired by the image acquiring unit; an image analysis unit that analyzes an image of the region of interest determined by the region of interest determination unit and calculates a property of a depicted object; a biomarker analysis unit that selects a value of a biomarker that is related to a property of the depicted object calculated by the image analysis unit; a determination unit that determines the consistency between the values ​​of the biomarkers selected by the biomarker analysis unit and the properties of the depicted object calculated by the image analysis unit; A medical information processing device comprising:

2. the biomarker value acquisition unit acquires values ​​of biomarkers associated with a lesion that may be present in the subject; The image analysis unit calculates a property of the depicted object according to the type of the lesion. The medical information processing device according to claim 1 .

3. The region of interest determining unit is capable of determining a plurality of the regions of interest on the medical image acquired by the image acquiring unit, The medical image processing apparatus according to claim 1 , further comprising a region-of-interest specifying unit that specifies the region of interest with high consistency from among the plurality of regions of interest determined by the region-of-interest determining unit.

4. The medical image processing apparatus according to claim 1 , wherein the biomarker analysis unit further analyzes past medical information to select a value of a biomarker related to a property of the depicted object.

5. The medical image processing device according to claim 1 , further comprising a score calculation unit that calculates a score indicating a possibility that the depicted object is the lesion based on the consistency determined by the determination unit.

6. The medical information processing apparatus according to claim 5 , further comprising an alert output unit that outputs an alert when the score calculated by the score calculation unit is lower than a threshold value.

7. 2. The medical information processing device according to claim 1, wherein when the value of the biomarker acquired by the biomarker value acquisition unit indicates that a lesion may be present in the subject, the region of interest determination unit determines the region of interest, and the image analysis unit calculates the properties of the depicted object.

8. The medical image processing apparatus according to claim 1 , wherein the biomarker analysis unit selects a value of a biomarker associated with a property of the depicted object after the region of interest is determined by the region of interest determination unit.

9. The medical information processing device of claim 8, wherein the biomarker analysis unit selects a biomarker value related to a property of the imaging object when the value of the biomarker acquired by the biomarker value acquisition unit indicates that a lesion may be present in the subject after the region of interest has been determined by the region of interest determination unit.

10. the biomarker analysis unit, when the value of the biomarker acquired by the biomarker value acquisition unit is not a value indicating that a lesion may exist in the subject, selects a value of the biomarker that is not related to a lesion that may exist in the subject; The medical image processing apparatus according to claim 9 , wherein the determining unit determines the consistency based on values ​​of biomarkers not related to the lesion selected by the biomarker analysis unit.

11. The medical image processing device according to claim 10 , wherein the values ​​of the biomarkers not related to a lesion include values ​​of biomarkers related to fibrosis.

12. The medical image processing apparatus according to claim 1 , wherein the region of interest determining unit determines the region of interest based on machine learning.

13. The medical image processing apparatus according to claim 12 , further comprising an input interface configured to accept an input operation for correcting the region of interest determined by the region of interest determining unit.

14. The medical information processing apparatus according to claim 1 , wherein the sample collected from the subject is a body fluid of the subject.

15. the biomarker value acquisition unit acquires values ​​of a plurality of types of biomarkers; the biomarker analysis unit selects values ​​of a plurality of biomarkers from the plurality of types of biomarkers whose values ​​have been acquired; The medical information processing device according to claim 1 .

16. The medical information processing apparatus according to claim 3 , wherein the determining unit outputs an estimation result of information relating to a condition of the subject based on the identified region of interest.

17. On the computer, A procedure for acquiring a medical image of a subject; obtaining values ​​of biomarkers contained in a sample taken from the subject; determining a region of interest on the acquired medical image; analyzing the image of the determined region of interest to calculate a property of the feature; selecting a biomarker value associated with the calculated characteristic of the image; and determining a consistency between the selected biomarker values ​​and the calculated characteristics of the representation; A program for executing the above.

Images