JP7486318B2 - Medical information processing device, medical information processing method, and medical information processing program - Google Patents

Description

This embodiment relates to a medical information processing device, a medical information processing method, and a medical information processing program.

Traditionally, when a doctor performs medical procedures such as treatment or examination on a patient, the doctor grasps the patient's condition based on medical knowledge and past experience, and then selects the next treatment/examination to be performed. In recent years, in order to achieve higher quality medical care, attention has been focused on analyzing medical information using trained models that have been trained using vast amounts of past medical information, and presenting doctors with information that is useful for their decision-making (e.g., the probability of side effects occurring after treatment).

However, because there are individual differences in the interpretation of inference results output by trained models, etc., the basis for diagnostic inferences between patients with similar cases may vary. This creates a problem of inconsistent diagnoses between patients with similar cases. In addition, even if diagnoses vary between patients with similar cases due to a lack of consistent inferences, there is a problem of not being able to recognize that diagnostic inferences have varied.

JP 2017-10577 A

The problem that this invention aims to solve is to provide the operator with information showing the consistency between the basis for diagnostic inferences regarding similar cases and the basis for diagnostic inferences regarding the target subject.

The medical information processing apparatus according to the present embodiment includes a collection unit, a quantification unit, a consistency evaluation unit, and a display unit. The collection unit collects first medical information of a first subject based on a diagnostic purpose for the first subject , and collects second medical information of at least one second subject related to a similar case of the first subject according to the diagnostic purpose. The quantification unit determines a first index in the first medical information for ranking the importance of a basis for a first inference in a first diagnosis for the first subject based on the diagnostic purpose for the first subject. The consistency evaluation unit generates consistency information related to an evaluation of the consistency between the first inference and the second inference based on a second index in the second medical information for ranking the importance of a basis for a second inference in a second diagnosis for the second subject and the first index. The display unit displays the consistency information.

FIG. 1 is a diagram showing an example of the configuration of a medical information processing apparatus according to an embodiment, together with a hospital information system. FIG. 2 is a flowchart illustrating an example of a procedure of a consistency information providing process according to the embodiment. FIG. 3 is a flowchart showing an example of a procedure of a consistency information providing process according to the embodiment. FIG. 4 is a diagram showing an example of subject list information according to the embodiment. FIG. 5 is a diagram showing an example of an integrated display of objective-related information displayed on a display according to the embodiment. 6 is a diagram showing an example of input of importance in an importance menu in the various medical information shown in FIG. 5 according to the embodiment. FIG. 7 is a diagram showing an example of the correspondence between the importance and the grounds order for various types of medical information in the purpose-related information according to the embodiment. FIG. 8 is a diagram showing an example of a first prediction result and a first prediction rank displayed on a display according to the embodiment. FIG. 9 is a diagram showing a first prediction result together with the granularity of medical information in the first predicted type of medical treatment according to the embodiment. FIG. 10 is a diagram showing an example of a change in a first index related to blood test results in an integrated display according to an embodiment. FIG. 11 is a diagram showing an example of type variation values based on a comparison of types of medical information between a first subject and a second subject according to the embodiment. FIG. 12 is a diagram showing an example of consistency information for each inference phase according to the embodiment. FIG. 13 is a diagram showing an example of a display showing a comparison of types of medical information by specifying inference phase 3 for consistency information and similar cases according to the embodiment. FIG. 14 is a diagram showing an example of consistency information for each inference phase according to the embodiment. FIG. 15 is a diagram showing an example of a display showing the type of medical information together with the order of reasons by specifying inference phase 1 for consistency information and two similar cases according to the embodiment. FIG. 16 is a diagram showing an example of the type and reference level of medical information that is the basis of the first inference in the embodiment. FIG. 17 is a diagram showing an example of consistency information for each inference phase according to the embodiment. FIG. 18 is a diagram showing an example of a display in which types of medical information and reference degrees are arranged in order of grounds ranking by specifying inference phase 3 for consistency information according to an embodiment. FIG. 19 is a diagram showing an example of consistency information indicating an average change amount for the type of assistive diagnosis and an inference basis change amount for a specified type of assistive diagnosis, according to an application example of the embodiment. FIG. 20 is a diagram showing an example of a display in which types of medical information before and after an assisted diagnosis are arranged in order of grounds by specifying a second subject and a type of assisted diagnosis for consistency information, according to an application example of the embodiment. FIG. 21 is a flowchart illustrating an example of a procedure of a consistency information providing process according to a modification of the embodiment. FIG. 22 is a diagram showing an example of consistency information for each inference phase according to a modified example of the embodiment.

The medical information processing method and the medical information processing program will be described in detail below with reference to the drawings. In the following embodiments, parts with the same reference numerals perform similar operations, and redundant descriptions will be omitted as appropriate. Note that the medical information processing device according to the present application is not limited to the embodiments shown below.

(Embodiment)
The overall configuration of a medical information processing device according to an embodiment will be described. FIG. 1 is a diagram showing an example of the configuration of a medical information processing device 1 according to an embodiment together with a hospital information system (HIS: Hospital Information System) 3. The medical information processing device 1 is connected to the hospital information system 3 via a communication interface 11. First, the hospital information system 3 related to the medical information processing device 1 will be described, and then the medical information processing device 1 will be described. The hospital information system 3 has an electronic medical record system, a radiology department information system, a medical image diagnostic device, a medical image management system, a clinical examination department information system, and other department systems. The electronic medical record system, the radiology department information system, the medical image diagnostic device, the medical image management system, the clinical examination department information system, and other department systems are connected to each other via a wired or wireless communication network.

The electronic medical record system is an information system that manages electronic medical records for recording medical treatment details. The electronic medical record system includes an electronic medical record server device and an electronic medical record terminal device. The electronic medical record server device is a computer device that executes processes related to the management of electronic medical records. The electronic medical record terminal device is used by doctors, nurses, etc. who input and refer to electronic medical records. The electronic medical record server device and the electronic medical record terminal device are connected to a communication network. The electronic medical record terminal device inputs an examination order for a subject at the instruction of a user. If the examination order is for imaging a subject for image diagnosis, the examination order includes information on the examination contents, such as the examination area, imaging target, disease name, examination purpose, and modality. The examination order may also include information such as the patient ID, patient name, examination ID, and examination date of the subject undergoing the examination. In response to the input of the examination order, the electronic medical record server device transmits the examination order to the radiology department information system together with the patient information of the subject in the examination order. If the test order is for a clinical test to be performed on a subject, the electronic medical record server device sends the test order to the clinical laboratory department information system.

The test results obtained by the test order are linked to the electronic medical record for the subject. At this time, the clinician who requested the test makes a diagnosis based on the test results and records the diagnosis in the subject's electronic medical record. The diagnosis includes information such as the diagnosis name, stage, and severity of the disease. The stage and severity of the disease are defined in a predetermined definition document. The electronic medical record terminal device inputs the diagnosis results based on various tests on the subject, for example, at the instruction of the clinician. The diagnosis results are stored in the electronic medical record server device.

Radiology information systems (RIS) are information systems that manage information in the radiology department of a hospital, and are equipped with a radiology information server device and a radiology information terminal device. The radiology information server device is a computer device that executes processes related to the management of information in the radiology department. The radiology information server device and the radiology information terminal device are connected to a communication network. As an example of the operation of the radiology information server device, the radiology information server device receives an examination order from an electronic medical record system and obtains information about a medical imaging diagnostic device from the examination order. The radiology information server device transmits the examination order to a medical imaging diagnostic device specified by the examination order. When the examination order is transmitted to the medical imaging diagnostic device, the radiology information server device may add information such as a patient ID and an examination date to a DICOM (Digital Imaging and Communications in Medicine) tag or the like. The radiology department information terminal device is used, for example, by radiology doctors who interpret medical images to create interpretation reports for medical images.

Medical imaging diagnostic devices are devices that acquire medical images, such as X-ray diagnostic devices, X-ray CT (Computed Tomography) devices, and magnetic resonance imaging devices. Based on an examination order from the radiology department information system, the medical imaging diagnostic device performs imaging of the subject in response to an operation by a technician or the like to start imaging. By performing imaging, the medical imaging diagnostic device collects data indicating the state of the subject's internal tissues. The medical imaging diagnostic device generates medical images based on the data, and transmits the generated medical images to a medical image management system together with the examination order.

Medical image management systems (PACS: Picture Archiving and Communication Systems) are information systems that manage medical images such as X-ray CT images and MR images. Medical image management systems are equipped with a medical image management server device. The medical image management server device is a computer device that executes processes related to the management of medical images and is connected to a communication network. When the medical image management server device receives medical images and order information from a medical image diagnostic device, it associates the medical images with the examination orders and stores them.

The Laboratory Information System (LIS) is an information system that manages information in the clinical laboratory of a hospital, and is equipped with a Laboratory Information Server, a Laboratory Information Terminal, and testing equipment. In the Laboratory Information System, clinical technologists perform specimen testing and physiological function testing based on test orders sent from the electronic medical record system. Sample testing may also be outsourced to external testing organizations such as hygiene laboratories. The Laboratory Information Server is a computer device that executes processes related to the management of information in clinical testing. The Laboratory Information Terminal is used by laboratory technologists and others in the clinical laboratory. The Laboratory Information Server and the Laboratory Information Terminal are connected to a communication network.

The testing equipment is equipment related to specimen testing, in which blood, urine, and other specimens obtained from patients are analyzed by clinical laboratory technicians, and physiological function testing, in which the patient's electroencephalogram, electrocardiogram, and the like are measured. The specimen testing includes, for example, pathological testing, blood and biochemistry testing, general testing of urine and stool, immunoserological testing, microbiological testing, and tests related to blood transfusion and organ transplantation. The physiological function testing includes, for example, electroencephalogram testing, respiratory function testing, cardiac testing, and fundus photography testing. A plurality of testing equipment is installed according to the type of specimen testing and the type of physiological function testing. At least a portion of the specimen testing may be performed by an external testing institution. In this case, the clinical testing department information server device receives and stores the test results related to the specimen testing from the external testing institution. The clinical testing department information server device also transmits the test results to the electronic medical record system.

Examples of other department systems include a surgery department system, a rehabilitation department system, or a dialysis department system. As with the radiology department information system and clinical laboratory department information system described above, the other department systems also create surgery implementation reports, rehabilitation implementation reports, and the like in response to orders issued by doctors on electronic medical record terminal devices. The other department systems send the created surgery implementation reports, rehabilitation implementation reports, and the like to the electronic medical record system.

The medical information processing device 1 is connected to the hospital information system 3 via a communication interface 11. The medical information processing device 1 may be incorporated into the hospital information system 3. The medical information processing device 1 may also be realized as a client-server system. The medical information processing device 1 shown in FIG. 1 has a communication interface 11, a memory 13 (storage unit), a display 15 (display unit), an input interface 17 (input unit), and a processing circuit 19 (processing unit). Data communication between the communication interface 11, the memory 13, the display 15, the input interface 17, and the processing circuit 19 is performed via a bus (BUS).

The communication interface 11 performs data communication with the hospital information system 3. The standard for communication between the communication interface 11 and the hospital information system 3 may be any standard, such as HL7 (Heart Level 7), DICOM, or both.

The memory 13 is a storage device such as a hard disk drive (HDD), a solid state drive (SSD), or an integrated circuit storage device that stores various information. The memory 13 stores data generated by executing, for example, the collection function 193, the quantification function 195, the estimation function 197, the consistency evaluation function 199, and the like. The generated data and these functions will be described later. The memory 13 stores various types of medical information collected from the hospital information system 3 by the collection function 193. The medical information includes clinical information such as various test results, medical images, and radiology reports linked to the subject's electronic medical record. The memory 13 may be a drive device that reads and writes various information between the memory 13 and a portable storage medium such as a compact disc (CD), a digital versatile disc (DVD), or a flash memory, or a semiconductor memory element such as a random access memory (RAM). The memory 13 also stores the control program according to this embodiment.

The display 15 displays various information. For example, the display 15 outputs a GUI (Graphical User Interface) for accepting various operations from the user. For example, the display 15 may be a liquid crystal display (LCD), a cathode ray tube (CRT), an organic electroluminescence display (OELD), a plasma display, or any other display, as appropriate. The display 15 may be a desktop type, or may be configured as a tablet terminal capable of wireless communication with the medical information processing device 1 main body.

The input interface 17 accepts various input operations from the user, converts the accepted input operations into electrical signals, and outputs them to the processing circuit 19. For example, the input interface 17 accepts selection instructions and various conditions from the user. As the input interface 17, for example, a mouse, keyboard, trackball, switch, button, joystick, touchpad, touch panel display, etc. can be used as appropriate. In this embodiment, the input interface 17 is not limited to one equipped with physical operation parts such as a mouse, keyboard, trackball, switch, button, joystick, touchpad, and touch panel display. 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 this electrical signal to the processing circuit 19 is also included as an example of the input interface 17. In addition, the input interface 17 may be configured as a tablet terminal or the like that can wirelessly communicate with the medical information processing device 1 main body.

The processing circuitry 19 controls the operation of the entire medical information processing device 1 in response to electrical signals of input operations output from the input interface 17. For example, the processing circuitry 19 has, as hardware resources, processors such as a CPU (Central Processing Unit), MPU (Micro Processing Unit), and GPU (Graphics Processing Unit), and memories such as a ROM (Read Only Memory) and RAM. The processing circuit 19 may also be realized by an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), other complex programmable logic devices (CPLDs), simple programmable logic devices (SPLDs), etc.

The processing circuit 19 has a system control function 191, a collection function 193, a quantification function 195, an estimation function 197, and a consistency evaluation function 199. Each function of the system control function 191, the collection function 193, the quantification function 195, the estimation function 197, and the consistency evaluation function 199 is stored in the memory 13 in the form of a program executable by a computer. The processing circuit 19 executes the system control function 191, the collection function 193, the quantification function 195, the estimation function 197, and the consistency evaluation function 199 by a processor that executes the programs deployed in the memory.

That is, the processing circuit 19 corresponds to a processor that realizes a function corresponding to each program by reading and executing the program from the memory 13. In other words, the processing circuit 19 in a state in which each program has been read has a function corresponding to the read program. Note that each function (191, 193, 195, 197, 199) is not limited to being realized by a single processing circuit. A processing circuit may be configured by combining multiple independent processors, and each processor may realize each function (191, 193, 195, 197, 199) by executing a program. The processing circuit 19 that realizes the system control function 191, collection function 193, quantification function 195, estimation function 197, and consistency evaluation function 199 is an example of a system control unit, collection unit, quantification unit, estimation unit, and consistency evaluation unit.

The processing circuit 19 controls each function of the processing circuit 19 based on the input operation received from the user via the input interface 17 by the system control function 191. Specifically, the processing circuit 9 reads out the control program stored in the memory 13, expands it on the memory in the processing circuit 19, and controls each part of the medical information processing device 1 according to the expanded control program. The processing circuit 19 controls the display 15 for various displays on the display 15. The collection function 193, quantification function 195, estimation function 197, and consistency evaluation function 199 realized by the processing circuit 19 will be described later along with the procedure for executing a process (hereinafter referred to as a consistency information providing process) for generating and displaying consistency information indicating the consistency between the basis for inference regarding the diagnosis of similar cases and the basis for inference regarding the diagnosis of the target subject.

The above describes the overall configuration of the medical information processing device 1 and the hospital information system 3. Below, the procedure for the consistency information provision process will be described. Figures 2 and 3 are flowcharts showing an example of the procedure for the consistency information provision process.

(Consistency information provision process)
(Step S201)
The display 15 displays the subject list information under the control of the system control function 191. The subject list information is information showing, as a list, items related to selection such as the subject (patient) ID to be diagnosed, the patient name, whether or not assisted diagnosis is required (hereinafter referred to as whether or not assisted diagnosis is required), the purpose of diagnosis, whether or not priority is given to the input of importance (hereinafter referred to as whether or not priority is given to the input of importance), and display of medical information. The importance corresponds to, for example, an index (hereinafter referred to as a first index) indicating the degree of importance regarding the grounds of inference in the diagnosis of the subject, that is, the weight of importance. The larger the value of the importance, the greater the influence on the grounds of inference. Hereinafter, for the sake of concrete explanation, the importance is assumed to be in five stages from 1 to 5. At this time, the medical information corresponding to an importance of 5 corresponds to the most important information on the grounds of inference in the diagnosis of the subject.

Figure 4 is a diagram showing an example of subject list information PLI. When the operator performs a diagnosis on each of multiple subjects, the operator gives instructions via the input interface 17 to input a selection for each subject in the subject list information regarding the need for an assisted diagnosis, the purpose of the diagnosis, the need for priority of importance input, and the display of medical information. When inputting the need for an assisted diagnosis, the purpose of the diagnosis, and the need for priority of importance input, a selection menu corresponding to these items is displayed, for example, by a context menu or a pull-down menu. For ease of explanation, the subject to be diagnosed will be referred to as the first subject below.

The collection function 193 collects medical information on the first subject (hereinafter referred to as the first medical information) from the hospital information system 3 based on the patient ID of the first subject selected in the subject list information. The collection function 193 stores the collected first medical information in the memory 13.

The selection of whether or not an assisted diagnosis is required for the first subject and the selection of whether or not importance input priority is required correspond to the operator selecting an inference process (hereinafter referred to as an inference path) regarding the diagnosis of the first subject. For example, if No is selected for whether or not an assisted diagnosis is required, the input selection menu for the diagnostic purpose and whether or not importance input priority is required is hidden or disabled. Hereinafter, the inference path at this time is referred to as the first inference path. The first inference path corresponds to performing a diagnostic inference (hereinafter referred to as the first inference) for the first subject without using an assisted diagnosis algorithm for the assisted diagnosis.

The inference path when Yes is selected for whether assisted diagnosis is required, a diagnostic purpose is input, and Yes is selected for whether importance input priority is required is called the second inference path. The second inference path corresponds to the process of determining a first index for medical information related to the diagnostic purpose (hereinafter referred to as purpose-related information) among medical information on a first subject (hereinafter referred to as first medical information), then outputting a result of a diagnostic prediction for the first subject (hereinafter referred to as a first diagnostic prediction result) by an assisted diagnosis algorithm, and finally inputting a change to the determined first index or maintaining the determined first index.

The inference path when Yes is selected for whether assisted diagnosis is required, a diagnosis purpose is input, and No is selected for whether importance input priority is required is called the third inference path. The third inference path corresponds to the process of outputting the first diagnosis prediction result using the assisted diagnosis algorithm, and then determining the first index for the purpose-related information.

In view of the above, in this step, one of the first to third inference paths is selected by inputting various selection instructions for the first subject in the subject list information. The method of selecting an inference path is not limited to the above. For example, an inference path selection menu may be displayed on the display 15 together with the subject list information. At this time, in response to the selection of an inference path, the collection function 193 collects the first medical information from the hospital information system 3. In the following, for the sake of concrete explanation, the diagnostic objective is described as weight loss, one of the side effects after radiation therapy, as shown in FIG. 4.

(Step S202)
The collection function 193 selects search conditions for similar cases of the first subject according to an instruction from the operator via the input interface 17. The search conditions are, for example, the disease name, the International Classification of Diseases and Related Health Problems (hereinafter referred to as ICD code), the medical department, the doctor in charge, image pattern matching, clinical findings text matching, and the like for the first subject. The disease name, ICD code, medical department, the doctor in charge, images, findings, and the like are linked to the electronic medical record. The search conditions are displayed, for example, by a context menu or a pull-down menu. The operator selects the search conditions displayed by the context menu or the pull-down menu, and the like, via the input interface 17.

(Step S203)
The collection function 193 collects at least one medical information of a second subject related to a similar case of the first subject (hereinafter referred to as second medical information). Specifically, the collection function 193 identifies a patient ID of the second subject based on a search condition. Next, the collection function 193 collects the second medical information from the hospital information system 3 based on the identified patient ID of the second subject. The collection function 193 stores the collected second medical information in the memory 13. The second medical information has a second index indicating the degree of grounds for the second inference in the second diagnosis for the second subject. The second index is an index indicating the degree of importance regarding the grounds for the inference of the second diagnosis for the second subject.

(Step S204)
If the inference path selected in step S201 is the first inference path (Yes in step S204), the process proceeds to step S213. If the inference path selected in step S201 is not the first inference path (No in step S204), that is, if the inference path selected in step S201 is the second or third inference path, the process proceeds to step S205.

(Step S205)
If the inference path selected in step S201 is the second inference path (Yes in step S205), the process proceeds to step S206. If the inference path selected in step S201 is not the second inference path (No in step S205), that is, if the inference path selected in step S201 is the third inference path, the process proceeds to step S208.

(Step S206)
The collection function 193 extracts purpose-related information from the first medical information based on the diagnostic purpose selected for the first subject. For example, the collection function 193 checks the selected diagnostic purpose against a correspondence table of types of medical information for diagnostic purposes (hereinafter, a purpose-related LUT (Look Up Table)). The purpose-related LUT is stored in the memory 13 in advance. The collection function 193 identifies the type of medical information corresponding to the diagnostic purpose. Next, the collection function 193 uses the identified type of medical information to extract purpose-related information from the first medical information. The system control function 191 displays the extracted purpose-related information on the display 15 together with patient information related to the first subject. That is, the display 15 displays medical information according to the diagnostic purpose in an integrated manner. The system control function 191 may also display the collected first medical information in an integrated manner on the display 15. This limits the first medical information displayed on the display 15 according to the diagnostic purpose.

Figure 5 is a diagram showing an example of an integrated display IGD of objective-related information displayed on the display 15. Since the diagnostic objective is weight loss as a side effect after radiation therapy, medical information related to the diagnostic objective is integrated and displayed on the display 15 as objective-related information. As shown in Figure 5, the display 15 displays image information II on the first subject, the result of drug administration by injection IR, the distribution of dose in radiation therapy DD, and the result of blood test BR in an integrated manner. The distribution of dose in radiation therapy DD is, for example, a differential or integral type DVH (Dose Volume Histogram). Note that the integrated and displayed medical information and display layout are not limited to those described above, and can be set and changed as appropriate depending on the diagnostic objective, etc.

(Step S207)
The quantification function 195 determines, in the first medical information, a first index indicating the degree of grounds for the first inference in the first diagnosis for the first subject based on the diagnostic purpose for the first subject. For example, the quantification function 195 determines the first index by an instruction from the operator for the purpose-related information displayed on the display 15. Specifically, in the purpose-related information, the medical information that is the grounds for the first inference is selected by an instruction from the operator via the input interface 17. Next, the quantification function 195 causes the display 15 to display a menu regarding the selection of the importance (hereinafter referred to as the importance selection menu) for the selected medical information, triggered by a predetermined operation via the input interface 17. The predetermined operation is, for example, a right click or double click on the mouse. The quantification function 195 may cause the display 15 to display the importance selection menu in response to the movement of the cursor on the selected medical information in the integrated display IGD.

The operator instructs the input interface 17 to input a selection of importance in the importance menu. As a result, the quantification function 195 quantifies the degree of justification for the first inference in the first diagnosis for the first subject. The quantification function 195 may record all information referred to by the operator (hereinafter referred to as reference information) in the memory 13 using the operator's operation history and gaze tracking technology such as eye tracking for the operator. The reference information is, for example, the order in which the medical information was referred to, the frequency of operations on the medical information, the time the medical information was referred to, etc. in the integrated display IGD of the objective-related information displayed on the display 15. At this time, the quantification function 195 may determine the importance corresponding to the first index based on the reference information.

FIG. 6 is a diagram showing an example of the input of the importance in the importance menu IM for the various medical information shown in FIG. 5. As shown in FIG. 6, the importance of the image information II is input as 4, the importance of the distribution of the dose in radiation therapy DD is input as 4, and the importance of the blood test result BR is input as 2. The quantification function 195 may determine the importance corresponding to the combination of medical information selected by the operator's instruction via the input interface 17. The combination of medical information is realized by various operations on the user interface (UI), such as a + button or a shift key and left click displayed for the medical information to be combined. The quantification function 195 displays one importance menu for the selected medical information on the display 15. At this time, the quantification function 195 highlights the medical information selected in the integrated display IGD. The quantification function 195 stores the quantified importance as a first index in the memory 13 in association with the type of the corresponding medical information. In addition, the quantification function 195 may display, in an importance menu, the importance level recommended according to the combination of numerical values of medical information in a combination of medical information related to a diagnostic purpose.

The quantification function 195 determines the order of importance (hereinafter referred to as the grounds order) based on the determined importance value. For example, as shown in FIG. 6, if the importance of image information II is 4, the importance of dose distribution DD in radiation therapy is 4, and the importance of blood test result BR is 2, the quantification function 195 determines the grounds order of both image information II and dose distribution DD as 1, and the grounds order of blood test result BR as 3. The quantification function 195 stores the determined grounds order in the memory 13 in association with the type of corresponding medical information. The quantification function 195 may also store the grounds order and the importance together as a first index in the memory 13. In this case, the first index corresponds to a value indicating the grounds order or the importance.

Figure 7 is a diagram showing an example of the correspondence between the importance and the evidence order for various types of medical information in the purpose-related information. As shown in Figure 7, if the importance is 2, 1, 4, 3, and 5 for CPT-11 (irinotecan injection), blood test for albumin, lung CT image, tumor DVH, and clinical record, respectively, the evidence order will be 4, 5, 2, 3, and 1, respectively.

(Step S208)
The estimation function 197 reads out an assisted diagnostic algorithm corresponding to the diagnostic purpose from the memory 13. The assisted diagnostic algorithm is, for example, a trained model that uses purpose-related information as an input and outputs a result of prediction regarding the diagnosis of the first subject corresponding to the diagnostic purpose (hereinafter referred to as a first predicted result). The first predicted result has, for example, a first predicted diagnosis for the first subject, a first prediction index indicating the degree of the basis of the first predicted inference in the first predicted diagnosis, and a type of medical information related to the first prediction index (hereinafter referred to as a first predicted medical type). The first predicted medical type corresponds to an explanatory variable used for inference by the trained model and contributed to the first predicted diagnosis. The first predicted diagnosis is, for example, a probability related to the diagnostic purpose (hereinafter referred to as a predicted probability). In addition, the first prediction index corresponds to a weight of importance regarding the basis of inference in the first predicted diagnosis. That is, the larger the value of the first prediction index, the greater the influence on the basis of inference of the first predicted diagnosis. The assisted diagnosis algorithm is not limited to a trained model such as a support vector machine (SVM), and may be an algorithm capable of executing assisted diagnosis such as a model using logistic regression or a decision tree. The estimation function 197 inputs the objective-related information to the assisted diagnosis algorithm read from the memory 13, and outputs a first prediction result.

The prediction function 197 determines the order of importance of the first predicted medical type in the first predicted diagnosis (hereinafter referred to as the first prediction order) based on the first prediction index. The prediction function 197 displays the first prediction result and the first prediction order on the display 15. The prediction function 197 associates the first prediction result and the first prediction order with the patient information of the first subject, stores them in the memory 13, and transmits them to the hospital information system 3 via the communication interface 11. The hospital information system 3 stores the first prediction result and the first prediction order in association with the electronic medical record of the first subject using the patient information of the first subject.

Figure 8 is a diagram showing an example of the first prediction result and the first prediction rank displayed on the display 15. Figure 8 shows that the predicted probability of weight loss for the first subject as a side effect after radiation therapy is 85%. Also, as shown in Figure 8, the first prediction indexes for the injection of cis-diamminedichloroplatinum (CDDP), blood test for albumin, CT image of the left lung, distribution of dose to the tumor, and basic patient information are 0.05, 4.3, 2.1, 8.5, and 0.6, respectively, and the first prediction ranks are 5, 2, 3, 1, and 4, respectively.

Figure 9 shows the first prediction result together with the granularity of the medical information in the first predicted medical type. As shown in Figure 9, the granularity of the medical information in the first predicted medical type is shown as major items, medium items, and minor items. The minor items are included in the medium items, and the medium items are included in the major items. For example, if the first predicted medical type is injection: CDDP, the first predicted medical type is classified as medication as a major item, injection as a medium item, and CDDP as a minor item. The granularity of the medical information is not limited to the three types of major items, medium items, and minor items, and can be set arbitrarily. The memory 13 stores a correspondence table (hereinafter referred to as a granularity correspondence table) between combinations of types of medical information and diagnostic purposes according to diagnostic purposes, and granularity according to tests in the medical information, etc.

(Step S209)
The collection function 193 collects prediction results (hereinafter referred to as second prediction results) for the diagnosis of the second subject according to the diagnosis purpose from the hospital information system 3 based on the second medical information. In addition, when the second prediction result is included in the second medical information, the collection function 193 extracts the second prediction result from the second medical information. The second prediction result has, for example, a second predicted diagnosis for the second subject, a second prediction index indicating the degree of the basis of the second prediction inference in the second predicted diagnosis, a type of medical information related to the second prediction index (hereinafter referred to as the second predicted medical type), and a ranking of the importance of the second predicted medical type in the second predicted diagnosis (hereinafter referred to as the second prediction ranking). The second predicted diagnosis is, for example, a prediction probability for the second subject. In addition, the second prediction index corresponds to the weight of the importance of the basis of the inference in the second predicted diagnosis. That is, the larger the value of the second prediction index, the greater the influence on the basis of the inference of the second predicted diagnosis.

If the second prediction result cannot be collected from the hospital information system 3, the estimation function 197 inputs the second medical information into the assisted diagnosis algorithm used in the processing in step S208, and outputs the second prediction result. Next, the estimation function 197 determines the second prediction ranking based on the second prediction index.

(Step S210)
If the inference path selected in step S201 is the second inference path (Yes in step S210), the process proceeds to step S211. If the inference path selected in step S201 is not the second inference path (No in step S210), that is, if the inference path selected in step S201 is the third inference path, the process proceeds to step S216.

(Step S211)
The input interface 17 inputs an instruction to change the first index in response to an instruction from the operator for the objective-related information displayed on the display 15, i.e., in the integrated display IGD. The quantification function 195 stores the changed first index (hereinafter, referred to as the changed index) in the memory 13 in association with the type of medical information corresponding to the changed index. Note that if the first index is not changed, the processing of this step is omitted.

Figure 10 is a diagram showing an example of a change in the first index related to blood test result BR in the integrated display IGD. Compared to Figure 6, in Figure 10, the importance related to blood test result BR has been changed from 2 to 4 by inputting a change in importance to the importance menu CIM related to blood test result BR. Since inputting a change instruction is similar to inputting the importance, a description will be omitted. Note that in this step, an importance may be input for medical information for which an importance was not set in step S207, and the determination of the importance may be canceled for medical information for which an importance was determined in step S207.

(Step S212)
The consistency evaluation function 199 generates consistency information based on the first index, the second index, the changed first index, the first prediction index, and the second prediction index. The consistency evaluation function 199 displays the generated consistency information on the display 15. For example, the consistency evaluation function 199 generates the consistency information by calculating a difference between the basis for the first inference and the basis for the second inference as a type variation value indicating the degree of variation in the type of medical information based on the first index and the second index.

Specifically, the consistency evaluation function 199 first adjusts the granularity of the medical information in the consistency information to the larger granularity using a granularity correspondence table, which is a correspondence table regarding the granularity of medical information, for the medical information on which the first inference is based, the medical information on which the second inference is based, the medical information related to the first prediction index, and the medical information related to the second prediction index. For example, if the medical information on which the first prediction inference and the second prediction inference are based is the pixel intensity of a CT image, and the medical information on which the first inference and the second inference are based is a CT image, the consistency evaluation function 199 adjusts the granularity of the medical information to the larger granularity, i.e., the CT image.

After adjusting the granularity of the medical information, the consistency evaluation function 199 calculates a type variation value by counting the difference between the basis for the second inference and the basis for the first inference for each type of medical information. The consistency evaluation function 199 generates consistency information by calculating the difference between the basis for the second predictive inference and the basis for the first predictive inference as a type prediction variation value indicating the degree of variation in the type of medical information based on the first prediction index and the second prediction index. The consistency evaluation function 199 also generates consistency information by calculating the difference between the basis for the second inference and the basis for the first inference as a type variation value indicating the degree of variation in the type of medical information based on the changed first index and second index.

The display 15 displays the consistency information. Specifically, the display 15 displays at least one of the type variability value and the type predicted variability value together with the name of the second subject.

Figure 11 is a diagram showing an example of a type variation value based on a comparison between the type of medical information related to the four first indexes for the first subject and the type of medical information related to the four second indexes for the second subject. As shown in the comparison in Figure 11, the blood test for albumin is used as the basis for the first inference in the first diagnosis for the first subject, but is not used as the basis for the second inference in the second diagnosis for the second subject. In addition, the presence or absence of medication is not used as the basis for the first inference in the first diagnosis for the first subject, but is used as the basis for the second inference in the second diagnosis for the second subject. Therefore, as a type variation value, 1 is counted as the number of pieces of medical information that refer only to the first subject, and 1 is counted as the number of pieces of medical information that refer only to the second subject.

12 is a diagram showing an example of consistency information for each inference phase. Inference phase 1 shown in FIG. 12 shows an example of a difference in inference grounds by a doctor who is an operator. The + side of the vertical axis of the graph shown in FIG. 12 indicates the type variation value referring only to the first subject. The - side of the vertical axis of the graph shown in FIG. 12 indicates the type variation value or type prediction variation value referring only to the second subject. Inference phase 1 shown in FIG. 12 shows an example of a difference in inference grounds by a doctor who is an operator (type variation value). Note that the consistency information corresponding to inference phase 1 may be generated immediately after the processing of step S207 and displayed on the display 15. Inference phase 2 shown in FIG. 12 shows an example of a difference in inference grounds by an assisted diagnosis algorithm (type prediction variation value). Note that the consistency information corresponding to inference phase 2 may be generated immediately after the processing of step S208 and displayed on the display 15. Inference phase 3 shown in FIG. 12 shows an example of a difference in inference grounds (type variation value) due to a change in the first index after the first prediction result is displayed. As shown in FIG. 12, the type variability value in inference phase 3 is smaller than the type variability value in inference phase 1. Therefore, the consistency of the diagnosis for the first subject is improved.

In the consistency information shown in FIG. 12 displayed on the display 15, the input interface 17 may input the designation of at least one name of the second subject at the instruction of the operator. When at least one name of the second subject is designated, more specifically, when an inference phase and at least one similar case are designated, the display 15 displays multiple types of medical information that are the basis of the first inference among the first medical information, and multiple types of medical information that are the basis of the second inference among the second medical information related to the second subject corresponding to the designated name.

Figure 13 is a diagram showing an example of a display DIK of the type of medical information related to the inference basis for the first and second subjects, by specifying inference phase 3 and a similar case (black dotted frame) in the consistency information shown in Figure 12. As shown in Figure 13, details of the difference in the inference basis for the first and second subjects for the specified inference phase 3 and second subject are displayed on the display 15. At this time, the consistency evaluation function 199 may use a granularity correspondence table to display the contents of the type of medical information in a tree (structure display) on the display 15.

After adjusting the granularity of the medical information, the consistency evaluation function 199 generates consistency information by calculating an index variation value indicating the degree of variation of the second index relative to the first index based on the first index and the second index. Specifically, the consistency evaluation function 199 calculates the absolute value of the difference between the first index and the second index for each type of medical information having the same inference basis. Next, the consistency evaluation function 199 adds up the calculated absolute value over the total number of types of medical information related to the inference basis. Finally, the consistency evaluation function 199 calculates the index variation value by dividing the addition result by the total number. The display 15 displays the calculated index variation value together with the name of the second subject.

After adjusting the granularity of the medical information, the consistency evaluation function 199 generates consistency information by calculating an index prediction variation value indicating the degree of variation of the second prediction index relative to the first prediction index based on the first prediction index and the second prediction index. Specifically, the consistency evaluation function 199 calculates the absolute value of the difference between the first prediction index and the second prediction index for each type of medical information having the same inference basis. Next, the consistency evaluation function 199 calculates the index prediction variation value by adding the calculated absolute value over the total number of types of medical information related to the inference basis and dividing the sum by the total number. The display 15 displays the calculated index prediction variation value together with the name of the second subject. The calculation of the index variation value or the index prediction variation value is shown, for example, by the following formula.

In the left side of the formula (1), diff corresponds to an index variation value or an index prediction variation value. In the right side of the formula (1), N is the total number of types of medical information related to the inference grounds. In the right side of the formula (1), i is a variable indicating the type of medical information related to _the inference grounds. In the right side of the formula (1), x _i corresponds to the first index or the first prediction index.

corresponds to the second index or the second prediction index.

Figure 14 is a diagram showing an example of consistency information for each inference phase. The vertical axis of the graph shown in Figure 14 indicates the index variation value or the index prediction variation value. The index variation value corresponding to inference phase 1 shown in Figure 14 may be generated immediately after the processing of step S207 and displayed on the display 15. The index prediction variation value corresponding to inference phase 2 shown in Figure 14 may be generated immediately after the processing of step S208 and displayed on the display 15. As shown in Figure 14, the index variation value in inference phase 3 is smaller than the index variation value in inference phase 1. This improves the consistency of the diagnosis for the first subject.

In the consistency information displayed on the display 15, the input interface 17 may input the designation of at least one of the names of the second subject and inference phase 2 in response to an instruction from the operator. At this time, the display 15 displays multiple types of medical information that are the basis of the first predictive inference from among the first medical information, and multiple types of medical information that are the basis of the second predictive inference from among the first medical information.

Figure 15 is a diagram showing an example of a display TDIK that shows the type of clinical information related to the inference basis for the first subject and two second subjects together with the basis rank, by specifying inference phase 1 and two similar cases (black dotted frame) in the consistency information shown in Figure 14. As shown in Figure 15, for the specified inference phase 1 and two second subjects, details of the differences in the inference basis between the first subject and the two second subjects are displayed on the display 15 in the order of the basis rank.

After adjusting the granularity of the medical information, the consistency evaluation function 199 calculates a reference degree indicating the degree to which the basis for the second inference is referenced to the basis for the first inference based on the type of medical information among the basis for the second inference that overlaps with the basis for the first inference and the total number of similar cases. The display 15 displays the type of medical information among the first medical information or the purpose-related information that is the basis for the first inference and the calculated reference degree as consistency information.

In the following, for the sake of concrete explanation, the types of medical information corresponding to the basis of the first inference, i.e., the first indicator, are assumed to be five types: CT images, dose plan, age, whether or not medication is administered, and blood test for albumin. The total number of similar cases is assumed to be 10. Of the 10 second subjects, the number of second subjects who used CT images as the basis of the second inference is assumed to be 9. Of the 10 second subjects, the number of second subjects who used dose plans as the basis of the second inference is assumed to be 8. Of the 10 second subjects, the number of second subjects who used age as the basis of the second inference is assumed to be 7. Of the 10 second subjects, the number of second subjects who used whether or not medication is administered as the basis of the second inference is assumed to be 7. Of the 10 second subjects, the number of second subjects who used blood test for albumin as the basis of the second inference is assumed to be 2.

In this case, the consistency evaluation function 199 calculates the reference degree for the CT images as 9/10 = 90%, the reference degree for the dose plan as 8/10 = 80%, the reference degree for the age as 7/10 = 70%, the reference degree for the presence or absence of medication as 3/10 = 90%, and the reference degree for the albumin blood test as 2/10 = 20%.

Figure 16 is a diagram showing an example of the type of medical information that is the basis of the first inference and the calculated reference level displayed on the display 15. Note that the display 15 may display the number of similar cases that were referenced instead of the reference level.

After adjusting the granularity of the medical information, the consistency evaluation function 199 calculates a degree of inference reference indicating the degree to which the basis of the second predictive inference is referenced to the basis of the first predictive inference based on the type of medical information among the basis of the second predictive inference that overlaps with the basis of the first predictive inference and the total number of similar cases. The calculation of the inference reference degree is the same as the calculation of the reference degree described above, so a description thereof will be omitted. The display 15 displays the type of medical information among the second medical information that is the basis of the second predictive inference and the inference reference degree as consistency information.

Figure 17 is a diagram showing an example of consistency information for each inference phase. The vertical axis of the graph shown in Figure 17 indicates the reference degree (%) or the estimated reference degree (%). The reference degree corresponding to inference phase 1 shown in Figure 17 may be generated immediately after the processing of step S207 and displayed on the display 15. The estimated reference degree corresponding to inference phase 2 shown in Figure 14 may be generated immediately after the processing of step S208 and displayed on the display 15.

In the consistency information shown in FIG. 17 displayed on the display 15, the input interface 17 may input a designation of an inference phase in response to an instruction from the operator. At this time, the display 15 displays the type of medical information that is the basis of the first inference and the reference degree or estimated reference degree in descending order of importance in the designated inference phase.

Figure 18 is a diagram showing an example of a displayed DRI in which the type of medical information and the reference level relating to the inference basis for the first subject are arranged in order of the basis rank by specifying inference phase 3, which is indicated by a black dotted line frame, in the consistency information shown in Figure 17. As shown in Figure 18, for the specified inference phase 3, the type of medical information that is the basis for the first inference and the reference level are arranged in order of the basis rank and displayed on the display 15. Note that PS in Figure 18 stands for performance status.

(Step S213)
The display 15 displays the collected first medical information in an integrated manner. The amount of information displayed in this step is greater than the amount of information displayed in step S206. The display layout of the first medical information may be the same as that shown in FIG. 5, for example, or may be different.

(Step S214)
The quantification function 195 determines a first index indicating the degree of evidence of the first inference in the first diagnosis for the first subject for the displayed first medical information. For example, the quantification function 195 determines the first index based on an instruction from the operator for the first medical information displayed on the display 15. The method of determining the first index is similar to the process in step S207, and therefore will not be described.

(Step S215)
The consistency evaluation function 199 generates consistency information based on the first index and the second index. The display 15 displays the generated consistency information. The generation and display of the consistency information in this step corresponds to the parts related to the inference phase 1 in Figures 12 to 15, 17 and 18, and Figure 16, so a description thereof will be omitted.

(Step S216)
The collection function 193 extracts purpose-related information from the first medical information based on the selected diagnostic purpose for the first subject. The system control function 191 displays the extracted purpose-related information together with the patient information for the first subject on the display 15. The process in this step is similar to that in step S206, and therefore a detailed description thereof will be omitted.

(Step S217)
The quantification function 195 determines, in the first medical information, a first index indicating the degree of grounds for the first inference in the first diagnosis for the first subject based on the diagnostic purpose for the first subject. The process in this step is similar to that in step S207, and therefore a detailed description thereof will be omitted.

(Step S218)
The consistency evaluation function 199 generates consistency information based on the first index, the second index, the first prediction index, and the second prediction index. The consistency evaluation function 199 displays the generated consistency information on the display 15. The generation and display of the consistency information in this step corresponds to the parts relating to inference phase 2 and inference phase 3 in Figures 12 to 15, 17 and 18, and Figure 16, so a description thereof will be omitted. This completes the consistency information providing process. After completing the consistency information providing process, the operator makes a final decision regarding the diagnosis of the first subject.

According to the medical information processing device 1 of the embodiment described above, the first medical information of the first subject is collected, the second medical information of at least one second subject related to a similar case of the first subject is collected, and a first index indicating the degree of the basis of the first inference in the first diagnosis of the first subject is determined in the first medical information based on the diagnostic purpose of the first subject, and consistency information regarding the evaluation of the consistency between the first inference and the second inference is generated based on the second index indicating the degree of the basis of the second inference in the second diagnosis of the second subject and the first index in the second medical information, and the consistency information is displayed. That is, according to the medical information processing device 1, information indicating the consistency between the basis of the inference regarding the diagnosis of the similar case and the basis of the inference regarding the diagnosis of the target subject can be visualized and provided to the operator. The operator can easily grasp the difference between the basis of the inference regarding the subject to be diagnosed and the basis of the inference regarding the similar case by the consistency information, and can make a final decision regarding the diagnosis of the subject to be diagnosed by referring to the consistency information. From the above, the medical information processing device 1 can improve the consistency of the basis for diagnostic inferences between the subject to be diagnosed and similar cases.

In addition, according to the medical information processing device 1 of the embodiment, the operator inputs an instruction to change the first index, and consistency information is generated based on the changed first index and second index. As a result, according to the medical information processing device 1, it is possible to visualize and provide to the operator information indicating the consistency between the basis for inference regarding the diagnosis of similar cases and the basis for inference after the change regarding the diagnosis of the target subject.

Furthermore, according to the medical information processing device 1 of the embodiment, the granularity of the medical information in the consistency information is adjusted to the larger granularity by using a correspondence table regarding the granularity of medical information for the medical information that is the basis for the first inference and the medical information that is the basis for the second inference. According to the medical information processing device 1, the granularity of the medical information in the consistency information is adjusted to the larger granularity by using a correspondence table regarding the granularity of medical information for the medical information related to the first prediction index and the medical information related to the second prediction index. As a result, the granularity of the medical information can be adjusted to provide information indicating consistency to the operator.

In addition, according to the medical information processing device 1 of the embodiment, consistency information is generated by calculating an index variation value based on the first index and the second index, and the index variation value is displayed together with the name of the second subject. According to the medical information processing device 1, consistency information is generated by calculating an index prediction variation value indicating the degree of variation of the second prediction index relative to the first prediction index based on the first prediction index and the second prediction index, and the index prediction variation value is displayed together with the name of the second subject. In this way, the quantified index variation value and index prediction variation value can be visualized and provided to the operator as information indicating consistency.

In addition, according to the medical information processing device 1 of the embodiment, when one of the names of the second subject is specified, multiple types of medical information from the first medical information that are the basis for the first inference and multiple types of medical information from the second medical information related to the second subject corresponding to the specified name that are the basis for the second inference are displayed. According to the medical information processing device 1, multiple types of medical information from the first medical information that are the basis for the first predictive inference and multiple types of medical information from the first medical information that are the basis for the second predictive inference are displayed. As a result, the types of medical information referenced by the doctor or assisted diagnosis for the first subject and the second subject can be visualized and presented to the operator.

According to the medical information processing device 1 according to the embodiment, the difference between the basis of the second inference and the basis of the first inference is calculated as a type variation value based on the first index and the second index, thereby generating consistency information, and displaying the type variation value together with the name of the second subject. According to the medical information processing device 1 according to the embodiment, the first medical information is applied to at least one assisted diagnosis algorithm according to the diagnosis purpose, and the first predicted diagnosis, the first prediction index, and the type of medical information related to the first prediction index are output, and the second predicted diagnosis, the second prediction index, and the type of medical information related to the second prediction index are collected based on the second medical information, and the difference between the basis of the second predictive inference and the basis of the first predictive inference is calculated as a type prediction variation value based on the first prediction index and the second prediction index, thereby generating consistency information, and displaying the type prediction variation value together with the name of the second subject. As a result, the quantified type variation value and the type prediction variation value can be visualized and provided to the operator as information indicating consistency.

In addition, according to the medical information processing device 1 of the embodiment, a reference degree indicating the degree to which the basis of the second inference is referenced to the basis of the first inference is calculated based on the type of medical information among the basis of the second inference that overlaps with the basis of the first inference and the total number of similar cases, and the type of medical information among the first medical information that is the basis of the first inference and the reference degree are displayed as consistency information. According to the medical information processing device 1, a guess reference degree indicating the degree to which the basis of the second predictive inference is referenced to the basis of the first predictive inference is calculated based on the type of medical information among the basis of the second predictive inference that overlaps with the basis of the first predictive inference and the total number of similar cases, and the type of medical information among the first medical information that is the basis of the first predictive inference and the guess reference degree are displayed as consistency information. As a result, the quantified reference degree and guess reference degree can be visualized and provided to the operator as information indicating consistency.

(Application example)
In this application example, the change in the variation of importance before and after the execution of the assisted diagnosis (hereinafter referred to as the inference basis change amount) and the average of the inference basis change amounts over a number of similar cases (hereinafter referred to as the average change amount) are calculated, and the inference basis change amount and the average change amount are provided to the operator. The process in this application example is executed, for example, after step S211 or step S212 shown in FIG. 3.

When the first index is changed by the operator after the type predicted variation value is displayed, the consistency evaluation function 199 calculates a changed variation value indicating the degree of variation of the second index relative to the changed first index based on the changed first index and second index. Specifically, the consistency evaluation function 199 calculates the changed variation value as the left side of the formula (1) by calculating the formula (1) using the first index changed as x _i on the right side of the formula (1). The calculation content of the formula (1) is similar to the description in the embodiment, so the description is omitted. The consistency evaluation function 199 calculates the change amount between the index variation value and the changed variation value as the inference basis change amount according to the type of the assisted diagnostic algorithm. The consistency evaluation function 199 stores the inference basis change amount in the memory 13 in association with the type of the assisted diagnostic algorithm (hereinafter referred to as the assisted diagnostic type) and the name of the patient. Furthermore, the consistency evaluation function 199 transmits the inference basis change amount and the type of assisted diagnosis to the hospital information system 3 so that they are linked to the patient information or electronic medical record of the first subject and stored.

The collection function 193 extracts the inference basis change amount and the assisted diagnosis type corresponding to the assisted diagnosis type from the second medical information related to the second subjects corresponding to the searched similar cases.

The consistency evaluation function 199 calculates the average change amount by adding up the inference basis change amounts for similar cases and dividing the sum by the total number of similar cases. The consistency evaluation function 199 associates the average change amount with the type of assisted diagnosis and stores it in the memory 13.

The display 15 displays the inference basis change amount and the average change amount along with the type of assisted diagnosis and the name of the second subject. When one of the names of the second subject and the type of assisted diagnosis are designated by the operator via the input interface 17, the display 15 displays multiple types of clinical information related to the first inference before and after execution of the designated assisted diagnosis algorithm, and multiple types of clinical information related to the second inference in the second subject corresponding to the designated name.

Figure 19 is a diagram showing an example of consistency information showing the average change amount for the assisted diagnosis type and the inference basis change amount for the first subject and two second subjects for the specified assisted diagnosis type. As shown in Figure 19, the average change amount in the radiation side effect prediction model by logistic regression is 25, which is larger than the average change amount in the other radiation side effect prediction models. Therefore, the second prediction result by the radiation side effect prediction model by logistic regression has a greater influence on the first index, which is the importance of the basis for the doctor's inference, compared to the second prediction result by the other radiation side effect prediction models. As shown in Figure 19, the radiation side effect prediction model by logistic regression is selected in the list showing the average change amount for each assisted diagnosis type. At this time, the display 15 displays the inference basis change amount for each of the three second subjects by the radiation side effect prediction model by logistic regression together with the name of the second subject.

Figure 20 is a diagram showing an example of a display in which the types of medical information for the first and second subjects before and after the assisted diagnosis are arranged in order of basis rank when the name of the second subject (Patient Taro) and a radiation side effect prediction model using logistic regression are specified in the consistency information shown in Figure 19. At this time, the consistency evaluation function 199 highlights the type of medical information with the greatest change in importance before and after the assisted diagnosis.

According to the medical information processing device 1 according to the application example described above, an index variation value indicating the degree of variation of the second index relative to the first index is calculated based on the first index and the second index, and when the first index is changed by the operator after the display of the type predicted variation value, a changed variation value indicating the degree of variation of the second index relative to the changed first index is calculated based on the changed first index and second index, and the amount of change between the index variation value and the changed variation value is calculated according to the type of assisted diagnostic algorithm, and the calculated amount of change is displayed as consistency information together with the type of assisted diagnostic algorithm and the name of the second subject. In addition, according to the medical information processing device 1, when one of the names of the second subject and the type of assisted diagnostic algorithm are specified, multiple types of medical information related to the first inference and multiple types of medical information related to the second inference in the second subject corresponding to the specified name are displayed before and after the execution of the specified assisted diagnostic algorithm.

This allows the operator to see the changes in the basis for inference between the subject being diagnosed and similar cases before and after the assisted diagnosis is performed, improving the consistency of the basis for inference regarding the diagnosis.

(Modification)
The consistency information providing process in the modified example is to generate and display consistency information regarding the second subject before determining the first index. The procedure of the consistency information providing process in the modified example will be described below. Fig. 21 is a flowchart showing an example of the procedure of the consistency information providing process.

(Consistency information processing)
(Step S21)
The collection function 193 selects search conditions for similar cases of the first subject in response to an instruction from the operator via the input interface 17. The processing content of this step is similar to the processing content of step S202 in FIG. 2, and therefore detailed description thereof will be omitted.

(Step S22)
The collection function 193 extracts purpose-related information from the first medical information based on the diagnostic purpose selected for the first subject. The system control function 191 displays the extracted purpose-related information together with the patient information for the first subject on the display 15. The processing content of this step is similar to the processing content of step S206 in FIG. 2, and therefore a detailed description thereof will be omitted.

(Step S23)
The collection function 193 collects second medical information and a second prediction result of a second subject related to a similar case of the first subject based on the first medical information and the search conditions. The collection of the second medical information and the second prediction result is similar to the processing contents in steps S203 and S209 in FIG. 2, and therefore a detailed description thereof will be omitted.

(Step S24)
The consistency evaluation function 199 generates consistency information regarding the evaluation of the consistency of the second inference in the second diagnosis for the second subjects in the second medical information based on a second index indicating the degree of grounds for the second inference in the second diagnosis for the second subjects in the second medical information. The consistency information generated in this step corresponds to, for example, the reference degree between similar cases (hereinafter referred to as the similar reference degree) and the inference reference degree (hereinafter referred to as the similar inference reference degree). Specifically, the consistency evaluation function 199 calculates the similarity reference degree by dividing the count number of the type across similar cases for each type of medical information that is the basis of the second inference by the total number of similar cases after adjusting the granularity of the second medical information between similar cases. The consistency evaluation function 199 calculates the similarity inference reference degree by dividing the count number of the type across similar cases for each type of medical information that is the basis of the second prediction inference by the total number of similar cases after adjusting the granularity of the second medical information between similar cases.

The display 15 displays, as consistency information, the type of medical information from the second medical information that is the basis for the second inference and the calculated similarity reference degree. The display 15 displays, as consistency information, the type of medical information from the second medical information that is the basis for the second predictive inference and the similarity inference reference degree.

Figure 22 is a diagram showing an example of consistency information for each inference phase. The vertical axis of the graph shown in Figure 22 indicates the similarity reference degree (%) or the similarity inference reference degree (%). In the consistency information shown in Figure 22 displayed on the display 15, the input interface 17 may input a specification of an inference phase indicated by a black dotted frame at the instruction of the operator. At this time, the display 15 displays, in descending order of importance, the type of medical information that is the basis for the second inference or second predictive inference, and the similarity reference degree or similarity inference reference degree in the specified inference phase.

(Step S25)
The prediction function 197 applies the objective-related information to an assisted diagnosis algorithm according to the diagnostic objective, and outputs a first prediction result. The prediction function 197 displays the first prediction result on the display 15. The processing content of this step is similar to the processing content of step S208 in FIG. 2, and therefore a detailed description thereof will be omitted.

(Step S26)
The quantification function 195 determines the first index based on the operator's instruction for the objective-related information displayed on the display 15. The processing content of this step is similar to the processing content of step S207 in Fig. 2, so detailed description will be omitted. With the above, the consistency information providing processing in this modified example ends. Note that after this step, the processing of step S212 in Fig. 2 may be executed. After the consistency information providing processing ends, the operator makes a final decision regarding the diagnosis of the first subject.

In view of the above, according to the medical information processing device 1 of the modified example, second medical information of each of multiple second subjects related to similar cases of the first subject is collected based on the first medical information related to the first subject, consistency information regarding the evaluation of the consistency of inference in the multiple second subjects is generated based on an index indicating the degree of basis of inference in the diagnosis for the multiple second subjects in the second medical information, and the consistency information is displayed. As a result, according to the present medical information processing device 1, consistency information regarding similar cases can be presented to the operator before the first index related to the first subject is determined and before the first prediction result is output, and consistency in the basis of inference regarding diagnosis can be improved between the subject to be diagnosed and the similar cases.

As another variation of this embodiment, the medical information processing device 1 may be incorporated into the hospital information system 3. The medical information processing device 1 may also be realized by a client-server system, a thin client, cloud computing, or the like. In this case, the communication interface 11, memory 13, and processing circuitry 19 are mounted on the server, and the display 15 and input interface 17 are mounted on the client.

When the technical idea of this embodiment is realized in a medical information processing method, the medical information processing method collects first medical information of a first subject, collects second medical information of at least one second subject related to a similar case of the first subject, determines a first index in the first medical information indicating the degree of justification for a first inference in a first diagnosis for the first subject based on a diagnostic purpose for the first subject, generates a second index in the second medical information indicating the degree of justification for a second inference in a second diagnosis for the second subject and consistency information regarding an evaluation of the consistency between the first inference and the second inference based on the first index, and displays the consistency information. The processing procedure and effects of the medical information processing method are similar to those of the embodiment, and therefore will not be described.

When the technical idea of this embodiment is realized by a medical information processing program, the medical information processing program causes a computer to collect first medical information of a first subject, collect second medical information of at least one second subject related to a similar case of the first subject, determine a first index in the first medical information indicating the degree of justification for a first inference in a first diagnosis for the first subject based on a diagnostic purpose for the first subject, generate a second index in the second medical information indicating the degree of justification for a second inference in a second diagnosis for the second subject, generate consistency information regarding an evaluation of the consistency between the first inference and the second inference based on the first index, and display the consistency information.

For example, consistent information provision processing can be realized by installing a medical information processing program in a computer in the hospital information system 3 and expanding it in memory. In this case, the program that can cause the computer to execute the method can also be stored and distributed on a storage medium such as a magnetic disk (such as a hard disk), an optical disk (such as a CD-ROM or DVD), or a semiconductor memory. The processing procedure and effects of the medical information processing program are the same as those in the embodiment, so a description thereof will be omitted.

According to at least one of the embodiments, application examples, and modified examples described above, it is possible to provide the operator with information indicating the consistency between the basis for inference regarding the diagnosis of similar cases and the basis for inference regarding the diagnosis of the target subject. This makes it possible to reduce the variability in the basis for diagnosis of the subject to be diagnosed and multiple subjects in similar cases.

Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. These embodiments and their modifications are within the scope of the invention and its equivalents as set forth in the claims, as well as the scope and gist of the invention.

REFERENCE SIGNS LIST 1 Medical information processing device 3 Hospital information system 11 Communication interface 13 Memory 15 Display 17 Input interface 19 Processing circuit 191 System control function 193 Acquisition function 195 Quantification function 197 Estimation function 199 Consistency evaluation function

Claims (15)

an acquisition unit that acquires first medical information of a first subject based on a diagnostic purpose for the first subject, and acquires second medical information of at least one second subject related to a similar case of the first subject according to the diagnostic purpose;
a quantification unit that determines a first index for ranking the importance of a basis for a first inference in a first diagnosis for the first subject, based on the first medical information;
a consistency evaluation unit that generates consistency information regarding an evaluation of the consistency between the first inference and the second inference based on a second index and the first index, the consistency evaluation unit generating consistency information regarding an evaluation of the consistency between the first inference and the second inference based on the second index and the first index, the consistency evaluation unit generating consistency information regarding an evaluation of the consistency between the first inference and the second inference based on the second index and the second medical information;
A display unit that displays the consistency information;
A medical information processing device comprising: An input unit for inputting a change instruction for the first index,
The consistency assessor generates the consistency information based on the changed first index and the changed second index.
The medical information processing device according to claim 1 . The consistency evaluation unit includes:
Calculating a difference between the basis of the first inference and the basis of the second inference as a type variation value indicating a degree of variation in types of medical information based on the first medical information having the first index and the second medical information having the second index;
the display unit displays the type variation value together with the name of the second subject.
The medical information processing device according to claim 1 . The consistency evaluation unit includes:
Calculating an absolute value of a difference between the first index and the second index for each type of medical information having the same inference basis in the first inference and the second inference;
Adding up the calculated absolute value over the total number of types of medical information related to the inference basis;
Calculating an index variability value by dividing the result of the addition by the total number;
the display unit displays the index variation value together with the name of the second subject.
The medical image processing device according to claim 1 . The display unit is
When one name of the names of the second subject is designated, a plurality of types of medical information that are the basis of the first inference among the first medical information, and a plurality of types of medical information that are the basis of the second inference among the second medical information related to the second subject corresponding to the designated name are displayed.
The medical information processing device according to claim 3 . and a prediction unit that applies the first medical information to at least one assisted diagnostic algorithm corresponding to the diagnostic purpose, and outputs a first predicted diagnosis for the first subject, a first prediction index for ranking the importance of a basis for a first prediction inference in the first predicted diagnosis, and a type of medical information related to the first prediction index,
the collecting unit collects, based on the second medical information, a second predictive diagnosis for the second subject, a second predictive index for ranking the importance of a basis for a second predictive inference in the second predictive diagnosis, and a type of medical information related to the second predictive index;
The consistency evaluation unit calculates a type prediction variation value indicating a degree of variation in the type of medical information based on a difference in type of medical information between the prediction result having the first prediction indicator and the prediction result having the second prediction indicator ;
the display unit displays the type predicted variation value together with the name of the second object.
The medical image processing device according to claim 1 . The consistency evaluation unit includes:
adjusting the granularity of the medical information in the consistency information to a larger granularity by using a correspondence table regarding the granularity of the medical information in the medical information on which the first inference is based, the medical information on which the second inference is based, the medical information related to the first prediction index, and the medical information related to the second prediction index;
The medical information processing device according to claim 6 . The consistency evaluation unit includes:
In the first prediction inference and the second prediction inference, an absolute value of a difference between the first prediction index and the second prediction index is calculated for each type of medical information having the same inference basis;
Adding the calculated absolute value over the total number of types of medical information related to the inference basis;
Calculate the index forecast variance value by dividing the result of the addition by the total number;
The display unit displays the index predicted variability value together with the name of the second subject.
The medical information processing device according to claim 6 or 7. The consistency evaluation unit includes:
Calculating an absolute value of a difference between the first index and the second index for each type of medical information having the same inference basis in the first inference and the second inference;
Adding up the calculated absolute value over the total number of types of medical information related to the inference basis;
Calculating an index variability value by dividing the result of the addition by the total number;
if the first index is changed after the type predicted variation value is displayed, a post-change variation value indicating a degree of variation of the second index relative to the changed first index is calculated based on the changed first index and the second index;
calculating a change amount between the index variation value and the changed variation value in accordance with a type of the assisted diagnostic algorithm;
the display unit displays the amount of change together with the type of the assisted diagnostic algorithm and the name of the second subject as the consistency information.
The medical image processing device according to claim 6 . The display unit is
when one of the names of the second subject and the type of the assisted diagnostic algorithm are designated, displaying a plurality of types of clinical information related to the first inference and a plurality of types of clinical information related to the second inference in the second subject corresponding to the designated name, before and after execution of the designated assisted diagnostic algorithm.
The medical information processing device according to claim 9 . the consistency evaluation unit calculates a degree of inference reference indicating a degree to which the basis for the second predictive inference is referenced to the basis for the first predictive inference based on a type of medical information among the basis for the second predictive inference that overlaps with the basis for the first predictive inference and the total number of similar cases;
The display unit displays, as the consistency information, a type of medical information that is the basis of the first predictive inference among the first medical information and the estimated reference degree.
The medical image processing device according to claim 6 . the consistency evaluation unit calculates a reference degree indicating an extent to which the basis for the second inference is referenced to the basis for the first inference based on a type of medical information that overlaps with the basis for the first inference among the basis for the second inference and a total number of the similar cases;
The display unit displays, as the consistency information, a type of medical information that is the basis of the first inference among the first medical information and the reference degree.
The medical image processing device according to claim 1 . an acquisition unit that acquires second medical information of each of a plurality of second subjects related to similar cases of the first subject based on first medical information related to the first subject in accordance with a diagnostic purpose for the first subject;
a consistency evaluation unit that generates consistency information regarding an evaluation of the consistency of the inference in the second subjects based on an index for ranking the importance of grounds for inference in the diagnosis of the second subjects in the second medical information;
A display unit that displays the consistency information;
A medical information processing device comprising: a collection unit collects first medical information of a first subject based on a diagnostic purpose for the first subject, and collects second medical information of at least one second subject related to a similar case of the first subject according to the diagnostic purpose;
A quantification unit determines a first index in the first medical information for ranking the importance of a basis for a first inference in a first diagnosis for the first subject based on a diagnostic purpose for the first subject;
a consistency evaluation unit generates consistency information regarding an evaluation of the consistency between the first inference and the second inference based on a second index for ranking the importance of a basis for the second inference in the second diagnosis for the second subject in the second medical information and the first index;
A medical information processing method, wherein a display unit displays the consistency information. On the computer,
a collection unit collects first medical information of a first subject based on a diagnostic purpose for the first subject, and collects second medical information of at least one second subject related to a similar case of the first subject according to the diagnostic purpose;
A quantification unit determines a first index in the first medical information for ranking the importance of a basis for a first inference in a first diagnosis for the first subject based on a diagnostic purpose for the first subject;
a consistency evaluation unit generates consistency information regarding an evaluation of the consistency between the first inference and the second inference based on a second index for ranking the importance of a basis for the second inference in the second diagnosis for the second subject in the second medical information and the first index;
a display unit displays the consistency information;
A medical information processing program that makes this possible.

Images