Medical information processing system and method

Abstract

To output, even when a possibility related to a specific disease largely changes in a short time, a determination result on which the changed possibility is reflected.SOLUTION: A medical information processing system comprises a storage unit, a receiving unit, an output unit, an acquisition unit, and an update unit. The storage unit stores a determination model for determining a possibility related to a specific disease. The receiving unit receives first subject information based on the history of the first subject. The output unit outputs first information on the first subject related to a disease. The acquisition unit acquires second subject information based on the history of a second subject. The update unit updates the determination model based on the second subject information. The output unit outputs second information including a possibility of the first subject based on the updated determination model and the first subject information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] There is known a medical information processing system that aggregates information on various test results and diagnostic results of many other subjects to determine the possibility of a subject having a specific disease, such as the diagnosis, treatment method, prognosis, and risk of the disease for the subject.

[0003] Such medical information processing systems include, for example, technologies for estimating future disease risk based on health checkup results, genes, age, and disease history. This technology estimates disease risk by estimating risk from easily measurable items for items that are difficult to test frequently. The underlying technology involves clustering the population distribution to obtain a disease distribution in advance as a prior probability distribution, and then estimating a posterior probability distribution under actual test results using Bayes' theorem. This method estimates a prior probability distribution based on factor data from a group of subjects, and then calculates the posterior probability under the conditions under which the test results are obtained based on the prior probability distribution. Factors that remain stable over time are input when collecting factor data, and a prior distribution of susceptibility is calculated based on these factors. This is because, when collecting factor data from a group of subjects to calculate a prior distribution of susceptibility, if the susceptibility of each factor changes during the period in which the group data is collected, it becomes impossible to reliably estimate susceptibility.

[0004] On the other hand, with epidemic diseases, the risk of contracting a disease can fluctuate significantly over a few days depending on the season and circumstances. In some cases, the behavior of the subject is involved in determining whether or not the disease is present and in determining post-test treatment. The subject's behavior can be investigated and tracked during the interview at the time of testing. However, among the behaviors investigated, high-risk behaviors may not be identified on the day of testing, but may become apparent several days after testing.

[0005] For example, suppose an initial test detects O-157 using a rapid antigen test and comes back negative, albeit just barely. Then, during a recent dietary interview, the patient mentions eating a salad containing radish sprouts a few days prior. Now, imagine a situation where, a few days after the test, radish sprouts at a specific facility emerge as the source of O-157 infection. Considering a general epidemic, in such a case, depending on the possible cause, the patient must be contacted and asked to return for testing. In such cases, the doctor in charge typically contacts the patient and urges them to retest. However, if there are a large number of people being tested or if information about local risk behaviors, such as food intake, is not widely publicized, the doctor in charge may be unable to determine the necessary treatment and quickly halt the progression of the patient's disease.

[0006] In this way, when the risk of contracting a disease changes significantly in a short period of time, it may not be possible to change the treatment of the subject quickly, and appropriate treatment may not be available for the patient. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Laid-Open No. 2004-305674 Summary of the Invention [Problem to be solved by the invention]

[0008] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to output a determination result that reflects the changed possibility, even when the possibility of a specific disease changes significantly over a short period of time. However, the problem to be solved by the embodiments disclosed in this specification and the drawings is not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described below can also be positioned as other problems. [Means for solving the problem]

[0009] A medical information processing system according to an embodiment includes a storage unit, a reception unit, an output unit, an acquisition unit, and an update unit. The storage unit stores a judgment model for determining the possibility of a specific disease. The reception unit receives first subject information based on the history of the first subject. The output unit outputs first information about the first subject related to the disease based on the judgment model and the first subject information. The acquisition unit acquires second subject information based on the history of the second subject. The update unit updates the judgment model based on the second subject information. The output unit outputs second information including the possibility of the first subject based on the updated judgment model and the first subject information. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing an example of the configuration of a medical information processing system according to the first embodiment. [Figure 2] FIG. 2 is a schematic diagram showing a data flow in the medical information processing system according to the first embodiment. [Figure 3] FIG. 3 is a diagram showing an example of a medical questionnaire used in the medical information processing system according to the first embodiment. [Figure 4] FIG. 4 is a diagram showing an example of a record used in the medical information processing system according to the first embodiment. [Figure 5] FIG. 5 is a flowchart illustrating a processing procedure of medical decision support processing by the medical information processing system according to the first embodiment. [Figure 6] FIG. 6 is a diagram showing an example of a data management screen output by the medical decision support process performed by the medical information processing system according to the first embodiment. [Figure 7] FIG. 7 is a flowchart illustrating the processing procedure of the determination processing by the medical information processing system according to the first embodiment. [Figure 8] FIG. 8 is a diagram showing an example of the configuration of a medical information processing system according to the second embodiment. [Figure 9]FIG. 9 is a flowchart illustrating a processing procedure of a determination process by a medical information processing system according to a modified example of the second embodiment. [Figure 10] FIG. 10 is a diagram showing an example of a data management screen output by the medical decision support process performed by the medical information processing system according to the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, an embodiment of a medical image processing apparatus will be described in detail with reference to the drawings. In the following description, components having substantially the same functions and configurations are designated by the same reference numerals, and redundant description will be given only when necessary.

[0012] (First embodiment) 1 is a diagram showing the configuration of a medical information processing system 1 of this embodiment. The medical information processing system 1 includes a determination device 10. The determination device 10 is connected to a medical information system 30 and an examination database 40 via a network 20. Here, the determination device 10, the medical information system 30, and the examination database 40 are described as being installed in the same facility, but the determination device 10 may be installed in a facility separate from the medical information system 30 and the examination database 40.

[0013] The network 20 is, for example, a LAN (Local Area Network), and the connection to the network 20 may be either a wired connection or a wireless connection. Also, as long as security is ensured by a VPN (Virtual Private Network) or the like, the connection line is not limited to a LAN. It may also be possible to connect to a public communication line such as the Internet.

[0014] The medical information system 30 manages information related to medical facilities such as hospitals. The medical information system 30 is, for example, a Hospital Information System (HIS). In the medical information system 30, the subject's electronic medical record, information on various tests, examination results, etc. are recorded in a storage device. In the examination results, the name of the suspected disease is recorded as an item of "suspected diagnosis."

[0015] The examination database 40 includes a storage device that stores multiple records related to patients (hereinafter referred to as subjects) suspected of having a specific disease (hereinafter referred to as a target disease). Each record is generated for each subject. For example, a record is generated each time a subject undergoes an examination or consultation at a medical facility. For this reason, multiple records may be recorded for one subject. The records store medical information about the subject. The medical information includes, for example, the name, patient ID, information about the examination facility, background factors, behavioral history, interview results, examination information, and diagnosis information. For example, if the target disease is "pneumonia," the examination database 40 may be called a pneumonia examination database.

[0016] The determination device 10 can transmit and receive various information to and from the medical information system 30 and the examination database 40 via the network 20. The determination device 10 acquires information (hereinafter referred to as history information) related to the history of the subject to be determined (hereinafter referred to as the subject to be determined), determines information related to the possibility of the subject to be determined having a specific disease based on the acquired history information, and outputs the determination result. In this embodiment, the behavioral history or interview results of the subject to be determined are used as the history information. Furthermore, in this embodiment, information related to the possibility of the subject to be determined having a specific disease is referred to as risk information.

[0017] The target disease is, for example, pneumonia. The target disease may also be various diseases such as food poisoning caused by Escherichia coli, Creutzfeldt-Jakob disease caused by ingesting specified risk parts of beef within a certain period of time, etc.

[0018] The risk information includes, for example, health status, the presence or absence of a target disease, and the degree of possibility of having the target disease. The risk information may also include the selection of a treatment method, a prognosis prediction, a prediction of future disease risk (risk of onset), and a prediction of drug effectiveness. When the risk information includes the degree of possibility of having the target disease, the determination device 10 estimates the probability that the subject has the target disease (hereinafter referred to as the disease probability) based on the test results and behavioral history of the subject, and determines whether the subject has a "possible disease," "low possibility of disease," or "high possibility of disease" based on the estimated disease probability. For example, "low possibility of disease" indicates that the possibility of disease is less than a predetermined value. For example, "high possibility of disease" indicates that the possibility of disease is equal to or greater than a predetermined value. For example, "possible disease" indicates that the degree of possibility of disease is unknown. The disease probability may also be referred to as disease probability.

[0019] In the present embodiment, an example will be described below in which the target disease is "pneumonia" and the risk information is the degree of possibility that the person being assessed has pneumonia. Fig. 2 is a schematic diagram showing a data flow in the medical information processing system 1 of the present embodiment.

[0020] The determination device 10 includes a memory 11, a communication interface 12, a display 13, an input interface 14, and a processing circuit 15. Although the determination device 10 will be described below as a single device that executes multiple functions, the multiple functions may be executed by separate devices. For example, the functions executed by the determination device 10 may be distributed and installed on different console devices or workstation devices.

[0021] The memory 11 is a storage device such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or an integrated circuit that stores various information. The memory 11 may also be a portable storage medium such as a CD (Compact Disc), a DVD (Digital Versatile Disc), or a flash memory, in addition to an HDD or SSD. The memory 11 may also be a drive device that reads and writes various information from and to semiconductor memory elements such as flash memory and RAM (Random Access Memory). The storage area of ​​the memory 11 may be located within the determination device 10, or may be located in an external storage device connected via a network.

[0022] The memory 11 stores a determination model for at least one disease. The memory 11 also stores a program executed by the processing circuit 15, various data used in the processing of the processing circuit 15, and the like. As the program, for example, a program that is installed in advance on a computer from a network or a non-transitory computer-readable storage medium and causes the computer to realize each function of the processing circuit 15 is used. Note that the various data handled in this specification are typically digital data. The memory 11 is an example of a storage unit.

[0023] The determination model estimates the probability of a subject having a target disease based on the subject's behavioral history or interview results, and determines the possibility that the subject has a target disease. The determination model uses a known classification algorithm such as linear discriminant analysis.

[0024] The communication interface 12 is a network interface that controls transmission of communications with the medical information system 30 and other external devices via the network 20.

[0025] The display 13 displays various types of information. For example, the display 13 outputs medical information generated by the processing circuitry 15, a GUI (Graphical User Interface) for receiving various operations from an operator, and the like. For example, the display 13 is a liquid crystal display or a CRT (Cathode Ray Tube) display. The display 13 also displays a data management screen, which will be described later. The display 13 is an example of a display unit.

[0026] The input interface 14 accepts various input operations from the operator, converts the accepted input operations into electrical signals, and outputs them to the processing circuitry 15. For example, the input interface 14 accepts input of medical information, input of various command signals, etc. from the operator. The input interface 14 is realized by a mouse, keyboard, trackball, switch buttons, a touch screen integrating a display screen and a touchpad, a non-contact input circuit using an optical sensor, a voice input circuit, etc., for performing various processes in the processing circuitry 15. The input interface 14 is connected to the processing circuitry 15 and converts input operations received from the operator into electrical signals and outputs them to the control circuit. Note that, in this specification, the input interface is not limited to those equipped with physical operating components such as a mouse and keyboard. For example, an electrical signal processing circuit that receives electrical signals corresponding to input operations from an external input device provided separately from the device and outputs the electrical signals to the processing circuitry 15 is also an example of an input interface. The input interface 14 is an example of an input unit.

[0027] The processing circuitry 15 controls the overall operation of the determination device 10. The processing circuitry 15 is a processor that executes a record generation function 151, a reception function 152, a determination function 153, an acquisition function 154, an update function 155, a comparison function 156, and a display control function 157 by calling and executing programs in the memory 11.

[0028] 1 illustrates the record generation function 151, reception function 152, determination function 153, acquisition function 154, update function 155, comparison function 156, and display control function 157 implemented by a single processing circuit 15. However, a processing circuit may be configured by combining multiple independent processors, and each processor may execute a program to implement each function. Furthermore, the record generation function 151, reception function 152, determination function 153, acquisition function 154, update function 155, comparison function 156, and display control function 157 may be referred to as a record generation circuit, reception circuit, determination circuit, acquisition circuit, update circuit, comparison circuit, and display control circuit, respectively, or may be implemented as separate hardware circuits. The above description of the functions executed by the processing circuit 15 also applies to the following embodiments and modifications.

[0029] The term "processor" used in the above description refers to a circuit such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an ASIC, a programmable logic device (e.g., a Simple Programmable Logic Device (SPLD)), a Complex Programmable Logic Device (CPLD), or a Field Programmable Gate Array (FPGA). The processor realizes its function by reading and executing a program stored in the memory 11. Note that instead of storing the program in the memory 11, the program may be configured to be directly embedded in the circuit of the processor. In this case, the processor realizes its function by reading and executing the program embedded in the circuit. Note that each processor in this embodiment is not limited to being configured as a single circuit, but may be configured as a single processor by combining multiple independent circuits to realize its function. Furthermore, the multiple components in FIG. 1 may be integrated into a single processor to realize its function. The above description of the "processor" also applies to the following embodiments and modified examples.

[0030] The processing circuitry 15 acquires medical information about the subject using the record generation function 151 and generates a record of the subject based on the acquired medical information. The generated record is output to, for example, the examination database 40. The processing circuitry 15 that realizes the record generation function 151 is an example of a report generation unit. The medical information includes the name, patient ID, information about the examination facility, background factors, behavioral history, examination information, diagnostic information, and assessment results. The medical information is acquired, for example, from the medical information system 30. Alternatively, the medical information is acquired by the subject marking or filling out a medical questionnaire printed on paper and scanning the completed questionnaire. The medical information may also be acquired from medical questionnaire results obtained from a diet management app, an exercise management app, a health management app, or the like. FIG. 3 is a diagram showing an example of a medical questionnaire used to acquire the behavioral history and medical questionnaire results of the subject. FIG. 4 is a diagram showing an example of a generated record.

[0031] Background factors are information about factors whose prior probability in the population changes little over time. Background factors include, for example, the subject's age, gender, health condition, medical history, and family medical history. Health conditions include, for example, information about exercise habits, living environment, smoking / drinking habits, and subjective symptoms.

[0032] Behavioral history is information about behaviors that affect the risk of contracting a target disease (hereinafter referred to as risk behaviors). Behavioral history is information about factor items whose prior probability changes significantly over time. Behavioral history includes place of residence, type of residence, dietary history, visit history, stay history, etc. Dietary history includes the type of food eaten, ingredients used in the meal, the date and time the meal was eaten, the name of a product consumed, the date and time the food was consumed, etc. Visit history includes the name of a facility visited, the date and time of the visit, etc. Stay history includes the name of the area stayed, the time of stay, the duration of stay, etc. In a medical questionnaire, for example, each item in the behavioral history is entered with a code indicating the content of the risk behavior and the date and time of the risk behavior. The date and time of the behavior includes the date and time the risk behavior was performed. In this case, a code table is attached to the medical questionnaire. Alternatively, the medical questionnaire may include an item asking whether or not a specific behavior was performed during a specified period, for example.

[0033] The test information is information about the results of various tests used by doctors when examining and diagnosing subjects. The test information is obtained, for example, from the medical information system 30. The test information includes the name of the test, the test results, the test date and time, etc. For example, the test results include body temperature, heart rate, weight, blood pressure, etc., along with the measurement date and time. For example, the test results of a blood test include a number of items such as white blood cell count and CPR (C-peptide) value. For example, the test results of a rapid antigen test or rapid antibody test include detection results of bacteria such as O-157 and other viruses. Furthermore, the test results of imaging tests using CT (Computed Tomography), MRI (Magnetic Resonance Imaging), ultrasound diagnostic equipment, etc. include information about imaging diagnostic findings and measurement values ​​such as cardiac output, left ventricular ejection fraction, and the volume ratio of pneumonia findings to the whole lung.

[0034] The diagnostic information includes information regarding the estimated result of the possibility that the subject has the target disease. The diagnostic information is obtained, for example, from the medical information system 30. The diagnostic information is information indicating, for example, "diagnosed with pneumonia," "diagnosed not with pneumonia," or "no determination as to whether or not pneumonia is present." A diagnosis is not determined based on the results of a single test, but is determined comprehensively based on multiple tests, patient background, and interview findings. Therefore, a definitive diagnosis is made several days to several weeks after the initial consultation. The result of the diagnosis determined at the initial consultation (hereinafter referred to as the initial diagnosis result) is recorded, for example, as a disease name or natural language in the "suspected diagnosis" field of the medical information system 30. The initial diagnosis result is also recorded in the "diagnosis" field of the record. The definitive diagnosis result is recorded, for example, as a disease name or natural language in the "definitive diagnosis" field of the medical information system 30. The definitive diagnosis result is also recorded, for example, in the "definitive diagnosis of suspected pneumonia" field of the record. The initial diagnosis result may also be referred to as a diagnostic finding.

[0035] If the result of the definitive diagnosis is not recorded in the medical information system 30, the processing circuitry 15 may determine the definitive diagnosis based on the subject's medical information. As an example, a case will be described in which a definitive diagnosis of pneumonia is determined based on the test results of a pathological test, the test results of a rapid antibody test, and the disease name of the initial diagnosis. First, if the test results of the pathological test indicate the presence or absence of pneumonia, the processing circuitry 15 determines the test results of the pathological test as the definitive diagnosis of pneumonia. If the test results of the pathological test indicate the presence or absence of pneumonia, or if a pathological test has not been performed, the processing circuitry 15 determines the definitive diagnosis of pneumonia based on the test results of the rapid antibody test and the disease name of the initial diagnosis. In this case, if the virus "a" is detected as the test result of the rapid antibody test, the processing circuitry 15 determines the definitive diagnosis of pneumonia as "pneumonia present." If the test results of the rapid antibody test indicate the virus "b" and the disease name of the provisional diagnosis is some type of "pneumonia," the processing circuitry 15 determines the definitive diagnosis of pneumonia as "pneumonia present." If the rapid antibody test results are other than those described above, the processing circuit 15 does not determine a definitive diagnosis of pneumonia.

[0036] The judgment result includes the judgment result based on the judgment model. The judgment result is recorded in the "Estimated probability of pneumonia" field of the record. For example, "Possible pneumonia" or "Low possibility of pneumonia" is recorded in the record. For example, if no judgment has been made on the person to be judged, the judgment result is not recorded in the "Estimated probability of pneumonia" field. Furthermore, if judgments have been made multiple times using the judgment model, the judgment result and the date and time of the judgment are recorded for each of the multiple judgments in the "Estimated probability of pneumonia" field.

[0037] The processing circuit 15 receives information about the subject to be assessed using the reception function 152. Specifically, the processing circuit 15 extracts records of subjects for which a definitive diagnosis of the target disease has not been recorded from all records stored in the memory 11, and acquires the information recorded in the extracted records as information about the subject to be assessed. The subject to be assessed is an example of a first subject. The information about the subject to be assessed is an example of first subject information. The processing circuit 15 that realizes the reception function 152 is an example of a reception unit.

[0038] Furthermore, the processing circuit 15 acquires attribute values ​​based on the behavioral history of the person being assessed. The attribute values ​​are variables that are set based on various behavioral items, various test results, health status values, etc. Each subject is represented by multidimensional data that combines attribute values ​​(variables) for the number of behavioral items and attribute values ​​(variables) for the number of other items. Some of these dimensions have discrete values. For example, since behavioral items are date and time values, the behavioral items are treated as continuous values. Also, for example, the gender item is treated as a discrete value.

[0039] The processing circuit 15 uses the judgment function 153 to output a judgment result regarding an estimate of the probability of affliction with the target disease based on the judgment model and information about the person to be judged. The processing circuit 15 outputs the judgment result obtained from the judgment model to the examination database 40, the display 13, a printing device connected to the judgment device 10, etc. The judgment result is an example of first information. The judgment result may also be called first risk information. The processing circuit 15 that realizes the judgment function 153 is an example of an output unit.

[0040] For example, the processing circuit 15 determines whether the attribute value of the subject falls within the range of attribute values ​​of the "low probability of disease" cluster or the range of attribute values ​​of the "possible disease" cluster, based on the attribute value of the subject. The "low probability of disease" cluster is a group of subjects whose probability of pneumonia is less than a predetermined value. The "possible disease" cluster is a group of subjects whose probability of pneumonia is equal to or greater than a predetermined value.

[0041] For example, if the value of the variable of the action item or the value of another item variable is included in the range of the "possible disease" cluster, the processing circuit 15 determines that the condition is "possible disease." In this case, for example, "possible pneumonia" is recorded in the "estimated probability of pneumonia" field of the record. Also, if the value of the variable of the action item or the value of another item variable is included in the range of the "low probability of disease" cluster, the processing circuit 15 determines that the condition is "low probability of disease." In this case, for example, "low possibility of pneumonia" is recorded in the "estimated probability of pneumonia" field of the record.

[0042] The processing circuit 15 acquires information about a subject (hereinafter referred to as a confirmed diagnosis subject) for whom a definitive diagnosis of a target disease has been recorded, using the acquisition function 154. Specifically, the processing circuit 15 extracts records of the confirmed diagnosis subject from all records stored in the memory 11, and acquires information about the confirmed diagnosis subject from the extracted records. The information about the confirmed diagnosis subject includes at least one of the behavioral history and interview results of the confirmed diagnosis subject. The information about the confirmed diagnosis subject also includes a definitive diagnosis result of the confirmed diagnosis subject. The confirmed diagnosis subject is an example of a second subject. The information about the confirmed diagnosis subject is an example of second subject information. The definitive diagnosis result of the confirmed diagnosis subject is an example of diagnostic information of the second subject. The processing circuit 15 realizing the acquisition function 154 is an example of an acquisition unit. The processing circuit 15 also acquires attribute values ​​based on the behavioral history of the confirmed diagnosis subject.

[0043] The processing circuitry 15 updates the determination model based on information about the diagnosed individual and the diagnosis information of the diagnosed individual using the update function 155. At this time, the processing circuitry 15 updates the determination model using at least one of the behavioral history and interview results of the diagnosed individual and the definitive diagnosis result of the diagnosed individual. The processing circuitry 15 that realizes the update function 155 is an example of an update unit.

[0044] New records for many subjects are generated in the examination database 40, and various behavioral items and health conditions of many subjects are recorded daily. In addition, confirmed diagnoses are added to the examination database 40 daily. The processing circuit 15 updates the conditions (hereinafter referred to as judgment conditions) for making judgments regarding the estimation of the probability of morbidity of the target disease by performing the classification and aggregation process described below on all confirmed diagnoses, including the newly added confirmed diagnoses, using the update function 155. The judgment model is updated by updating the judgment conditions.

[0045] Specifically, the processing circuit 15 first classifies the diagnosed individuals into multiple clusters based on their behavioral items and behavioral dates and times. At this time, the processing circuit 15 calculates the disease probability for each of the multiple clusters and classifies the clusters so that the difference in disease probability between the multiple clusters is large.

[0046] The processing circuitry 15 outputs an updated judgment result based on the updated judgment model and information about the person to be judged using the judgment function 153. The processing circuitry 15 performs a judgment again using the judgment model, updates the judgment result based on the re-judgment result of the judgment model, and outputs the updated judgment result to the inspection database 40, the display 13, a printing device connected to the judgment device 10, etc. The updated judgment result is an example of second information. The updated judgment result may also be called second risk information.

[0047] Specifically, the processing circuit 15 determines which of a plurality of clusters the subject of judgment belongs to based on the behavior date and time of the subject of judgment, and outputs a determination result regarding an estimate of the probability of morbidity based on the disease probability of that cluster.

[0048] The processing circuitry 15 uses the comparison function 156 to compare the updated judgment result with the judgment result before the update for each subject to be judged, and outputs the comparison result for subjects whose updated judgment result differs from their pre-update judgment result. For example, if the judgment result has changed, the processing circuitry 15 outputs information indicating that the judgment result has changed to the display 13 or a printing device connected via the network 20, etc. In this case, the processing circuitry 15 may create a list extracting only subjects whose judgment results have changed, and output the list to the display 13 or the printing device. The processing circuitry 15 that realizes the comparison function 156 is an example of a comparison unit.

[0049] The processing circuitry 15 uses the display control function 157 to display a GUI (hereinafter referred to as a data management screen) for managing data stored in the examination database 40 on the display 13. The processing circuitry 15 that realizes the display control function 157 is an example of a display control unit.

[0050] Next, we will explain the operation of the medical decision support processing executed by the determination device 10. The medical decision support processing is processing that collects behavioral histories and definitive diagnosis results for a target disease for multiple subjects, estimates the probability of the subject having the target disease based on the collected results, makes a determination regarding the estimated probability of having the disease, and, if the determination result has changed from the previous determination result, outputs information indicating that the determination result has changed.

[0051] The following describes a case where the processing of each step of the medical decision support processing is executed in response to a user's instruction input via the input interface 14. FIG. 5 is a flowchart showing an example of the procedure of the medical decision support processing. Note that the processing procedure for each process described below is merely an example, and each process can be modified as appropriate as possible. Furthermore, steps can be omitted, replaced, or added to the processing procedures described below as appropriate depending on the embodiment.

[0052] The processing of each step of the medical decision support processing may be automatically executed at regular intervals. The medical decision support processing may be executed, for example, once a day at a predetermined time late at night. The medical decision support processing may also be executed once a week (every seven days). The medical decision support processing may also be executed every time a new subject record is generated. The medical decision support processing may also be called batch processing.

[0053] (Medical Decision Support Processing) (Step S101) The processing circuitry 15 uses the display control function 157 to display a data management screen 50 on the display 13 based on the records stored in the memory 11. Fig. 6 is a diagram showing an example of the data management screen 50. The data management screen 50 has a data display section 51, a period setting section 52, a record creation instruction input section 53, a data collection instruction input section 54, and an update instruction input section 55.

[0054] The data display unit 51 displays information about the records stored in the memory 11 of the determination device 10 in a list for each subject. The data display unit 51 displays, for example, the name, patient ID, information about the testing facility, background factors, behavioral history, test information, diagnostic information, and determination results. The test information display field displays, for example, the test results of a rapid antibody test. The diagnostic information display field displays, for example, the initial diagnostic results. The test results of the rapid antibody test and the initial diagnostic results are preferably displayed because they are used for a definitive diagnosis.

[0055] In the period setting unit 52, an instruction to set the records to be displayed in the data display unit 51 is input. For example, by setting "one month" in the period setting unit 52, only records created in the most recent month are displayed in the data display unit 51.

[0056] An instruction to create a new record is input in the record creation instruction input unit 53. An instruction to acquire test information and diagnosis information is input in the data collection instruction input unit 54. An instruction to update the assessment result regarding the target disease of the subject is input in the update instruction input unit 55.

[0057] (Step S102) When an operation is input in the record creation instruction input unit 53, the processing circuit 15 causes the record creation function 151 to scan the medical questionnaire and create a new record using the scanned data.

[0058] (Step S103) When an operation is input to the data collection instruction input unit 54, the processing circuit 15 obtains the electronic medical record of the subject displayed on the data display unit 51 from the medical information system 30, and collects the test information, behavioral history, and diagnostic information of the subject for whom "pneumonia" is recorded in the "suspected diagnosis" field. This allows new attributes such as test information and diagnostic information that have not previously been recorded in the record to be obtained. Only the extracted records are displayed on the data display unit 51. The example in Figure 6 shows that, among records created in the last month, there are four subjects for whom "pneumonia" is recorded in the "suspected diagnosis" field of their electronic medical records.

[0059] (Step S104) When an operation is input in the update instruction input unit 55, the processing circuitry 15 extracts, from the records displayed on the data display unit 51, subjects for whom a definitive diagnosis does not exist, as subjects to be assessed. Then, the processing circuitry 15 acquires a result of assessment by executing a process for making a judgment regarding an estimate of the probability of morbidity of the target disease (hereinafter referred to as the assessment process) using the reception function 152, the assessment function 153, the acquisition function 154, and the update function 155. The assessment result is stored in the record. The details of the assessment process will be described later.

[0060] (Step S105) The processing circuit 15 uses the comparison function 156 to compare the last two judgment results stored in the record of the person being judged. If the judgment results differ, the processing circuit 15 extracts the changes in the judgment results and displays them on the data display unit 51. In the example of FIG. 6, the judgment result display field on the data display unit 51 displays the judgment result before the update, the judgment result after the update, and a comparison result between the judgment result before the update and the judgment result after the update. If only one judgment result is stored in the record, the data display unit 51 does not display the judgment result before the update, and only displays the judgment result after the update. If two or more judgment results are stored in the record, the data display unit 51 displays the previous judgment result as the judgment result before the update, and the latest judgment result as the judgment result after the update. If the previous judgment result and the latest judgment result differ, the data display unit 51 displays the changes in the judgment results or the latest judgment result as the comparison result. In addition, in the data display unit 51, the judgment results for subjects with a definitive diagnosis are displayed as the judgment results before the update and the latest judgment results after the update, and the comparison results are displayed with a symbol or the like indicating that the subject is not subject to comparison.

[0061] Next, a detailed description will be given of the operation of the determination process executed by the determination device 10. Fig. 7 is a flowchart showing an example of the procedure of the determination process.

[0062] (Determination process) (Step S111) The processing circuit 15 uses the acquisition function 154 to refer to the examination database 40 and acquire the records of all subjects who have been definitively diagnosed with pneumonia as records of confirmed diagnoses. Next, the processing circuit 15 acquires the attribute values ​​of each confirmed diagnose and the definitive diagnosis result of each confirmed diagnose based on the records of the confirmed diagnoses. At this time, the processing circuit 15 acquires the attribute values ​​of only action items that have an action date and time within the expiration date and time. Here, either "with disease" or "without disease" is acquired as the definitive diagnosis result.

[0063] (Step S112) Next, the processing circuit 15 updates the judgment model using the update function 155. In this case, the processing circuit 15 first applies various classification methods to the attribute values ​​of each diagnosed individual to classify each diagnosed individual into either a "low probability of disease" cluster or a "possible disease" cluster. In this case, various variable selection methods may be used to narrow down the variables used for classification to a small number. Examples of variable selection methods include variable elimination and variable addition. Other variable selection methods may also be used. Furthermore, if the number of variables is not enormous, it is preferable to use the exhaustive classification method from the standpoint of accuracy.

[0064] Furthermore, the processing circuitry 15 divides the attribute value ranges so that the ratio of the number of subjects in the group of subjects whose definitive diagnosis is "disease-present" (hereinafter referred to as the disease-present group) to the number of subjects in the group of subjects whose definitive diagnosis is "disease-free" (hereinafter referred to as the disease-free group) is as different as possible between the "low probability of disease" cluster and the "possible disease" cluster. That is, the processing circuitry 15 determines conditions (hereinafter referred to as classification conditions) for classifying diagnosed subjects based on attribute values ​​so that subjects whose definitive diagnosis is "disease-present" are classified into the "possible disease" cluster and subjects whose definitive diagnosis is "disease-free" are classified into the "low probability of disease" cluster.

[0065] When linear discriminant analysis is used as a classification method (clustering method), the division is performed so as to maximize the Mahalanobis distance between clusters. Many clustering methods determine the division surface based on some distance between two groups. For example, in a support vector machine, the division surface is determined so as to maximize the distance between the group and the division surface. The distance is the distance in attribute value space. Another classification method is the decision tree method. In the decision tree method, classification is performed so as to minimize the impurity of the data when it is divided.

[0066] Alternatively, a classification method (clustering method) such as that described in the unpublished prior patent application, Patent Application No. 2020-039545, may be used. This prior application describes a method for determining clusters so that the upper limit of the confidence interval for the disease probability is minimized or the lower limit of the confidence interval for the disease probability is maximized, in order to separate the disease probabilities between clusters as much as possible. This method is effective for problems with poorly separated clusters. For example, if the disease probability of a "probable disease" cluster is much lower than 100% (e.g., about 5%), but sufficient data indicates that it is higher than the disease probability of a "low disease probability" cluster (e.g., 0.1%), the range of the cluster can be determined. The range of the cluster includes the date and time range. It is difficult to classify such poorly separated clusters using distance-based or impurity-based partitioning methods.

[0067] As a hypothetical example, the following classification results are presented for the behavioral item of eating raw venison in Okayama Prefecture. Among subjects whose behavioral dates and times were between January 10 and January 14, 1997, a large number of subjects were diagnosed with a definitive "disease," while among subjects whose behavioral dates and times were outside of the above range, a small number were diagnosed with a definitive "disease." Subjects whose rapid antigen test values ​​were above a certain value were classified into the "possible disease" cluster, regardless of the behavioral item of eating raw venison in Okayama Prefecture. Furthermore, even if the rapid antigen test values ​​were below a certain value, subjects who ate raw venison between January 10 and January 14, 1997, were classified into the "possible disease" cluster. On the other hand, subjects whose rapid antigen test values ​​were below a certain value and who did not eat raw venison between January 10 and January 14, 1997, were classified into the "low probability of disease" cluster.

[0068] The processing circuit 15 performs a counting process for each cluster based on the classification results of the confirmed diagnoses, and calculates the disease probability for each cluster based on the counting results. The disease probability for each cluster is expressed by the following formula. Probability of disease = number of tests in the diseased group / (number of tests in the non-disease group + number of tests in the diseased group) Here, the number of disease-positive tests is the number of subjects in the disease-positive group among the subjects in this cluster. The number of disease-free tests is the number of subjects in the disease-free group among the subjects in this cluster. In other words, the disease probability is the ratio of the number of subjects in the disease-positive group among the subjects in this cluster to the total number of subjects in that cluster. The processing circuit 15 classifies each confirmed diagnosis so that the disease probabilities between clusters are as distinct as possible. In other words, the processing circuit 15 determines the classification criteria for subjects based on the disease probabilities of each cluster so that the difference in disease probabilities between clusters is as large as possible.

[0069] It is preferable that the processing circuit 15 uses only the attribute values ​​of action items within the expiration date for classification. Each action item has an expiration date. The expiration date is the period during which the action item affects the assessment of the subject. For example, an action item indicating the date and time of eating "shrimp (raw)" is set to 10 days as the expiration date. In this case, the date 10 days before the assessment date is set as the expiration date. An action item indicating that shrimp was eaten before the expiration date does not provide information useful for assessing the subject to be assessed. For this reason, the processing circuit 15 does not use the attribute values ​​of action items having an action date and time before the expiration date for classification, but only uses the attribute values ​​of action items having an action date and time between the expiration date and the assessment date for classification.

[0070] Furthermore, for action items for which the number of subjects used for classification is below a certain number, classification may be performed without using those action items. For example, if the number of subjects whose action dates and times are within the expiration date for a certain action item is below a certain number, the attribute values ​​of that action item are not used to classify confirmed diagnoses. In this case, by reducing the number of items used for classification, it is possible to prevent the number of action items for subjects used for classification from becoming enormously diverse, and it is possible to perform the classification process within a practical amount of time.

[0071] (Step S113) The processing circuit 15 acquires the record of the person to be judged by the reception function 152. Next, the processing circuit 15 acquires each attribute value of the person to be judged based on the record of the person to be judged.

[0072] (Step S114) The processing circuit 15 uses the determination function 153 to determine whether each subject of assessment is included in the "low probability of disease" cluster or the "possible disease" cluster based on the attribute values ​​of the subject of assessment and the classification conditions of the cluster. For a subject of assessment whose action item variables and other variable values ​​are included in the classification conditions of the "possible disease" cluster, the processing circuit 15 sets the disease probability of the "possible disease" cluster as the subject of assessment's morbidity probability, and determines the estimated result of the pneumonia morbidity probability as "possible disease." For a subject of assessment whose action item variables and other variable values ​​are included in the conditions of the "low probability of disease" cluster, the processing circuit 15 sets the disease probability of the "low probability of disease" cluster as the subject of assessment's morbidity probability, and determines the estimated result of the pneumonia morbidity probability as "low probability of disease."

[0073] As described above, by executing the processes of steps S101 to S105, it is possible to obtain the judgment result of the judgment process for a subject who has been judged for the first time. For a subject who has been judged for the second or subsequent time by the judgment process, the latest judgment result can be obtained using a judgment model that reflects the diagnostic results of confirmed diagnoses added from the date of the previous judgment to the current judgment. If the current judgment result has changed from the previous judgment result, for example, a data management screen 50 displaying the change in the judgment result is displayed on the display 13. The data management screen 50 may be output to a printing device connected via the network 20 or the like and printed on paper by the printing device. Alternatively, a list may be created that extracts only subjects whose judgment results have changed and output to the display 13 or the printing device.

[0074] The effects of the medical information processing system 1 having the determination device 10 according to this embodiment will be described below.

[0075] New records for many subjects are created in the test database 40, and various behavioral items and health conditions of many subjects are recorded daily. In addition, confirmed diagnoses are added to the test database 40 daily. In a judgment based on the initial test values, only risk factors known at that time are taken into consideration. The behavioral history of the subject himself / herself is involved in determining whether or not a disease is present and the post-test treatment. In addition, the relationship between a subject's behavior and the risk of contracting the disease may become clear later. For this reason, behavioral items that affect the probability of contracting the disease may change over a short period of time. For example, it may be discovered that people who behaved in a similar manner during a specific period of time have the same disease.

[0076] The medical information processing system 1 according to this embodiment accepts subject information based on the historical information of the subject, and outputs information regarding the possibility of a specific disease for the subject based on a determination model for a specific disease and the subject information of the subject. The information regarding the possibility of a specific disease may be referred to as risk information. The historical information includes at least one of a behavioral history and a medical interview result. The medical information processing system 1 also acquires subject information based on at least one of a behavioral history and a medical interview result of a confirmed diagnosis, and updates the determination model based on the subject information of the confirmed diagnosis. The medical information processing system 1 then outputs updated information about the subject based on the updated determination model and the subject information of the subject.

[0077] In this embodiment, the probability of having a specific disease is used as information regarding the possibility of a specific disease. Furthermore, the subject information of the confirmed diagnosis subject includes the confirmed diagnosis of the specific disease by the confirmed diagnosis subject. The medical information processing system 1 updates the determination model based on the confirmed diagnosis by the confirmed diagnosis subject.

[0078] Furthermore, the medical information processing system 1 according to this embodiment acquires the behavioral items and behavioral dates and times of the diagnosed individual based on the subject information of the diagnosed individual, and classifies the diagnosed individual into multiple clusters based on the behavioral items and behavioral dates and times of the diagnosed individual. In this case, the processing circuit 15 calculates the disease probability for each of the multiple clusters and classifies the multiple clusters so that the difference in disease probability between the multiple clusters is large. The medical information processing system 1 then determines which of the multiple clusters the subject belongs to based on the behavioral date and time of the subject, and outputs a determination result regarding the estimated probability of illness based on the disease probability of that cluster.

[0079] With the above configuration, the medical information processing system 1 according to this embodiment updates the judgment model using the behavioral items and behavioral dates and times of all diagnosed subjects, including subjects with newly added definitive diagnoses. This allows the judgment model to reflect the diagnostic results of subjects who have been definitively diagnosed between the date of the previous judgment and the date of the current judgment. Then, by using the updated judgment model to judge the risk of disease, it is possible to obtain judgment results that reflect changes in local prevalence and changes in risk factors. This allows the subject's treatment to be quickly changed and appropriate treatment for the patient to be taken, even if the risk of disease changes significantly in a short period of time.

[0080] Furthermore, the medical information processing system 1 according to this embodiment compares the updated assessment result with the assessment result before the update for each subject to be assessed, and outputs the comparison result for subjects whose updated assessment result differs from their pre-update assessment result. For example, if the current assessment result has changed from the previous assessment result, a data management screen 50 displaying the change in the assessment result is displayed on the display 13. By checking the data management screen 50 displaying the change in the assessment result, the user can identify subjects whose risk of developing the disease has changed and take appropriate measures.

[0081] The data management screen 50 displaying the changes in the assessment results may be output to a printer connected via the network 20 or the like and printed on paper by the printer. For example, the assessment process may be automatically performed late at night, and a list of subjects whose test results have changed in a subsequent reassessment may be printed. The attending physician can identify subjects whose assessment results have changed by checking the list the following day. For example, if a reexamination is necessary, the attending physician can contact the subject and schedule the next examination. Furthermore, if the test results have changed negatively, the physician can check the list of subjects whose assessment results have changed and take necessary measures, such as reexamination. This prevents subjects who may have a disease from being overlooked. Furthermore, if the assessment results of a subject have changed positively, and the subject's progress is good, the physician can determine that the actual risk at the time of the previous assessment was not significant, and can decide to terminate observation or treatment.

[0082] Furthermore, in this embodiment, the medical information processing system 1 generates a definitive diagnosis based on the test results and diagnostic findings of the diagnosed individual. As a result, even if the results of the definitive diagnosis are not recorded in the medical information system 30, a definitive diagnosis can be obtained based on, for example, the test results of the pathological test, the test results of the rapid antibody test, and the disease name of the diagnostic findings.

[0083] (Second embodiment) A second embodiment will be described. This embodiment is a modification of the configuration of the first embodiment as follows. Descriptions of the same configuration, operation, and effects as those of the first embodiment will be omitted. The medical information processing system 1 according to this embodiment uses the results of the classification and aggregation process in the aforementioned determination process to extract behavioral items that have a certain level of influence on the diagnosis.

[0084] FIG. 8 is a diagram showing the configuration of the medical information processing system 1 of this embodiment. In addition to the functions described in the first embodiment, the processing circuitry 15 executes a risk item extraction function 158. The processing circuitry 15 extracts high-risk action items (hereinafter referred to as risk items) based on information about the diagnosed individual and the diagnostic information of the diagnosed individual using the risk item extraction function 158, and outputs the extracted risk items to the examination database 40, the display 13, a printing device connected to the assessment device 10, etc. Risk items are action items that have a certain level of impact on the diagnosis. For example, the processing circuitry 15 determines action items included in the "possible disease" cluster as risk items. The processing circuitry that realizes the risk item extraction function 158 is an example of an extraction unit.

[0085] Next, the operation of the determination process executed by the determination device 10 of this embodiment will be described. Fig. 9 is a flowchart showing an example of the procedure of the determination process according to this embodiment. The processes of steps S201-S202 and steps S207-S208 are similar to the processes of steps S111-S114 in Fig. 7, respectively, and therefore will not be described again. Here, an example will be described in which each of a plurality of action items included in the "possible disease" cluster is extracted as a risk item, and the extracted risk items are output to the display 13.

[0086] (Determination process) (Step S203) The processing circuit 15 uses the risk item extraction function 158 to determine each of the multiple behavior items included in the "possible disease" cluster as a risk item.

[0087] (Step S204) The processing circuit 15 generates multiple sub-clusters by classifying the area of ​​the "possible disease" cluster into multiple areas centered on each action item using the risk item extraction function 158. Each of the generated sub-clusters corresponds to one of the action items included in the "possible disease" cluster. The range of the sub-cluster also includes the range of the action date and time of the risk item.

[0088] (Step S205) The processing circuit 15 uses the risk item extraction function 158 to apply the above-mentioned aggregation process to each sub-cluster, and calculates the disease probability and its confidence interval for each sub-cluster.

[0089] (Step S206) The processing circuitry 15 generates a list of risk items (hereinafter referred to as the risk item list) using the risk item extraction function 158. The risk item list includes, for each sub-cluster, the central risk item, the date and time range, the disease probability, the confidence interval for the disease probability, the number of subjects in the disease group, and the number of subjects in the disease-free group. The processing circuitry 15 outputs the generated risk item list to the display 13.

[0090] The risk item extraction staff member can perform a confirmation operation for each risk item displayed in the risk item list. For example, an "Approve" button and a "Disapprove" button are displayed for each sub-cluster on the risk item list display screen, and the user can select whether or not to use each risk item in the subsequent assessment process by selecting either the "Approve" button or the "Disapprove" button. In the subsequent assessment process, only risk items corresponding to sub-clusters for which the "Approve" button was selected are used, and risk items corresponding to sub-clusters for which the "Disapprove" button was selected are not used. For example, for risk items corresponding to sub-clusters for which the "Disapprove" button was selected, even if the person being assessed belongs to that sub-cluster in the subsequent assessment process, the risk item is judged to be "low possibility of disease" rather than "possible disease."

[0091] The effects of the medical information processing system 1 having the determination device 10 according to this embodiment will be described below.

[0092] The medical information processing system 1 according to this embodiment extracts risk items based on information about individuals with confirmed diagnoses and their diagnostic information, and outputs the extracted risk items. Risk items are behavior items that have a certain level of influence on the diagnostic result. For example, behavior items included in the "possible disease" cluster are extracted as risk items. The risk items are, for example, listed and displayed on the display 13.

[0093] With the above configuration, the medical information processing system 1 according to this embodiment allows the operator to know which action items have a certain level of influence on the diagnosis by checking the output risk items. In addition, the operator can arbitrarily select the action items to be used for the judgment regarding the estimation of the probability of morbidity from among the risk items.

[0094] The extracted risk items may be output to a medical questionnaire creation device that creates a medical questionnaire. In this case, the processing circuit 15 uses the risk item extraction function 158 to output the extracted risk items to the medical questionnaire creation device as questions to be added to the medical questionnaire. The medical questionnaire creation device receives the extracted risk items and adds questions to the medical questionnaire for collecting behavioral history related to the risk items. For example, questions related to constant risks that are not affected by epidemics, such as areas overseas that are constantly at risk of infection, are always included in the medical questionnaire. Furthermore, by adding questions related to behavioral items newly identified through the risk item extraction process to the medical questionnaire, behavioral history related to risk items newly identified in recent diagnostic results can be appropriately collected.

[0095] In the above embodiment, the test database 40 is installed within a facility and data is recorded for subjects who undergo testing at the facility. However, this is not limiting. The test database 40 may be configured to collect data from an entire region or country in order to collect data from a larger number of subjects. The test database 40 may also be a distributed database. In such cases, data must be anonymized when provided to an outside facility or when acquired from a regional database. For example, k-anonymization, which anonymizes less than k subjects so that they are indistinguishable, is used as a data anonymization method. For example, k-anonymization with k=3 is desirable. Unless at least several subjects are included in a cluster, the confidence interval for the disease probability becomes wide, making it impossible to calculate a reliable disease probability. Therefore, even if the cluster is limited to no more than three people, anonymity can be maintained without causing practical problems.

[0096] (Third embodiment) A third embodiment will be described. This embodiment is a modification of the configuration of the first embodiment as follows. Descriptions of the configuration, operation, and effects similar to those of the first embodiment will be omitted.

[0097] The determination device 10 estimates the effect of a specific drug on the subject (hereinafter referred to as drug effect) based on the historical information of the subject, and determines whether the subject has a "possibility of disease improvement," a "low possibility of disease improvement," or a "high possibility of disease improvement" based on the estimated drug effect. For example, "low possibility of disease improvement" indicates that the possibility of disease improvement by administering the drug is less than a predetermined value. For example, "high possibility of disease improvement" indicates that the possibility of disease improvement by administering the drug is equal to or greater than a predetermined value. For example, "possibility of disease improvement" indicates that the degree of possibility of disease improvement by administering the drug is unknown.

[0098] In this embodiment, test information and information about medication (hereinafter referred to as medication information) are used as the history information. The medication information is information about medications used in the treatment of the person being evaluated. The medication information includes the type, name, administration start date, administration date and time, dosage, etc. of the medication used in the treatment.

[0099] The test information is, for example, the test results of simple tests such as rapid antigen tests and rapid antibody tests. In this case, the test results are the detection results of bacteria or viruses.

[0100] The determination model determines the probability that the condition of a subject will improve if the subject is administered a specific drug. Specifically, the determination model estimates the drug effect of a specific drug on the subject based on the subject's history information, and determines the possibility that the subject's disease will improve.

[0101] The determination device 10 also collects definitive diagnostic results regarding drug effects for multiple subjects, estimates the drug effect of a specific drug on a specific subject based on the collected results, and determines the possibility of improvement of the disease of the subject. If the determination result has changed from the previous determination result, the determination device 10 notifies the user that the determination result has changed.

[0102] The diagnostic information acquired by the record generation function 151 includes information regarding estimated drug efficacy. For example, the diagnostic information may indicate, for a specific drug used to treat pneumonia, that "pneumonia is diagnosed as improved," "pneumonia is diagnosed as not improved," or "there is no judgment regarding improvement of pneumonia." For example, with epidemic diseases such as infectious diseases, even if test results indicate improvement in the disease, the patient may be monitored for several days. For example, if the improved state of the disease is maintained several days after test results indicating improvement are obtained, the administered drug is diagnosed as effective in treating the subject. On the other hand, if the disease worsens several days after test results indicating improvement are obtained, the administered drug is diagnosed as ineffective in treating the subject. The diagnostic result regarding drug efficacy determined during testing (hereinafter referred to as the initial diagnostic result of drug efficacy) is recorded in natural language, for example, in the "estimated drug efficacy" field of the medical information system 30. The initial diagnostic result of drug efficacy is also recorded in the "estimated drug efficacy" field of the record. The result of the definitive diagnosis regarding the drug effect is recorded, for example, as natural language in the item "Definitive diagnosis of drug effect" in the medical information system 30. In addition, the result of the definitive diagnosis of the drug effect is recorded, for example, in the item "Definitive diagnosis of drug effect" in the record.

[0103] The processing circuitry 15 further receives information about the subject to be evaluated by the reception function 152. Specifically, the processing circuitry 15 extracts records of subjects for which a definitive diagnosis of drug effect has not been recorded from all records stored in the memory 11, and acquires the information recorded in the extracted records as information about the subject to be evaluated. The subject to be evaluated is an example of a first subject. The information about the subject to be evaluated is an example of first subject information.

[0104] The processing circuit 15 outputs a judgment result regarding the estimation of drug effectiveness based on the judgment model and information about the subject of judgment using the judgment function 153. The processing circuit 15 outputs the judgment result obtained from the judgment model to the examination database 40, the display 13, a printing device connected to the judgment device 10, etc. The judgment result is an example of first information. The judgment result may also be called first risk information.

[0105] The processing circuit 15 acquires information about the confirmed diagnosis subject using the acquisition function 154. In this embodiment, information about the confirmed diagnosis subject is acquired as information about the confirmed diagnosis subject. Specifically, the processing circuit 15 extracts records of the confirmed diagnosis subject for the drug effect from all records stored in the memory 11, and acquires information about the confirmed diagnosis subject for the drug effect from the extracted records. The information about the confirmed diagnosis subject for the drug effect includes historical information about the confirmed diagnosis subject for the drug effect. The historical information includes test information and a definitive diagnosis result. The confirmed diagnosis subject is an example of a second subject. The information about the confirmed diagnosis subject is an example of second subject information. The definitive diagnosis result of the confirmed diagnosis subject is an example of diagnostic information about the second subject.

[0106] The processing circuit 15 updates the determination model based on information about the person who has been diagnosed with a confirmed drug effect and the diagnostic information of the person who has been diagnosed with a confirmed drug effect by the update function 155. At this time, the processing circuit 15 updates the determination model using the definitive diagnosis result of the person who has been diagnosed with a confirmed drug effect.

[0107] Daily, confirmed diagnoses with confirmed diagnoses of drug efficacy are added to the test database 40. The processing circuit 15 updates the judgment conditions for estimating drug efficacy by performing the above-mentioned classification and aggregation processes on all confirmed diagnoses, including the newly added confirmed diagnoses, using the update function 155. The judgment model is updated as the judgment conditions are updated.

[0108] The processing circuitry 15 outputs an updated judgment result based on the updated judgment model and information about the person to be judged using the judgment function 153. The processing circuitry 15 performs a judgment again using the judgment model, updates the judgment result based on the re-judgment result of the judgment model, and outputs the updated judgment result to the inspection database 40, the display 13, a printing device connected to the judgment device 10, etc. The updated judgment result is an example of second information. The updated judgment result may also be called second risk information.

[0109] The processing circuitry 15 uses the comparison function 156 to compare the updated judgment result with the pre-update judgment result for each subject, and outputs the comparison result for subjects whose updated judgment result differs from their pre-update judgment result.

[0110] Next, we will explain the operation of the medical decision support processing executed by the determination device 10. In this embodiment, the medical decision support processing is processing that collects test information and definitive diagnosis results regarding drug effects for multiple subjects, estimates the drug effects of a specific drug for the subjects based on the collected results, makes a determination regarding the estimated drug effects, and, if the determination result has changed from the previous determination result, outputs information indicating that the determination result has changed.

[0111] (Medical Decision Support Processing) (Step S101) The processing circuit 15 uses the display control function 157 to display a data management screen 50 on the display 13 based on the records stored in the memory 11. Fig. 10 is a diagram showing an example of the data management screen 50. In this embodiment, the display field for diagnostic information on the data display unit 51 further displays the initial diagnostic results of the drug effect.

[0112] (Step S102) When an operation is input to the record creation instruction input unit 53, the processing circuit 15 generates a new record using the record generation function 151, as in the first embodiment.

[0113] (Step S103) When an operation is input to the data collection instruction input unit 54, similar to the first embodiment, the processing circuitry 15 obtains the electronic medical records of the subjects displayed on the data display unit 51 from the medical information system 30, and collects test information and diagnostic information of the subjects whose "suspected diagnosis" field is recorded as "pneumonia." This results in newly acquired test information and diagnostic information that has not previously been recorded in the records. Only the extracted records are displayed on the data display unit 51. The example in Figure 10 shows that, among the records created in the last month, there are four subjects whose "suspected diagnosis" field is recorded as "pneumonia."

[0114] (Step S104) When an operation is input to the update instruction input unit 55, the processing circuit 15 extracts, from the records displayed on the data display unit 51, subjects for whom a definitive diagnosis of drug effect has not been made as subjects to be judged. Then, the processing circuit 15 acquires a judgment result by executing a judgment process, as in the first embodiment. The judgment process of this embodiment is a process for making a judgment regarding an estimation of drug effect. In the judgment process, for subjects for whom a judgment has been made for the first time, a judgment result can be obtained using a judgment model. For subjects for whom a judgment has been made for the second or subsequent time, the latest judgment result can be obtained using a judgment model that reflects the diagnostic results of confirmed diagnoses added from the date of the previous judgment to the date of the current judgment. The judgment result obtained by the judgment process is stored in the record.

[0115] (Step S105) The processing circuit 15 compares the last two judgment results stored in the record of the person to be judged using the comparison function 156. If the judgment results are different, the processing circuit 15 extracts the change in the judgment results and displays it on the data display unit 51.

[0116] The effects of the medical information processing system 1 having the determination device 10 according to this embodiment will be described below.

[0117] To evaluate the effectiveness of a newly developed drug for specific subjects suffering from an infectious disease or a new disease, new records for a large number of subjects are created in the test database 40, and the treatment progress and health status of the large number of subjects are recorded daily. Furthermore, confirmed diagnoses of subjects with confirmed diagnoses of the effectiveness of the newly developed drug are added to the test database 40 daily. As information on subjects with confirmed diagnoses of the drug's effectiveness is collected, the effectiveness of the newly developed drug may change within a few days. Furthermore, in the case of an epidemic disease, the predicted results of the therapeutic drug's effectiveness may change rapidly.

[0118] The medical information processing system 1 according to this embodiment also uses the test information of the target subject as historical information, accepts subject information based on the historical information of the subject to be assessed, and outputs information regarding the possibility of a specific disease for the subject to be assessed based on a judgment model related to a specific disease and the subject information of the subject to be assessed. In this embodiment, the drug effect of a drug used as a therapeutic agent for a specific disease is used as information regarding the possibility of a specific disease. The drug effect is the probability that the condition of a subject administered with a specific drug will improve. Furthermore, the test information of the target subject is used as historical information.

[0119] With the above configuration, the medical information processing system 1 according to this embodiment updates the determination model using the definitive diagnosis results of all subjects, including subjects for whom a definitive diagnosis of drug efficacy has been added. This allows the determination model to reflect the diagnostic results of subjects whose drug efficacy has been definitively diagnosed from the results of a simple test, such as a rapid antibody test, between the date of the previous drug efficacy assessment and the date of the current assessment. Then, by using the updated determination model to assess the drug efficacy of a specific drug based on the results of a simple test, such as a rapid antibody test, it is possible to obtain a determination result that reflects changes in the predicted drug efficacy of the therapeutic drug.

[0120] For example, if there are two drugs that may be effective for a specific disease, it is possible to determine which drug is more effective for the subject. This allows for the drug to be administered to the subject to be quickly changed and appropriate treatment taken for the patient, even if the predicted drug effect changes significantly in a short period of time.

[0121] Furthermore, the medical information processing system 1 according to this embodiment compares the updated assessment result with the assessment result before the update for each subject to be assessed, and outputs the comparison result for subjects whose updated assessment result differs from their pre-update assessment result. For example, if the current assessment result has changed from the previous assessment result, a data management screen 50 displaying the change in the assessment result is displayed on the display 13. By checking the data management screen 50 displaying the change in the assessment result, the user can identify subjects whose predicted drug effect results have changed and take appropriate measures.

[0122] According to at least one of the embodiments described above, even if the possibility of a particular disease changes significantly in a short period of time, it is possible to output a determination result that reflects the changed possibility.

[0123] 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 spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention described in the claims and their equivalents. [Explanation of symbols]

[0124] 1. Medical information processing system 10…Judgment device 11...Memory 12...Communication interface 13...Display 14...Input interface 15...Processing circuit 151...Record generation function 152...Reception function 153...Decision function 154...Acquisition function 155…Update function 156...Compare function 157...Display control function 158...Risk item extraction function 20…Network 30...Medical Information System 40...Inspection database 50...Data management screen 51...Data display section 52...Period setting section 53...Record creation instruction input section 54...Data collection instruction input section 55...Update instruction input section

Claims

1. a storage unit that stores a determination model for determining the possibility of a specific disease; a receiving unit that receives first subject information based on a history of the first subject; an output unit that inputs the first subject information into the determination model, thereby acquiring the possibility of the first subject output from the determination model, and outputs first information including the acquired possibility; an acquisition unit that acquires second subject information based on the history of the second subject and a diagnosis result of the second subject regarding the possibility; an updating unit that updates the determination model based on the second subject information; a comparison unit, the output unit inputs the first subject information into an updated determination model, thereby acquiring the possibility of the first subject output from the updated determination model, and outputs second information including the acquired possibility; the comparison unit compares the first information with the second information for each of the plurality of first subjects, and outputs a comparison result for a subject for which the first information and the second information are different. Medical information processing system.

2. The determination model determines the probability that a particular subject has the disease as the possibility. The medical information processing system according to claim 1 .

3. The history includes at least one of a behavioral history and a medical interview result. The medical information processing system according to claim 2 .

4. the second subject information includes a definitive diagnosis of the second subject with respect to the disease; the update unit updates the determination model based on a definitive diagnosis of the second subject. The medical information processing system according to claim 2 or 3.

5. the second subject information includes test results and diagnostic findings of the second subject; the acquisition unit generates a definitive diagnosis of the second subject based on the test results and diagnostic findings of the second subject. The medical information processing system according to claim 4 .

6. The device further includes an extraction unit that extracts behavioral items that have a certain level of influence on the diagnostic result based on at least one of the behavioral history of the second subject and the results of the interview of the second subject, and the definitive diagnosis of the second subject. The medical information processing system according to claim 4 .

7. the extraction unit adds the extracted behavior items to question items in a medical questionnaire. The medical information processing system according to claim 6.

8. the update unit calculates a disease probability for each of a plurality of clusters related to the possibility, determines classification conditions for the clusters so that the Mahalanobis distance between the plurality of clusters is maximized, or the upper limit of the confidence interval for the disease probability is minimized, or the lower limit of the confidence interval for the disease probability is maximized, and updates the determination model by classifying the second subject into the plurality of clusters so that the difference in disease probability between the plurality of clusters is increased; the output unit determines which of the plurality of clusters the first subject is included in, and outputs the second information based on a disease probability of the cluster to which the first subject is included. The medical information processing system according to any one of claims 2 to 7.

9. the acquiring unit acquires a behavior date and time of the second subject based on at least one of a behavior history and a medical interview result of the second subject; the update unit classifies the second subject into the plurality of clusters based on a behavior date and time of the second subject; the output unit outputs the second information based on a behavior date and time of the first subject. The medical information processing system according to claim 8.

10. The determination model determines the probability that the subject's condition will improve if a particular drug is administered to the subject. The medical information processing system according to claim 1 .

11. The history includes test information of the subject. The medical information processing system according to claim 10.

12. a comparison unit that compares the first information with the second information for each of the plurality of first subjects and outputs a comparison result for a subject whose first information and second information are different from each other; The medical information processing system according to claim 10 or 11.

13. the second subject information includes a definitive diagnosis of the second subject regarding the effect of the drug; the update unit updates the determination model based on a definitive diagnosis of the second subject. The medical information processing system according to any one of claims 10 to 12.

14. the receiving unit receiving first subject information based on the history of the first subject; an output unit inputting the first subject information into a determination model that determines a possibility of the first subject having a specific disease, thereby acquiring the possibility of the first subject output from the determination model, and outputting first information including the acquired possibility; an acquiring unit acquiring second subject information based on a history of the second subject and a diagnosis result of the second subject regarding the possibility; an updating unit updating the determination model based on the second subject information; the output unit inputs the first subject information into an updated determination model, thereby acquiring the possibility of the first subject output from the updated determination model, and outputs second information including the acquired possibility; A medical information processing method comprising:

Images