Information processor, method for extracting data, and data extraction program

Abstract

To make a user to easily use opinion data.SOLUTION: An information processor includes: a reception unit for receiving specification of a condition of extracting opinion data; and an extraction unit for starting processing of extracting opinion data which fits the extraction condition of opinion data recorded after the specification is received and recording the extracted opining data into the database, with the reception of the specification as a trigger. The opining data recorded in the database can be used to make decisions for infection prevention.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus, a data extraction method, and a data extraction program.

Background Art

[0002] Development of technologies for making medical information more easily usable has been underway. For example, Patent Document 1 discloses a medical information management server that assigns search metadata to a radiology report in which text data of findings by a radiologist associated with an image generated by MRI (Magnetic Resonance Imaging) or the like is described.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By assigning metadata by the medical information management server described in Patent Document 1, the convenience in using the accumulated past findings data is improved. However, this medical information management server is not sufficient from the viewpoint of enabling easy use of the findings data generated every moment, and there is room for improvement.

[0005] For example, in the recent worldwide pandemic of viral infectious diseases, it has been necessary to monitor and analyze the spread situation. However, in order to perform such monitoring and analysis, it has been necessary to continuously collect and classify the findings data generated every moment by hand at each medical institution, health center, etc., and the human and time costs have been enormous. The medical information management server described in Patent Document 1 does not contribute to solving such problems.

[0006] An exemplary object of the present disclosure is to provide a technique that enables easy utilization of findings data generated moment by moment.

Means for Solving the Problem

[0007] An information processing apparatus according to an exemplary aspect of the present disclosure includes a reception unit that receives a specification of extraction conditions for findings data indicating the content of a patient's examination, and an extraction unit that, upon receiving the specification, starts a process of extracting, from among the findings data recorded in each of one or more target medical institutions after receiving the specification, those that conform to the specified extraction conditions and recording them in a database.

[0008] In a data extraction method according to an exemplary aspect of the present disclosure, one or more processors execute a reception process of receiving a specification of extraction conditions for findings data indicating the content of a patient's examination, and an extraction process of extracting, from among the findings data recorded in each of one or more target medical institutions after receiving the specification, those that conform to the specified extraction conditions and recording them in a database, and the extraction process is started upon the reception process.

[0009] A data extraction program according to an exemplary aspect of the present disclosure causes a computer to function as a reception unit that receives a specification of extraction conditions for findings data indicating the content of a patient's examination, and an extraction unit that, upon receiving the specification, starts a process of extracting, from among the findings data recorded in each of one or more target medical institutions after receiving the specification, those that conform to the specified extraction conditions and recording them in a database.

Advantages of the Invention

[0010] According to an exemplary aspect of the present disclosure, there is an exemplary effect that a technique can be provided that enables easy utilization of findings data generated moment by moment.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Mode for Carrying Out the Invention

[0012] Hereinafter, embodiments of the present invention will be exemplified. However, the present invention is not limited to the following exemplary embodiments, and various modifications are possible within the scope shown in the claims. For example, embodiments obtained by appropriately combining the technical means employed in the following exemplary embodiments may also be included in the scope of the present invention. In addition, embodiments obtained by appropriately omitting some of the technical means employed in the following exemplary embodiments may also be included in the scope of the present invention. Also, the effects mentioned in the following exemplary embodiments are examples of the effects expected in those exemplary embodiments, and do not define the extension of the present invention. That is, embodiments that do not exhibit the effects mentioned in the following exemplary embodiments may also be included in the scope of the present invention.

[0013] [First Exemplary Embodiment] A first exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. This exemplary embodiment is a basic form for each of the exemplary embodiments described later. Note that the scope of application of each technical means adopted in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technical means adopted in this exemplary embodiment can be adopted in other exemplary embodiments included in the present disclosure as long as there are no particular technical obstacles. Also, each technical means shown in the drawings referred to for explaining this exemplary embodiment can be adopted in other exemplary embodiments included in the present disclosure as long as there are no particular technical obstacles.

[0014] (Configuration of Information Processing Apparatus 1) The configuration of information processing apparatus 1 will be described with reference to FIG. 1. FIG. 1 is a block diagram showing the configuration of information processing apparatus 1. As shown in the figure, information processing apparatus 1 includes a reception unit 101 and an extraction unit 102.

[0015] Reception unit 101 receives a specification of extraction conditions for finding data indicating the examination content of a patient. Note that the method of receiving the specification is not particularly limited. For example, if information processing apparatus 1 includes an input device, reception unit 101 may receive the specification via the input device. Also, for example, if information processing apparatus 1 includes a communication function, reception unit 101 may receive the specification from an external device using the communication function.

[0016] The finding data may be any data indicating the examination content of a patient. For example, the finding data may include data indicating the examination content and results of a doctor, data indicating examination results, data indicating medication instructions, and vital data of the patient. In addition to this, for example, the finding data may include data indicating the gender, date of birth, examination date and time, and medical institution where the examination was performed of the patient.

[0017] Upon receiving the specification of the extraction conditions by the reception unit 101, the extraction unit 102 starts the process of extracting, from the findings data recorded in each of the one or more target medical institutions after the reception of the specification, the data that conforms to the specified extraction conditions and recording it in the database. Therefore, the user of the information processing apparatus 1 only needs to specify the extraction conditions for extracting the desired findings data, and the desired findings data among the findings data generated every moment in a plurality of medical institutions will be accumulated in the database.

[0018] Note that the database may be provided in the information processing apparatus 1 or may be provided outside the information processing apparatus 1. Also, the above "medical institution" means an institution that generates findings data indicating the examination content of a patient. For example, a health center that examines whether a patient is infected with a specific infectious disease and generates findings data indicating the examination results is also included in the scope of the above "medical institution".

[0019] As described above, the information processing apparatus 1 includes a reception unit 101 that receives the specification of the extraction conditions for the findings data indicating the examination content of the patient, and an extraction unit 102 that, upon receiving the specification by the reception unit 101, starts the process of extracting, from the findings data recorded in each of the one or more target medical institutions after the reception of the specification, the data that conforms to the specified extraction conditions and recording it in the database. Therefore, according to the information processing apparatus 1 according to this exemplary embodiment, an effect of being able to easily use the findings data generated every moment can be obtained.

[0020] For example, when the spread of a specific infectious disease is suspected, the user can accumulate the findings data of the patients with that infectious disease in the database by specifying the symptoms typical of that infectious disease as the extraction conditions. The user in this case may be, for example, a public institution. Thereby, the findings data to be targeted when monitoring or analyzing the spread of that infectious disease can be narrowed down to the data accumulated in the database, so the usability of the findings data is improved. Also, for example, the findings data recorded in the database can be used for decision-making in infectious disease countermeasures.

[0021] (Data extraction program) The functions of the above-described information processing apparatus 1 can also be realized by a program. The data extraction program according to this exemplary embodiment causes a computer to function as a reception means for receiving a designation of extraction conditions for finding data indicating a patient's examination content, and, upon receiving the designation, as an extraction means for starting a process of extracting, from the finding data recorded in each of the one or more target medical institutions after receiving the designation, those that conform to the designated extraction conditions and recording them in a database. Therefore, according to the data extraction program according to this exemplary embodiment, an effect can be obtained that it becomes possible to easily use the finding data that occurs every moment.

[0022] (Flow of data extraction method) The flow of the data extraction method will be described with reference to FIG. 2. FIG. 2 is a flowchart showing the flow of the data extraction method. Note that the execution subject of each step in this data extraction method may be a processor included in the information processing apparatus 1, may be a processor included in another apparatus, or may be processors provided in different apparatuses for each step.

[0023] In S1 (reception process), one or more processors receive a designation of extraction conditions for finding data indicating a patient's examination content.

[0024] In S2 (extraction process), one or more processors, after receiving the designation in S1, extract, from the finding data recorded in each of the one or more target medical institutions, those that conform to the designated extraction conditions and record them in a database.

[0025] The process of S2 may be performed, for example, by collectively acquiring the observation data recorded after the acceptance of the designation in S1 and extracting from it the data that matches the extraction conditions. Also, for example, the process of S2 may be performed by repeating a series of processes in which each time new observation data is recorded after the acceptance of the designation in S1, the observation data is acquired, and if the acquired observation data matches the extraction conditions, it is recorded in the database.

[0026] As described above, in the data extraction method according to the present exemplary embodiment, one or more processors execute a reception process for receiving a designation of extraction conditions for finding data indicating the contents of a patient's medical examination, and an extraction process for extracting, after the designation is received, finding data that matches the designated extraction conditions from among the finding data recorded in each of the one or more target medical institutions, and recording the data in a database, the extraction process being started by the reception process, thereby obtaining an effect of making it possible to easily use finding data that is generated from moment to moment.

[0027] Second Exemplary Embodiment A second exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. The scope of application of each technical means employed in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technical means employed in this exemplary embodiment can be employed in other exemplary embodiments included in this disclosure to the extent that no particular technical hindrance occurs. In addition, each technical means shown in each drawing referred to for explaining this exemplary embodiment can be employed in other exemplary embodiments included in this disclosure to the extent that no particular technical hindrance occurs.

[0028] (Configuration of data extraction system 7A) The configuration of the data extraction system 7A will be described based on FIG. 3. FIG. 3 is a diagram showing an overview of the data extraction system 7A. As shown in the figure, the data extraction system 7A includes an information processing device 1A, an output device 2A, databases 3A-1 to 3A-3 of medical institutions, and a database 4A for recording the extracted finding data. When it is not necessary to distinguish each of the databases 3A-1 to 3A-3 of medical institutions, they are simply referred to as "database 3A". Also, the number of databases 3A included in the data extraction system 7A, in other words, the number of medical institutions for which the finding data is to be extracted, is arbitrary and is not limited to the illustrated example.

[0029] Details will be described below, but the data extraction system 7A has a function of recording, in the database 4A, the finding data that meets the extraction conditions specified by the user among the finding data generated at each target medical institution and recorded in the databases 3A-1 to 3A-3 as needed. This function is realized by the information processing device 1A.

[0030] The output device 2A is a device having a function of outputting data. The data output mode of the output device 2A is not particularly limited. For example, the output device 2A may be one that outputs data by display, or one that outputs data by printing, or one that outputs data by voice, or may be one that uses a combination of multiple of these. Below, an example will be described in which the output device 2A is a device that outputs data by display, that is, a display device. When the information processing device 1A includes an output device, the data may be output to that output device and the output device 2A may be omitted.

[0031] Database 3A is a database that records the findings data generated by the target medical institutions. As described above, the number of target medical institutions and the number of databases 3A are arbitrary. For example, the data extraction system 7A can also extract findings data targeting medical institutions nationwide. Note that one database 3A may be provided for each medical institution, or the findings data of multiple medical institutions may be collectively recorded in one database 3A. Also, the data recorded in database 3A may be input to the information processing device 1A via another device.

[0032] Database 4A is a database for recording the findings data extracted by the information processing device 1A. In the data extraction system 7A shown in FIG. 3, an example where the database 4A is provided outside the information processing device 1A is shown, but the database 4A may also be provided inside the information processing device 1A.

[0033] Here, for example, as shown in FIG. 3, it is assumed that extraction conditions are specified such as a body temperature of 39 degrees or more, a positive result of a COVID-19 PCR (Polymerase Chain Reaction) test, and administration of remdesivir. In this case, upon receiving the specification of the extraction conditions, the information processing device 1A starts a process of extracting, from the findings data recorded in the database 3A after receiving the specification, those that conform to the specified extraction conditions and recording them in the database 4A.

[0034] As a result, the database 4A accumulates the findings data of patients with a body temperature of 39 degrees or more, a positive result of a COVID-19 PCR test, and administration of remdesivir. And this makes it easier to analyze patients as described above. For example, the information processing device 1A can also generate a graph as shown in the figure using the findings data accumulated in the database 4A and display it on the output device 2A.

[0035] The graph shown in Fig. 3 shows the age - by - age trend of the number of patients (specifically, the number of times such a diagnosis was made) with a body temperature of 39 degrees or higher, a positive result in the COVID - 19 PCR test, and who were administered remdesivir, with the three axes of time, age, and count. By displaying such a graph, it becomes possible to easily grasp the trend of the tendency of dosing instructions for patients in each age group. Also, the data extraction system 7A can extract and accumulate finding data not only related to COVID - 19 but also other infectious diseases and illnesses such as influenza. That is, the data extraction system 7A can be used for monitoring and analyzing any infectious disease or illness.

[0036] (Configuration of the information processing device 1A) Based on Fig. 4, the configuration of the information processing device 1A will be described. Fig. 4 is a block diagram showing a configuration example of the information processing device 1A. As shown in the figure, the information processing device 1A includes a control unit 10A that comprehensively controls each part of the information processing device 1A, and a storage unit 11A that stores various data used by the information processing device 1A. Also, the information processing device 1A includes a communication unit 12A for the information processing device 1A to communicate with other devices, an input unit 13A that receives inputs to the information processing device 1A, and an output unit 14A for the information processing device 1A to output various data. Also, the control unit 10A includes a reception unit 101A, an extraction unit 102A, a monitoring unit 103A, a natural language processing unit 104A, an output data generation unit 105A, and an output control unit 106A.

[0037] The reception unit 101A, similar to the reception unit 101 in the exemplary embodiment 1, receives the specification of the extraction conditions for the finding data indicating the examination content of the patient. For example, at least any one of the patient's chief complaint, the time when the symptoms started, the nature of the symptoms, the intensity, the presence or absence of remission, the presence or absence of exacerbation, the site where the symptoms appeared, accompanying symptoms, the time - series change of the symptoms, the treatment plan such as dosing instructions, and the examination results may be used as the extraction conditions.

[0038] The extraction unit 102A, similar to the extraction unit 102 of the exemplary embodiment 1, upon receiving the designation by the reception unit 101A, after receiving the designation, starts the process of extracting, from the findings data recorded in each of the one or more target medical institutions, those that conform to the designated extraction conditions and recording them in the database 4A. In order to facilitate the use of the recorded findings data, it is desirable for the extraction unit 102A to structure and record the findings data in a predetermined format.

[0039] The user of the data extraction system 7A may appropriately specify extraction conditions according to the usage purpose of the findings data. For example, when monitoring or analyzing a specific symptom or syndrome, the extraction conditions indicating that symptom may be specified. Thereby, the extraction unit 102A records in the database 4A the findings data indicating that the patient corresponds to that symptom. Therefore, according to the information processing apparatus 1A, in addition to the effects exhibited by the information processing apparatus 1, an effect is obtained that it becomes possible to facilitate the monitoring and analysis of a specific symptom.

[0040] The monitoring unit 103A monitors the recording of the findings data in the database 3A. Also, when the monitoring unit 103A detects that new findings data has been recorded in the database 3A, it acquires the recorded findings data from the database 3A. The monitoring by the monitoring unit 103A is started upon receiving the designation of the extraction conditions by the reception unit 101A.

[0041] Note that the monitoring unit 103A may access the database 3A after a predetermined time from when the reception unit 101A receives the designation of the extraction conditions and collectively acquire the findings data recorded at that predetermined time. Also, the monitoring unit 103A may repeat this process every predetermined time. Further, the monitoring unit 103A may, when a predetermined operation is performed by the user, collectively acquire the findings data recorded during the period from when the reception unit 101A receives the designation of the extraction conditions to when the operation is performed.

[0042] The natural language processing unit 104A is configured to determine whether the finding data including the examination content described in the natural language conforms to the extraction conditions. The details of the natural language processing unit 104A will be described in the item of "Example 2 of the method for extracting finding data" described later.

[0043] The output data generation unit 105A generates output data using the finding data recorded in the database 4A. The output data may be anything according to the usage purpose of the finding data. For example, when it is desired to monitor a specific symptom or syndrome, the symptom may be specified as an extraction condition. In this case, the output data generation unit 105A may generate output data indicating the characteristics of the symptom.

[0044] For example, the output data generation unit 105A may generate data (such as a graph) showing the transition of the number of patients diagnosed as corresponding to the syndrome to be monitored (or the number of times the diagnosis of corresponding to the syndrome has been made) as output data. In addition to this, for example, the output data generation unit 105A may generate statistical data (such as the average age of patients diagnosed as corresponding to a specific syndrome, etc.) about the finding data recorded in the database 4A. The transition of the number of patients and the statistical data can also be data classified by age group, gender, region, etc. Further, for example, the output data generation unit 105A may perform causal analysis on the finding data recorded in the database 4A and generate output data (such as a causal graph) showing the analysis result. Thereby, it can contribute to the investigation of the cause of the syndrome and the study of effective countermeasures.

[0045] The output control unit 106A causes the output device 2A to output data. For example, if the output device 2A is a display device, the output control unit 106A causes the output device 2A to display characters and images. The content of the data to be output to the output device 2A is not particularly limited. For example, the output control unit 106A may cause the output device 2A to display an image or a character string that prompts the input of extraction conditions, or may cause the output device 2A to display the output data generated by the output data generation unit 105A. As described above, this output data may be, for example, data indicating the characteristics of symptoms shown in the specified extraction conditions. Note that the output control unit 106A may cause the output unit 14A to output data.

[0046] As described above, the information processing apparatus 1A includes an output control unit 106A that causes the output device 2A to output output data indicating the characteristics of symptoms shown in the extraction conditions, which is generated using the finding data recorded in the database 4A. Here, when the extraction unit 102A records new finding data in the database 4A, the output data generation unit 105A may update the output data using the recorded new finding data. Then, the output control unit 106A may update the output data to be output to the output device 2A to reflect the new finding data in response to the extraction unit 102A recording new finding data in the database 4A. Thereby, in addition to the effects exhibited by the information processing apparatus 1, an effect is obtained in that it becomes possible to monitor the finding data in real time.

[0047] (Example 1 of Method for Extracting Finding Data) The method for extracting finding data will be described with reference to FIG. 5. FIG. 5 is a diagram for explaining a method for determining whether or not extraction conditions are satisfied when the finding data is an electronic medical record. Note that the extraction conditions in the example of FIG. 5 are the same as those in the example of FIG. 3.

[0048] The findings data shown in FIG. 5 includes examination items for COVID-19. Specifically, this findings data includes a plurality of data items related to examinations such as the route of coming to the hospital, and the examination content is input into each data item. This findings data is generated by inputting the examination content into an electronic medical record including a plurality of data items. That is, by having doctors and the like use an electronic medical record that includes a plurality of data items and is structured so that the examination content can be input into each data item, the findings data as shown in the figure can be easily generated. Note that the electronic medical record with the examination content input may be used as the findings data as it is, or the values of each data item may be extracted from the electronic medical record and associated with the corresponding data items to be used as the findings data.

[0049] In the findings data shown in FIG. 5, for each data item, the form is such that the one corresponding to the examination content is selected from a plurality of options. However, the data items of the findings data may include those in which numerical values (such as body temperature, etc.) are input. Further, the findings data may include personal information such as the patient's name, date of birth, gender, place of residence, etc., and data indicating the examination facility, the area where the examination was conducted, the place of infection, the examination date and time, and the attending doctor, etc.

[0050] As described above, the findings data may be generated by inputting the examination content into each of the plurality of data items in the electronic medical record. In this case, the extraction unit 102A determines whether the findings data conforms to the extraction conditions from the values of the data items corresponding to the extraction conditions in the electronic medical record. Thereby, in addition to the effect exerted by the information processing apparatus 1, an effect is obtained in that it becomes possible to accurately extract the findings data that conforms to the extraction conditions.

[0051] For example, as shown in FIG. 5, assume that extraction conditions are specified such that the body temperature is 39 degrees or higher, the result of a PCR test is positive, and remdesivir is administered. In this case, the extraction unit 102A refers to the data items corresponding to these extraction conditions in the electronic medical record, that is, the values of the data items of "findings", "PCR test", and "medication instruction". Then, the extraction unit 102A determines whether the findings data conforms to the extraction conditions from those values. In the findings data shown in FIG. 5, since "findings" is "body temperature 39 degrees or higher", "PCR test" is "(+) positive", and "medication instruction" is "remdesivir", this findings data is determined to conform to the extraction conditions and is recorded in the database 4A.

[0052] Note that the extraction unit 102A does not necessarily need to record all the data included in the findings data (which can also be referred to as the input electronic medical record) in the database 4A, and may record only the data to be used. For example, the extraction unit 102A may record in the database 4A the anonymized findings data excluding personal information such as the patient name from the findings data. Thereby, the effective use of the findings data considering personal information protection can be realized. Also, when a syndrome is specified as an extraction condition, the extraction unit 102A may extract the data related to the syndrome from the findings data and record it in the database 4A. Further, the user may be able to specify the data items to be recorded in the database 4A.

[0053] By the way, since the data items in the electronic medical record can be set as appropriate, the data items set for each type of electronic medical record may be different. For example, in the electronic medical record of a certain vendor, the body temperature is recorded in the data item of "findings" as in the example of FIG. 5, while in the electronic medical record of another vendor, it may be recorded in other data items such as "body temperature".

[0054] Therefore, the extraction unit 102A may apply a conversion rule according to the type of the electronic medical record, which converts the data items included in the electronic medical record into standard data items, and determine whether the finding data conforms to the extraction conditions based on the converted data items. As a result, in addition to the effect exerted by the information processing apparatus 1, an effect is obtained in that it becomes possible to accurately extract the finding data that conforms to the extraction conditions from a plurality of types of electronic medical records.

[0055] (Example 2 of the method for extracting finding data) Further, the finding data may be such that part or all of the examination content is described in natural language. In this case, by using the function of the natural language processing unit 104A, it is possible to determine whether the finding data satisfies the extraction conditions. This will be described with reference to FIG. 6. FIG. 6 is a diagram for explaining a method of determining whether the finding data described in natural language satisfies the extraction conditions.

[0056] In the finding data shown in FIG. 6, the examination content is described in natural language, but the content is the same as the finding data shown in FIG. 5. Also, the extraction conditions in the example of FIG. 6 are the same as those in the example of FIG. 5. When such finding data is acquired by the monitoring unit 103A, the natural language processing unit 104A inputs the examination content described in natural language included in the finding data into the machine-learned language model, and outputs data indicating whether the finding data conforms to the extraction conditions.

[0057] Here, the "language model" used by the natural language processing unit 104A is a machine-learned model constructed by machine learning the arrangement of its components (such as words) in a sentence expressed in natural language and the arrangement of sentences in a text. For example, the above language model may be a model that is machine-learned to generate an answer sentence for a question sentence expressed in natural language. Further, the above language model may be generated by machine learning using training data in which, for a pair of finding data described in natural language and extraction conditions, whether or not the finding data conforms to the extraction conditions is associated as correct answer data. Alternatively, the above language model may be a general-purpose language model fine-tuned with the above training data.

[0058] For example, as shown in FIG. 6, the natural language processing unit 104A may input, to the language model, in addition to the examination content shown in the finding data and the extraction conditions, a query asking whether or not the examination content satisfies the extraction conditions. Then, when the answer output by the language model indicates that the extraction conditions are satisfied, the extraction unit 102A records the finding data including the examination content in the database 4A (that is, the finding data is extracted). On the other hand, when the output of the language model indicates that the extraction conditions are not satisfied, the extraction unit 102A does not record the finding data including the examination content in the database 4A (that is, the finding data is not extracted).

[0059] When the electronic medical record includes examination content described in natural language in part, the natural language processing unit 104A may input that part to the language model and have it answer whether or not that part satisfies at least part of the extraction conditions. In this case, the extraction unit 102A may comprehensively determine whether or not the extraction conditions are satisfied by combining the other part (structured part) of the electronic medical record and the above answer. Further, the natural language processing unit 104A may input the entire electronic medical record including part of the examination content described in natural language to the language model and have it answer whether or not the electronic medical record satisfies the extraction conditions.

[0060] In this way, the information processing apparatus 1A includes a natural language processing unit 104A that inputs the examination content described in natural language included in the findings data into a pre-trained language model and outputs data indicating whether or not the examination content conforms to the extraction conditions. Then, the extraction unit 102A determines whether or not the above-described findings data conforms to the extraction conditions based on the data output by the language model. As a result, in addition to the effect exhibited by the information processing apparatus 1, an effect is obtained in that even when the findings data includes examination content described in natural language, it is possible to accurately extract the findings data that conforms to the extraction conditions.

[0061] Note that the natural language processing unit 104A may structure the examination content described in natural language using a language model. In this case, the extraction unit 102A can determine whether or not the findings data including the structured examination content conforms to the extraction conditions using the structured examination content. For example, the natural language processing unit 104A may input a query instructing the extraction of a description corresponding to the extraction conditions from the findings data, together with the findings data (described in natural language) and the extraction conditions, into the language model. As a result, since the description corresponding to the extraction conditions is output from the language model, the natural language processing unit 104A can structure by associating the extraction conditions with the output description.

[0062] For example, in the example of FIG. 6, the natural language processing unit 104A may generate a query "Extract the measurement result of body temperature from the findings data" using the keyword "body temperature" included in the extraction conditions and input it into the language model. In this case, the data output by the language model will indicate "39.0 degrees". Therefore, the natural language processing unit 104A can structure by associating the value "39.0 degrees" with the data item "body temperature". The structured findings data can be used to determine whether or not the extraction conditions are satisfied in the same manner as in the example of FIG. 5.

[0063] (Overall processing flow) The flow of the process executed by the information processing device 1A will be described with reference to Fig. 7. Fig. 7 is a flow diagram showing an example of the process executed by the information processing device 1A. Fig. 7 includes a data extraction method according to this exemplary embodiment.

[0064] In S11 (reception process), the reception unit 101A receives the designation of extraction conditions. Then, in S12, the monitoring unit 103A starts monitoring the finding data in the database 3A. In S13, the monitoring unit 103A judges whether or not the finding data has been added to the database 3A. If the judgment in S13 is YES, the process proceeds to S14, and if the judgment in S13 is NO, the process proceeds to S17.

[0065] In S14 (extraction process), the extraction unit 102A extracts the finding data that matches the specified extraction conditions from the recorded finding data after the acceptance of the designation in S11, and records the extracted data in the database 4A. S14 is a process that is performed in response to the process of S11. Details of S14 will be described later with reference to FIG. 8.

[0066] In S15, the output data generation unit 105A generates output data using the findings data recorded in S14. Then, in S16, the output control unit 106A causes the output device 2A to output the output data generated in S15. When the process of S15 is performed for the second or subsequent times, the output data generation unit 105A updates the output data using the newly recorded findings data. Then, in the following S16, the output control unit 106A updates the output data to be output by the output device 2A.

[0067] The configuration of performing the processes of S15 and S16 after the process of S14 is effective, for example, when a specific syndrome is constantly monitored. If constant monitoring is not necessary, the processes of S15 and S16 may be performed in response to a user instruction. Also, instead of monitoring for the addition of finding data, the finding data recorded after the acceptance of the specification in S11 may be obtained at a predetermined timing, and the finding data that satisfies the extraction conditions may be extracted from the finding data.

[0068] In S17, the monitoring unit 103A determines whether to end the monitoring. If it is determined to be YES in S17, the illustrated process ends. On the other hand, if it is determined to be NO in S17, the process returns to S13. Note that the conditions for ending the monitoring may be determined as appropriate. For example, it may be configured to accept the specification of the monitoring period together with the extraction conditions. In this case, the expiration of the specified period becomes the condition for ending the monitoring.

[0069] (Flow of processing for extraction and recording of findings data) The processing of S14 in FIG. 7, that is, the details of the processing for extraction and recording of findings data, will be described based on FIG. 8. FIG. 8 is a flowchart showing the flow of processing for extraction and recording of findings data.

[0070] In S141, the natural language processing unit 104A determines whether the added findings data includes examination content described in natural language. If it is determined to be NO in S141, the process proceeds to S144, and if it is determined to be YES in S141, the process proceeds to S142.

[0071] In S142, the natural language processing unit 104A generates a query asking whether the examination content described in natural language included in the added findings data satisfies the extraction conditions. Then, in S143, the natural language processing unit 104A inputs the query generated in S142 into the language model and obtains its output.

[0072] In S144, the extraction unit 102A determines whether the newly added finding data satisfies the extraction conditions (specified in S11 of FIG. 7). For example, when it is determined as NO in S141 and the process transitions to S144, the extraction unit 102A may determine whether the finding data conforms to the extraction conditions from the value of the data item corresponding to the extraction conditions in the finding data. Also, for example, when it is determined as YES in S141 and the process transitions to S144, the extraction unit 102A may determine whether the finding data conforms to the extraction conditions based on the output of the language model obtained in S143. Further, when the finding data is an electronic medical record, the extraction unit 102A may perform the determination in S144 after applying the conversion rule according to the type thereof.

[0073] When the extraction unit 102A determines as NO in S144, it ends the process without recording the newly added finding data in the database 4A. On the other hand, when the extraction unit 102A determines as YES in S144, it proceeds to S145 and records the newly added finding data in the database 4A. Thereby, the process of FIG. 8 ends.

[0074] 〔Modification Example〕 The execution subject of each process described in each of the above embodiments is arbitrary and is not limited to the above examples. That is, a data extraction system having the same functions as the information processing apparatus 1 or 1A can be constructed by a plurality of mutually communicable devices. For example, each process in the flowcharts shown in FIGS. 2, 7, and 8 can also be executed in parallel by a plurality of devices (which can also be referred to as processors).

[0075] 〔Example of Realization by Software〕 Some or all of the functions of the information processing apparatuses 1 and 1A may be realized by hardware such as an integrated circuit (IC chip) or may be realized by software.

[0076] In the latter case, the information processing apparatuses 1 and 1A are realized by, for example, a computer that executes instructions of a program which is software for realizing each function. An example of such a computer (hereinafter referred to as computer C) is shown in FIG. 9. FIG. 9 is a block diagram showing the hardware configuration of the computer C that functions as the information processing apparatus 1 or 1A.

[0077] The computer C includes at least one processor C1 and at least one memory C2. A program (data extraction program) P for operating the computer C as the above-described apparatuses is recorded in the memory C2. In the computer C, the processor C1 reads and executes the program P from the memory C2, whereby each function of the information processing apparatus 1 or 1A is realized.

[0078] As the processor C1, for example, a CPU (Central Processing Unit), a GPU (Graphic Processing Unit), a DSP (Digital Signal Processor), an MPU (Micro Processing Unit), an FPU (Floating point number Processing Unit), a PPU (Physics Processing Unit), a TPU (Tensor Processing Unit), a quantum processor, a microcontroller, or a combination thereof can be used. As the memory C2, for example, a flash memory, an HDD (Hard Disk Drive), an SSD (Solid State Drive), or a combination thereof can be used.

[0079] Note that the computer C may further include a RAM (Random Access Memory) for expanding the program P during execution or temporarily storing various data. Further, the computer C may further include a communication interface for transmitting and receiving data to and from other devices. Further, the computer C may further include an input / output interface for connecting input / output devices such as a keyboard, a mouse, a display, and a printer.

[0080] Further, the program P can be recorded on a non-transitory tangible recording medium M readable by the computer C. As such a recording medium M, for example, a tape, a disk, a card, a semiconductor memory, or a programmable logic circuit can be used. The computer C can obtain the program P via such a recording medium M. Further, the program P can be transmitted via a transmission medium. As such a transmission medium, for example, a communication network or a broadcast wave can be used. The computer C can also obtain the program P via such a transmission medium.

[0081] 〔Supplementary Note〕 This disclosure includes the technologies described in the following supplementary notes. However, the present invention is not limited to the technologies described in the following supplementary notes, and various modifications are possible within the scope indicated in the claims.

[0082] (Supplementary Note A1) An information processing apparatus comprising: a reception unit that receives a specification of extraction conditions for finding data indicating the examination content of a patient; and an extraction unit that, upon receiving the specification, starts a process of extracting, from the finding data recorded in each of one or more target medical institutions after receiving the specification, those that conform to the specified extraction conditions and recording them in a database.

[0083] (Supplementary Note A2) The said finding data is generated by inputting examination details into each of a plurality of data items in an electronic medical record, and the said extraction means determines whether or not the said finding data conforms to the extraction conditions from the values of the data items corresponding to the said extraction conditions in the said electronic medical record. The information processing apparatus according to Supplementary Note A1.

[0084] (Supplementary Note A3) The said extraction means applies a conversion rule corresponding to the type of the said electronic medical record, which is a conversion rule for converting the data items included in the said electronic medical record into standard data items, and determines whether or not the said finding data conforms to the extraction conditions based on the data items after conversion. The information processing apparatus according to Supplementary Note A2.

[0085] (Supplementary Note A4) The said finding data includes natural language processing means for inputting the examination details described in natural language into a machine-learned language model and outputting data indicating whether or not the said examination details conform to the extraction conditions, and the said extraction means determines whether or not the said finding data conforms to the extraction conditions based on the data output by the said language model. The information processing apparatus according to any one of Supplementary Notes A1 to A3.

[0086] (Supplementary Note A5) The said extraction conditions indicate one or more symptoms, and the said extraction means records the said finding data indicating that the patient corresponds to the said symptoms in the said database. The information processing apparatus according to any one of Supplementary Notes A1 to A4.

[0087] (Supplementary Note A6) An output control unit that outputs output data indicating the characteristics of the said symptoms, generated using the said finding data recorded in the said database, to an output device, and the said output control unit updates the output data output to the said output device to reflect the said new finding data in response to the said extraction means recording new finding data in the said database. The information processing apparatus according to Supplementary Note A5.

[0088] (Supplementary Note B1) One or more processors execute a reception process of receiving a specification of extraction conditions for finding data indicating the examination details of a patient, and after receiving the specification, extract, from the finding data recorded in each of the one or more target medical institutions, those that conform to the specified extraction conditions and record them in a database. The extraction process is started triggered by the reception process. A data extraction method.

[0089] (Appendix B2) The finding data is generated by inputting examination details into each of a plurality of data items in an electronic medical record. In the extraction process, the at least one processor determines whether the finding data conforms to the extraction conditions from the values of the data items corresponding to the extraction conditions in the electronic medical record. The data extraction method described in Appendix B1.

[0090] (Appendix B3) In the extraction process, the at least one processor applies a conversion rule according to the type of the electronic medical record, which converts the data items included in the electronic medical record into standard data items, and determines whether the finding data conforms to the extraction conditions based on the converted data items. The data extraction method described in Appendix B2.

[0091] (Appendix B4) The at least one processor includes a process of inputting the examination details described in natural language included in the finding data into a machine-learned language model and outputting data indicating whether the examination details conform to the extraction conditions. In the extraction process, the at least one processor determines whether the finding data conforms to the extraction conditions based on the data output by the language model. The data extraction method described in any one of Appendices B1 to B3.

[0092] (Appendix B5) The extraction conditions indicate one or more symptoms, and in the extraction process, the at least one processor records, in the database, the finding data indicating that the patient corresponds to the symptoms, according to the data extraction method described in any of Appendices B1 to B4.

[0093] (Appendix B6) Output control processing in which the at least one processor causes an output device to output output data indicating characteristics of the symptoms, the output data being generated using the finding data recorded in the database, and the data extraction method described in Appendix B5 in which the at least one processor updates the output data output to the output device to reflect the new finding data in response to new finding data being recorded in the database.

[0094] (Appendix C1) A data extraction program that causes a computer to function as reception means for receiving a specification of extraction conditions for finding data indicating the content of a patient's examination, and extraction means for starting, after receiving the specification, a process of extracting, from the finding data recorded in each of one or more target medical institutions, data that conforms to the specified extraction conditions and recording it in a database.

[0095] (Appendix C2) The finding data is generated by inputting examination content into each of a plurality of data items in an electronic medical record, and the extraction means determines whether the finding data conforms to the extraction conditions based on the values of the data items corresponding to the extraction conditions in the electronic medical record, according to the data extraction program described in Appendix C1.

[0096] (Appendix C3) The extraction means applies a conversion rule according to the type of the electronic medical record, the conversion rule for converting the data items included in the electronic medical record into standard data items, and determines whether the finding data conforms to the extraction conditions based on the converted data items, according to the data extraction program described in Appendix C2.

[0097] (Appendix C4) Function the computer as a natural language processing means for inputting the examination content described in natural language included in the finding data into a machine-learned language model and outputting data indicating whether or not the examination content conforms to the extraction condition, and the extraction means determines whether or not the finding data conforms to the extraction condition based on the data output by the language model, and is the data extraction program according to any one of Appendices C1 to C3.

[0098] (Appendix C5) The extraction condition indicates one or more symptoms, and the extraction means records the finding data indicating that the patient corresponds to the symptom in the database, and is the data extraction program according to any one of Appendices C1 to C4.

[0099] (Appendix C6) Function the computer as an output control means for outputting the output data indicating the characteristics of the symptom generated using the finding data recorded in the database to an output device, and the output control means updates the output data to be output to the output device to reflect the new finding data in response to the extraction means recording new finding data in the database, and is the data extraction program according to Appendix C5.

[0100] (Appendix D1) An information processing device including at least one processor, the at least one processor executing a reception process for receiving a specification of an extraction condition for finding data indicating a patient's examination content, and an extraction process for starting, after receiving the specification, extracting, from the finding data recorded in each of one or more target medical institutions, those that conform to the specified extraction condition and recording them in a database.

[0101] Incidentally, the information processing apparatus may further include a memory. Further, a program for causing the at least one processor to execute each of the above processes may be stored in the memory.

[0102] (Appendix D2) The findings data is generated by inputting examination contents into each of a plurality of data items in an electronic medical record. In the extraction process, the at least one processor determines whether the findings data conforms to the extraction conditions from the values of the data items corresponding to the extraction conditions in the electronic medical record. The information processing apparatus according to Appendix D1.

[0103] (Appendix D3) In the extraction process, the at least one processor applies a conversion rule according to the type of the electronic medical record, which converts the data items included in the electronic medical record into standard data items, and determines whether the findings data conforms to the extraction conditions based on the converted data items. The information processing apparatus according to Appendix D2.

[0104] (Appendix D4) The at least one processor executes natural language processing in which the examination content described in natural language included in the findings data is input into a machine-learned language model, and data indicating whether the examination content conforms to the extraction conditions is output. In the extraction process, based on the data output by the language model, it is determined whether the findings data conforms to the extraction conditions. The information processing apparatus according to any one of Appendix D1 to D3.

[0105] (Appendix D5) The extraction conditions indicate one or more symptoms. In the extraction process, the at least one processor records the findings data indicating that the patient corresponds to the symptoms in the database. The information processing apparatus according to any one of Appendix D1 to D4.

[0106] (Appendix D6) The at least one processor outputs, to an output device, output data indicating characteristics of the symptoms, the output data being generated using the finding data recorded in the database, and executes an update process for updating the output data output to the output device to reflect the new finding data in response to the new finding data being recorded in the database. The information processing apparatus according to Appendix D5.

[0107] (Appendix E) A non-transitory recording medium recording a data extraction program that causes a computer to execute a reception process for receiving a specification of extraction conditions for finding data indicating examination contents of a patient, and an extraction process for starting, upon receiving the specification, a process of extracting, from the finding data recorded in each of one or more target medical institutions after receiving the specification, data that conforms to the specified extraction conditions and recording the data in a database.

Explanation of Signs

[0108] 1 Information processing apparatus 101 Reception unit (reception means) 102 Extraction unit (extraction means) 1A Information processing apparatus 101A Reception unit (reception means) 102A Extraction unit (extraction means) 104A Natural language processing unit (natural language processing means) 106A Output control unit (output control means)

Claims

1. Receiving means for receiving a specification of extraction conditions for finding data indicating the examination details of a patient; Extraction means for starting a process of extracting, from the finding data recorded in each of one or more target medical institutions after receiving the specification, those that conform to the specified extraction conditions and recording them in a database, triggered by the reception of the specification; An information processing apparatus comprising:

2. The finding data is generated by inputting examination details into each of a plurality of data items in an electronic medical record, The extraction means determines whether the finding data conforms to the extraction conditions from the values of the data items corresponding to the extraction conditions in the electronic medical record. The information processing apparatus according to claim 1.

3. The extraction means applies a conversion rule according to the type of the electronic medical record, which converts the data items included in the electronic medical record into standard data items, and determines whether the finding data conforms to the extraction conditions based on the converted data items. The information processing apparatus according to claim 2.

4. Natural language processing means for inputting the examination details described in natural language included in the finding data into a machine-learned language model and outputting data indicating whether the examination details conform to the extraction conditions, The extraction means determines whether the finding data conforms to the extraction conditions based on the data output by the language model. The information processing apparatus according to any one of claims 1 to 3.

5. The extraction conditions indicate one or more symptoms, The extraction means records in the database the finding data indicating that the patient corresponds to the symptoms. The information processing apparatus according to any one of claims 1 to 3.

6. An output control unit for outputting output data indicating the characteristics of the symptoms, generated using the finding data recorded in the database, to an output device, The output control unit updates the output data to be output to the output device to reflect the new finding data in response to the extraction means recording new finding data in the database. The information processing apparatus according to claim 5.

7. One or more processors perform A reception process for receiving a specification of extraction conditions for finding data indicating the examination details of a patient; After receiving the specified acceptance, an extraction process is executed to extract, from the finding data recorded in each of the one or more target medical institutions, those that meet the specified extraction conditions and record them in a database. The extraction process is a data extraction method that starts with the reception process.

8. A computer a reception means for receiving a specification of extraction conditions for finding data indicating the content of a patient's examination, and a data extraction program that functions as an extraction means for starting a process of extracting, from the finding data recorded in each of the one or more target medical institutions after receiving the specification, those that meet the specified extraction conditions and recording them in a database, triggered by the reception of the specification.

Images