Medical information processing apparatus, medical information processing system, and medical information processing method

Abstract

To efficiently record and share important information for determining a medical action to be executed next to a subject.SOLUTION: A medical information processing apparatus includes an extraction unit, an identification unit, a calculation unit, and a presentation unit. The extraction unit extracts word elements from record data in which speech contents of a plurality of speakers about a subject are recorded. The specification unit specifies a word attribute of each of the extracted word elements based on a concept indicated by each of the word elements. The calculation part calculates importance of each of the word elements on the basis of at least one of the word attribute, a speaker attribute of a speaker related to the word element, a temporal relationship between a time related to a speech content including the word element and a reference time, and medical resource information of a medical resource related to the word element. The presentation unit presents a candidate for a medical action to be performed on the subject on the basis of the word element corresponding to the importance.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Traditionally, in intensive care units (ICUs) at hospitals and other medical facilities, a single patient is cared for 24 hours a day by multidisciplinary staff. When providing this type of care, timely collaboration and information sharing between the various professions is important. For this reason, it is common for multidisciplinary staff to meet in the ICU to make multidisciplinary rounds with the aim of determining the patient care policy for the day (setting to-do lists).

[0003] During multidisciplinary rounds, participants evaluate and summarize the patient's current condition from a large amount of medical data and give a presentation, but due to the short meeting time, this information may not be shared effectively between professions.In addition, in order to effectively discuss setting tasks, it is important to understand the patient's progress, but generally, collecting information to understand the patient's progress from electronic medical records, etc., is time-consuming and laborious.

[0004] For these reasons, there is a need for an efficient way to record and share important information for setting ToDos. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2019-149083 Summary of the Invention [Problem to be solved by the invention]

[0006] One of the problems to be solved by the embodiments disclosed in this specification and drawings is to efficiently record and share important information for determining the next medical procedure to be performed on a subject. However, the problems to be solved by the embodiments disclosed in this specification and drawings are 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]

[0007] A medical information processing apparatus according to an embodiment includes an extraction unit, an identification unit, a calculation unit, and a presentation unit. The extraction unit extracts word elements from recorded data recording the content of statements made by multiple speakers regarding a subject. The identification unit identifies word attributes of each of the extracted word elements based on a concept indicated by each of the extracted word elements. The calculation unit calculates the importance of each word element based on at least one of the word attributes, the speaker attributes of the speakers related to the word elements, the temporal relationship between a time related to the content of statements including the word elements and a reference time, and medical resource information of medical resources related to the word elements. The presentation unit presents candidate medical procedures to be performed on the subject based on the word elements according to their importance. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a medical information processing system according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of a text data generation process according to the embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of word segmentation according to the embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of fluctuation correction processing according to the embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of the attribute specification process according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of the importance calculation process according to the embodiment. [Figure 7]FIG. 7 is a diagram illustrating an example of a screen for presenting a ToDo list according to the embodiment. [Figure 8] FIG. 8 is a flowchart showing an example of processing executed by the medical image processing apparatus according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of a medical information processing apparatus, a medical information processing system, and a medical information processing method will be described in detail with reference to the drawings.

[0010] FIG. 1 is a diagram showing an example of the configuration of a medical information processing system according to this embodiment.

[0011] The medical information processing system according to this embodiment is installed in a medical facility such as a hospital, and efficiently records and shares important information for determining the next medical procedure to be performed on a subject.

[0012] Specifically, the medical information processing system of this embodiment extracts word elements from the remarks of participants in a meeting (hereinafter simply referred to as a meeting) held as part of a multi-disciplinary medical round, calculates the importance of the word elements, and presents medical procedures derived from word elements with high importance to the user as candidates (ToDo list) for the next medical procedure to be performed on the subject.

[0013] 1, a medical information processing system 100 according to this embodiment includes an information integration system 110 and a medical information processing device 120. The information integration system 110 and the medical information processing device 120 are connected to each other via a network 130 such as a LAN (Local Area Network) or a WAN (Wide Area Network) so as to be able to communicate with each other.

[0014] The information integration system 110 includes a patient information DB (Data Base) 111. For example, the information integration system 110 is realized by computer equipment such as a server or a workstation equipped with an electronic medical record system, a HIS (Hospital Information System), a RIS (Radiology Information System), or the like.

[0015] The patient information DB 111 stores various types of medical information about patients. A patient is an example of a subject. For example, the patient information DB 111 stores, as medical information, information about each patient's illness (pre-existing conditions, complications, side effects, etc.), symptoms (pathological conditions) of treatments (including administered drugs), connected devices, etc.

[0016] The medical information processing device 120 is a device used by medical professionals such as doctors who decide on medical procedures to be performed on patients, and has a function for presenting medical procedures to be performed on patients based on comments from conference participants. For example, the medical information processing device 120 is realized by a personal computer or the like.

[0017] Specifically, the medical information processing device 120 includes a network (NW) interface 121, a storage circuitry 122, an input interface 123, a display 124, a microphone 125, and a processing circuitry 127.

[0018] The NW interface 121 controls the transmission and communication of various data transmitted and received between the medical information processing device 120 and other devices connected via the network 130. Specifically, the NW interface 121 is connected to the processing circuitry 127, and outputs information received from the information integration system 110 to the processing circuitry 127. For example, the NW interface 121 is realized by a network card, a network adapter, a NIC (Network Interface Controller), or the like.

[0019] The storage circuitry 122 stores various data, various programs, etc. For example, the storage circuitry 122 is realized by a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, a hard disk, an optical disk, or the like.

[0020] Specifically, the storage circuitry 122 includes a rounds record DB 122a, which stores rounds record information that records comments and ToDo lists of participants in a meeting.

[0021] The input interface 123 receives input operations of various instructions and various information from an operator. Specifically, the input interface 123 is connected to the processing circuitry 127, converts the input operations received from the operator into electrical signals, and outputs the electrical signals to the processing circuitry 127.

[0022] For example, the input interface 123 is realized by a trackball, a switch button, a mouse, a keyboard, a touchpad that performs input operations by touching the operation surface, a touchscreen that integrates a display screen and a touchpad, a non-contact input circuit using an optical sensor, and a voice input circuit.

[0023] In this specification, the input interface 123 is not limited to an interface having physical operation parts such as a mouse, a keyboard, etc. For example, an example of the input interface 123 also includes an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the device and outputs this electrical signal to a control circuit.

[0024] The display 124 displays various types of information. Specifically, the display 124 is connected to the processing circuit 127 and displays various types of information output from the processing circuit 127. For example, the display 124 is realized by a liquid crystal monitor, a CRT (Cathode Ray Tube) monitor, a touch panel, or the like.

[0025] The microphone 125 records the voices uttered by the participants during the conference. Note that a plurality of microphones 125 may be provided. For example, one microphone 125 may be provided for each participant of the conference.

[0026] The processing circuitry 127 controls the components of the medical information processing device 120. Specifically, the processing circuitry 127 has a reception function 127a, a first acquisition function 127b, a second acquisition function 127c, a generation function 127d, an extraction function 127e, an identification function 127f, a calculation function 127g, and a presentation function 127h.

[0027] Here, the extraction function 127e is an example of an extraction unit, the identification function 127f is an example of an identification unit, the calculation function 127g is an example of a calculation unit and a reception unit, and the presentation function 127h is an example of a presentation unit and a notification unit.

[0028] The reception function 127a receives various operation inputs from the user. For example, the reception function 127a receives an input specifying a target patient who will be the target of a multi-disciplinary medical round from the user via the input interface 123. Specifically, the reception function 127a receives an input of the patient ID of the target patient from the user. Note that the reception function 127a may also receive an input from the user selecting the patient ID of the target patient from multiple patient IDs displayed in a list format.

[0029] Furthermore, for example, the reception function 127a receives an input to start a conference from the user. Specifically, the reception function 127a receives an input from the user pressing a conference start button displayed on the display 124.

[0030] Furthermore, for example, the reception function 127a receives an input to end the conference from the user. Specifically, the reception function 127a receives an input from the user pressing an end conference button displayed on the display 124.

[0031] The first acquisition function 127b acquires patient information about the target patient. For example, the first acquisition function 127b references the patient information DB 111 of the information integration system 110 via the NW interface 121 and the network 130, and acquires patient information corresponding to the patient ID accepted by the acceptance function 127a. Specifically, the first acquisition function 127b acquires information about the target patient, such as the department, name, age, number of days in the ICU, illness, purpose of admission, and treatment. Note that, in the following embodiments, ICU admission will be described as an example, but the number of days of hospitalization or the start date of a specific treatment may also be input.

[0032] The second acquisition function 127c acquires voice data that records the content of remarks about the patient made by multiple conference participants. Voice data is an example of recorded data. For example, when the reception function 127a receives an input to start the conference, the second acquisition function 127c acquires voice data recorded by the microphone 125. Furthermore, when the reception function 127a receives an input to end the conference, the second acquisition function 127c ends the voice data acquisition process. The voice data records the remarks of the participants separately for each speaker. This distinction between participants can be made by determining which microphone 125 collected the remarks if a microphone 125 is provided for each participant. Furthermore, it is also possible to distinguish which participant spoke from voice data containing a mixture of voices from multiple participants using voice recognition or the like.

[0033] In addition, if the conference is held in a chat format using a terminal device such as a smartphone or a tablet terminal, the second acquisition function 127c may acquire text data entered by the conference participants via the terminal device as recorded data.

[0034] The generation function 127d generates text data based on the voice data. For example, the generation function 127d generates the text data by converting the voice data acquired by the second acquisition function 127c into text data using a known voice recognition technology.

[0035] Here, Fig. 2 is a diagram illustrating an example of a process for generating text data. As shown in Fig. 2, text data TD is generated in which the statements of the meeting participants are expressed in text. In the example of Fig. 2, the text data TD is generated by converting the statements of the meeting participants, such as "CT scan scheduled for the morning, HD in the afternoon, and extubation," into text.

[0036] 1, the description will continue. The extraction function 127e extracts word elements from the generated text data. For example, the extraction function 127e uses a known natural language processing technique to divide a sentence included in the text data TD generated by the generation function 127d into words.

[0037] Here, Fig. 3 is a diagram illustrating an example of word segmentation. As shown in Fig. 3, a sentence included in text data TD is segmented into words. In the example of Fig. 3, the sentence "A CT scan is scheduled for the morning, ... HD is performed in the afternoon, ... and the tube is extubated" is segmented into words by dividing it into "morning", "to", "CT scan", "to", "schedule", ..., "afternoon", "to", "HD", "perform", ..., and "extubate".

[0038] Furthermore, for example, the extraction function 127e performs fluctuation correction for the divided words to unify synonyms into one expression.

[0039] FIG. 4 is a diagram illustrating an example of fluctuation correction. As shown in FIG. 4, a process is performed to unify words with the same meaning, such as "HD" and "hemodialysis," into a single expression. As an example, the extraction function 127e performs a process to unify synonyms into the expression that appears most frequently. In FIG. 4, a process is performed to convert "HD" into "hemodialysis." In the example of FIG. 4, "hemodialysis" becomes the word element WE, which is the target of the importance calculation described below. This process of unifying expressions can be performed by storing a dictionary file in which synonyms are listed in advance and determining whether the divided word corresponds to one of the synonyms.

[0040] The extraction function 127e extracts a plurality of word elements WE from the text data generated by the generation function 127d by performing the above-mentioned word division and fluctuation correction.

[0041] Returning to Figure 1, the description will continue. The identification function 127f identifies the word attributes of each extracted word element based on the concept indicated by each extracted word element. For example, the identification function 127f identifies the word attributes of each extracted word element by using a known medical ontology or the like to superordinately conceptualize each extracted word element.

[0042] As an example, when the word element "hemodialysis" is extracted, the identifying function 127f uses a known medical ontology to obtain hypernyms in the medical ontology, such as "hemodialysis" → "dialysis therapy" → "treatment." Similarly, when the word element "enteral nutrition" is extracted, the identifying function 127f obtains hypernyms, such as "enteral nutrition" → "nutritional management" → "treatment." Similarly, when the word element "CT" is extracted, the identifying function 127f obtains hypernyms, such as "CT" → "diagnosis."

[0043] Similarly, when the word element "respiratory rate" is extracted, the identifying function 127f acquires a hypernym such as "respiratory rate" → "vital signs" → "physical examination." Similarly, when the word element "heart failure" is extracted, the identifying function 127f acquires a hypernym such as "heart failure" → "disease." Similarly, when the word element "edema" is extracted, the identifying function 127f acquires a hypernym such as "edema" → "symptoms."

[0044] Here, Fig. 5 is a diagram illustrating an example of the attribute identification process. As shown in Fig. 5, the identification function 127f uses a known medical ontology to acquire a hypernym "diagnosis" on the medical ontology for the word element WE "CT examination." Similarly, for the word element WE "heart failure," it acquires a hypernym "disease." Similarly, for the word element WE "hemodialysis," it acquires a hypernym "treatment."

[0045] Then, the identifying function 127f identifies the attribute of "CT examination" as "diagnosis," the attribute of "heart failure" as "disease," and the word attribute of "hemodialysis" as "treatment" according to the conversion result.

[0046] Returning to Figure 1, the explanation will continue. The calculation function 127g calculates the importance of a word element. For example, the calculation function 127g calculates the importance using as an index the frequency of appearance of the word element, the number of times a medical procedure related to the word element has been included in a ToDo list in the past, etc.

[0047] Specifically, the calculation function 127g calculates the importance such that the higher the frequency of appearance of the word element during the meeting, the higher the importance value. Also, the calculation function 127g analyzes past data using a statistical method and calculates the importance such that the more times a medical procedure related to the word element has been included in the ToDo list in the past, the higher the importance value.

[0048] Furthermore, for example, the calculation function 127g corrects the calculated importance of a word element by taking into account the word attributes of the word element, the speaker attributes of the speaker related to the word element, the temporal relationship between the time related to the utterance content including the word element and the reference time, medical resource information of the medical resource related to the word element, and the severity of the underlying disease if the word element is a disease. In the following embodiments, an example will be described in which the importance is calculated based on the appearance frequency of the word element and then corrected, but this is not limited to this. Instead of the appearance frequency, the importance may be calculated directly using the word attributes of the word element, the speaker attributes, the temporal relationship, the medical resource information, and the severity.

[0049] In this embodiment, the calculation function 127g sets a first weight used to calculate the importance according to the word attribute of the word element. Specifically, the calculation function 127g analyzes past data using a statistical method, and sets the first weight so that the value of the first weight increases as the number of times that a medical practice related to a word element of the word attribute has been included in the ToDo list in the past increases.

[0050] Here, the speaker attribute is information that indicates the attribute of the speaker who made a statement related to the extracted word element. For example, the speaker attribute is information that indicates the speaker's occupation, such as attending physician, ICU doctor, ICU nurse, pharmacist, dietitian, physical therapist, or clinical engineer.

[0051] It should be noted that the identification of the speaker can utilize known voice recognition technology, etc. For example, the calculation function 127g uses a statistical method, a machine learning method, etc. to extract features of the speaker's voice, such as the pitch of the voice, the tone of the voice, the speaking speed, and speaking habits, from the voice data acquired by the second acquisition function 127c, and identifies the speaker based on the extracted features.

[0052] In addition, the calculation function 127g may identify the speaker by providing multiple microphones 125, associating the speaker's identification information with each microphone 125 before the start of the conference, and identifying which microphone 125 the audio data was recorded through.

[0053] In this embodiment, the calculation function 127g sets a second weight used to calculate the importance according to the speaker attribute. Specifically, the calculation function 127g analyzes past data using a statistical method, and sets the second weight so that the value of the second weight increases as the number of times that a medical procedure related to a word element spoken by a speaker with the speaker attribute has been included in a ToDo list in the past increases.

[0054] By calculating the importance by taking into account speaker attributes in this way, for example, even if a speaker whose job is not very involved in setting up a ToDo list frequently utters words related to medical procedures during a meeting, the medical procedures based on those words are less likely to be presented in the ToDo list. Therefore, the medical information processing device 120 according to this embodiment can present more appropriate medical procedures in the ToDo list.

[0055] The time relationship is information that indicates the time relationship between a time related to the extracted word element and a reference time. For example, the reference time is the time point at which the importance is calculated.

[0056] For example, if text data containing the sentence "Hemodialysis was performed yesterday" is generated, "yesterday" will be the time relation of the word element "hemodialysis." Similarly, if there is text data containing "A CT scan is scheduled for this afternoon," "this afternoon" will be the time information of the word element "CT scan." Similarly, if there is text data containing "Extubate after the patient's condition stabilizes," "after the patient's condition stabilizes" will be the time information of the word element "Extubate."

[0057] In this embodiment, the calculation function 127g sets the third weight used to calculate the importance according to the time information. Specifically, if the time information represents a time earlier than the time point at which the importance is calculated, the calculation function 127g sets the value of the third weight to a small value.

[0058] In addition, when the time information represents a time in the future relative to the time point at which the importance is calculated, the calculation function 127g sets the third weight so that the value of the third weight becomes larger the closer the time represented by the time information is to the time point at which the importance is calculated.

[0059] Furthermore, when the time information includes a condition, the calculation function 127g sets the third weight according to the achievement status of the condition.

[0060] Specifically, when the time information includes the condition "after the patient's condition stabilizes," the calculation function 127g identifies the patient's condition based on the text data generated by the generation function 127d. When a word element with a context indicating that the patient's condition is not stable appears in the text data with a frequency equal to or greater than a predetermined threshold, the calculation function 127g sets the third weight so that the value of the third weight becomes smaller. On the other hand, when a word element with a context indicating that the patient's condition is stable appears in the text data with a frequency equal to or greater than a predetermined threshold and it can be inferred that the patient's condition is likely to stabilize soon or that the patient's condition is already stable, the calculation function 127g sets the third weight so that the value of the third weight becomes larger.

[0061] By calculating the importance taking into account time relationships or future predictions in this way, even if a word related to a medical procedure that has already been performed appears frequently, the likelihood of a medical procedure based on that word being presented on the ToDo list is reduced. Therefore, the medical information processing device 120 according to this embodiment can present more appropriate medical procedures on the ToDo list.

[0062] In addition, if it is not possible to determine from the text data acquired by the second acquisition function 127c whether the conditions for presenting it as a Todo list are met, the calculation function 127g may have the user (conference participant) input whether the conditions are met or not after the end button is pressed.

[0063] The medical resource information is information about the medical resources held by the medical facility. For example, the medical resource information is information that indicates the relationship between the amount of medical resources (manpower, supplies, rooms, bed reservations, medical expenses, etc.) at the time of calculating the importance and the amount of required medical resources that is preset for each word element that can be a ToDo candidate.

[0064] In this embodiment, the calculation function 127g sets a fourth weight used in calculating the importance level according to the amount of medical resources.

[0065] Specifically, when the difference between the amount of medical resources at the time of calculating the importance and the amount of required medical resources is less than a predetermined threshold, the calculation function 127g sets the value of the fourth weight to a small value, whereas when the difference between the amount of medical resources at the time of calculating the importance and the amount of required medical resources is equal to or greater than a predetermined threshold, the calculation function 127g sets the value of the fourth weight to a large value.

[0066] By calculating the importance by taking medical resource information into account in this way, for example, medical procedures that cannot be performed due to a lack of medical resources such as manpower are less likely to be presented on the ToDo list. Therefore, the medical information processing device 120 according to this embodiment can present more appropriate medical procedures on the ToDo list. Furthermore, if the extracted word elements represent information about the name of a disease or a condition, the severity of the patient's condition can be extracted based on the word elements, and a higher weight can be set for presentation on the ToDo list. This severity can be determined by storing a dictionary file in advance that associates word elements with severity and referencing the dictionary file.

[0067] Here, Fig. 6 is a diagram illustrating an example of the calculation result of importance. In Fig. 6, the horizontal axis represents the attribute, and the vertical axis represents the importance. In the example of Fig. 6, the importance is shown to be increasing in the order of "hemodialysis" with the word attribute "treatment," "CT examination" with the word attribute "diagnosis," and "heart failure" with the word attribute "disease."

[0068] In this embodiment, the calculation function 127g corrects the importance of the calculated word element by taking into account all of the word attributes, speaker attributes, temporal relationships, and medical resource information, but the importance may also be corrected by taking into account any one of these, or a combination of any two or three of these.

[0069] Furthermore, in this embodiment, the calculation function 127g calculates the importance for all word attributes, but it may also calculate the importance only for word attributes that represent medical procedures that can be presented as a ToDo list, such as "treatment" and "diagnosis."

[0070] Returning to Figure 1, the explanation will be continued. The presentation function 127h presents candidates for medical procedures to be performed on a patient based on word elements corresponding to the calculated importance. For example, the presentation function 127h presents medical procedures related to a predetermined number (e.g., three) word elements in descending order of importance by displaying them as a ToDo list on the display 124.

[0071] Specifically, the presentation function 127h presents word attributes that represent medical procedures that can be presented as a ToDo list, such as "treatment" and "diagnosis," as a ToDo list as is. On the other hand, word attributes that cannot be presented as a ToDo list as is are converted into expressions that represent medical procedures and then presented as a ToDo list.

[0072] For example, in the case of "hypotension" whose word attribute is "symptom," the presentation function 127h uses a known natural language processing technique to identify a predicate for "hypotension" based on a dependency relationship from the text data generated by the generation function 127d. For example, if the identified predicate is "improve," the presentation function 127h presents a medical action, "improve hypotension," as a ToDo list.

[0073] In addition, in the case of "heart failure" whose word attribute is "disease," the presentation function 127h identifies a predicate for "heart failure" in the same manner as described above. For example, if the identified predicate is "evaluate," the presentation function 127h presents a medical action, "evaluate heart failure," as a ToDo list.

[0074] As another example, the presentation function 127h may present co-occurring medical procedures based on the co-occurrence relationship of word elements as a to-do list. For example, if "hypotension" and "intubation" are co-occurring, the presentation function 127h may present the medical procedure "intubation" together with the word element "hypotension" as a to-do list.

[0075] Here, Fig. 7 is a diagram showing an example of a ToDo list presentation screen RD. The presentation screen RD in Fig. 7 presents a ToDo list for the target patient for each date on which a meeting was held. The presentation screen RD in Fig. 7 has a patient information display field PD, a ToDo list display field TL, and a comment display field RC.

[0076] The patient information display field PD is a display field that displays the department, patient name, age, number of days admitted, disease, purpose of admission, and current treatment acquired by the first acquisition function 127b as patient information of the target patient.

[0077] The medical department is information that indicates the medical department of the target patient. The patient name is information that indicates the name of the target patient. The age is information that indicates the age of the target patient. The number of days of admission is information that indicates the number of days the target patient has been admitted to the ICU. The disease is information that indicates the disease that the target patient is suffering from. The purpose of admission is information that indicates the reason the target patient has been admitted to the ICU. The current treatment is information that indicates the treatment currently being given to the target patient.

[0078] The ToDo list display field TL is a display field that displays the ToDo list presented by the presentation function 127h. In the example of Fig. 7, medical procedures related to three word elements in descending order of importance, "hemodialysis," "consider laxatives," and "nutritional management," are displayed as a ToDo list. Furthermore, when a meeting is taking place, the presentation function 127h updates the ToDo list in real time.

[0079] Specifically, if the importance of a word element calculated in real time by the calculation function 127g exceeds the importance of at least one of the word elements that are the basis for the medical procedures currently displayed as a ToDo list, the ToDo list is updated so that the medical procedures related to the three word elements with the highest current importance are displayed.

[0080] The comment display field RC is a display field that displays comments made by participants in a conference. Comments CO1 to CO5 are displayed in the comment display field RC in association with information S1 to S5 that indicates the speaker. In the example of Fig. 7, comment CO1 indicates a comment made by an ICU doctor S1, comment CO2 indicates a comment made by the attending physician S2 of the medical department, comment CO3 indicates a comment made by a nurse S3, comment CO4 indicates a comment made by a pharmacist S4, and comment CO5 indicates a comment made by a dietitian S5.

[0081] If a comment no longer fits in the comment display field RC, the user can operate the slider bar SB to display comments made before the comment displayed in the comment display field RC. Also, when a conference is in progress, the text data generated by the generation function 127d is displayed in real time together with information indicating the speaker identified by the calculation function 127g as described above.

[0082] The generating function 127d, the extracting function 127e, the identifying function 127f, the calculating function 127g, and the presenting function 127h perform the above-mentioned processing on the voice data acquired by the second acquiring function 127c from the time when the receiving function 127a receives the input to start the conference until the time when the receiving function 127a receives the input to end the conference. After the series of processing is completed, the processing circuit 127 stores the ToDo list in the ToDo list display field TL and the comments in the comment display field RC as a rounds record in the rounds record DB 122a.

[0083] Next, a description will be given of the processing executed by the medical image processing apparatus 120 according to this embodiment.

[0084] First, the first acquisition function 127b acquires the patient information of the target patient (step S101).

[0085] For example, the reception function 127a receives input of the patient ID of the target patient from the user. The first acquisition function 127b refers to the patient information DB 111 of the information integration system 110 via the NW interface 121 and the network 130, and acquires the patient information corresponding to the patient ID accepted by the reception function 127a. The patient information acquired by the first acquisition function 127b is displayed by the processing circuit 127 on the ToDo list presentation screen.

[0086] Next, the reception function 127a receives an input to start the conference (step S102). For example, the reception function 127a receives an input of pressing a conference start button displayed on the display 124 via the input interface 123.

[0087] Next, the second acquisition function 127c acquires voice data (step S103). For example, the second acquisition function 127c acquires voice data recorded by the microphone 125. Next, the generation function 127d generates text data (step S104). For example, the second acquisition function 127c converts the voice data acquired in step S103 into text data using a known voice recognition technique.

[0088] Next, the extraction function 127e extracts word elements from the text data (step S105). For example, the extraction function 127e uses a known natural language processing technique to divide the sentences included in the text data generated in step S104 into words. The extraction function 127e also performs fluctuation correction on the divided words to unify synonyms into the same expression.

[0089] Next, the identifying function 127f identifies word attributes of the word elements (step S106). For example, the identifying function 127f uses a known medical ontology to acquire hypernyms in the medical ontology for each of the word elements extracted in step S104. The identifying function 127f identifies each of the acquired hypernyms as word attributes of each of the word elements.

[0090] Next, the calculation function 127g calculates the importance of the word element (step S107). For example, the calculation function 127g calculates the importance based on indicators such as the appearance frequency of the word element and the number of times a medical procedure related to the word element has been adopted in a to-do list. In addition, the calculation function 127g corrects the importance based on the word attribute of the word element, the speaker attribute related to the word element, the temporal relationship, and medical resource information.

[0091] Next, the presentation function 127h determines whether the importance calculated in step S106 exceeds the importance of the word element that is the basis of the medical action currently presented as a ToDo list (step S108). If the importance does not exceed the importance of the word element that is the basis of the medical action currently presented as a ToDo list (step S108: No), the process proceeds to step S110, which will be described later.

[0092] On the other hand, if the importance of the word element that is the basis of the medical action currently presented as the ToDo list is exceeded (step S108: Yes), the presentation function 127h updates the ToDo list (step S109). For example, the presentation function 127h updates the ToDo list so that medical actions related to the three word elements with the highest current importance are displayed.

[0093] Next, the reception function 127a determines whether an input to end the conference has been received (step S110). For example, the reception function 127a determines whether an input to press the conference end button displayed on the display 124 has been received via the input interface 123. If an input to end the conference has not been received (step S110: No), the process returns to the processing of step S103 described above.

[0094] On the other hand, if an input to end the conference is received (step S110: Yes), the processing circuitry 127 stores the rounds record (step S111) and ends this process. For example, the processing circuitry 127 performs a process of storing the ToDo list displayed on the display screen and the content of the speaker's remarks as a rounds record in the rounds record DB 122a of the storage circuitry 122. When the ToDo list is displayed on the display screen, information that served as the basis for listing each item on the ToDo list may be displayed. For a certain item in the ToDo list, word elements with high importance that served as the basis for extracting the item may be extracted and displayed next to the item. Furthermore, relevant parts of the audio data or text data of the conversation from which the word elements were extracted may be displayed side by side. This makes it possible to understand what conversation was the basis for suggesting the ToDo list item.

[0095] As described above, the medical information processing device 120 according to the embodiment generates text data from audio data recording the content of statements made by multiple speakers regarding a subject, extracts word elements from the generated text data, identifies the word attributes of each of the extracted word elements based on the concept indicated by each of the extracted word elements, calculates the importance of each of the word elements based on the word attributes, the speaker attributes of the speakers related to the word elements, temporal relationships, and medical resource information, and presents candidates for medical procedures to be performed on the patient based on the word elements according to their importance.

[0096] This allows the importance of word elements to be calculated taking into account word attributes, speaker attributes of speakers related to the word elements, temporal relationships, and medical resource information. This reduces the possibility of, for example, presenting a medical procedure that has already been performed as a candidate for a patient, or presenting a medical procedure that is impossible to perform due to a lack of manpower, etc., as a candidate for a patient. Therefore, the medical information processing device 120 according to this embodiment can present important information for determining the next medical procedure for a patient in an easy-to-understand format. Furthermore, presenting information in an easy-to-understand format facilitates information recording and allows efficient information sharing among multiple professions. In other words, the medical information processing device 120 according to this embodiment can efficiently record and share important information for determining the next medical procedure to be performed on a patient.

[0097] The above-described embodiment can be modified as needed by partially changing the configuration or functions of each device. Therefore, several modifications of the above-described embodiment will be described below as other embodiments. The following mainly focuses on differences from the above-described embodiment, and detailed descriptions of commonalities with the content already described will be omitted. The modifications described below may be implemented individually or in appropriate combination.

[0098] (Variation 1) In the above embodiment, the presentation function 127h presents only information about medical procedures as a ToDo list. In this modified example, a ToDo list is presented that presents information about "who," "when," and "whether the medical procedure will be performed."

[0099] In this modified example, when calculating the importance, the calculation function 127g analyzes the text data generated by the generation function 127d using known natural language processing technology, and identifies "who" and "when" will perform (or have performed) the medical procedure related to the word element for which the importance is to be calculated.

[0100] The calculation function 127g then compares the amount of medical resources at the time corresponding to the specified "when" with the amount of required medical resources, and sets a fourth weight based on the medical resource information. This makes it possible to more accurately evaluate the amount of medical resources and then calculate the importance of the word element.

[0101] Furthermore, if the calculation function 127g cannot identify at least one of "who" and "when" from the text data, the presentation function 127h presents the medical procedure as a confirmation-required ToDo list, separate from the normal ToDo list, with the "who" and "when" fields left blank and with input fields added, to prompt the user to enter information.

[0102] In this case, the reception function 127a receives input of "who" and "when" from the user, and the calculation function 127g calculates the importance of the word element according to the input. Once the importance is calculated, the presentation function 127h executes a process to delete the medical practice related to the word element from the confirmation ToDo list.

[0103] The presentation function 127h may also notify the user to check the schedule if a medical procedure that shares medical resources with the target patient is scheduled to be performed on a patient other than the target patient during the time period corresponding to the specified or input "when." This allows the user to review the schedule of the medical procedure depending on the condition of the target patient and other patients.

[0104] According to this modification, the next medical procedure to be performed on the patient can be presented in a more specific manner.

[0105] (Variation 2) In the above embodiment, a form has been described in which the medical information processing device 120 executes the above-mentioned processing via its own input interface 123, display 124, and microphone 125. In this modified example, a form will be described in which the above-mentioned processing is executed by a terminal device used by a conference participant and a medical information processing device communicably connected to the terminal device.

[0106] In this modification, the terminal device has the same configuration as the input interface 123, the display 124, and the microphone 125 described in the above embodiment as components of the medical information processing device 120. For example, the terminal device may be realized by a smartphone, a tablet terminal, a personal computer, a small dedicated device similar to a smartphone, or the like.

[0107] Furthermore, the medical information processing device executes, via a terminal device, the processes that are executed in the above-described embodiment via the input interface 123, the display 124, and the microphone 125. For example, the medical information processing device is realized by computer equipment such as a server or a workstation.

[0108] The terminal device and the medical information processing device are connected to each other so that they can communicate with each other via the Internet, for example. In this case, the medical information processing system may be constructed in the form of cloud computing, in which the medical information processing device provides services in response to requests from the terminal device.

[0109] Furthermore, the above-described embodiment has described an example in which the acquisition unit, extraction unit, identification unit, calculation unit, and presentation unit in this specification are realized by the second acquisition function 127c, extraction function 127e, identification function 127f, calculation function 127g, and presentation function 127h of the processing circuit 127, respectively, but the embodiment is not limited to this.

[0110] For example, the acquisition unit, extraction unit, identification unit, calculation unit, and presentation unit in this specification may be realized by the second acquisition function 127c, extraction function 127e, identification function 127f, calculation function 127g, and presentation function 127h of the processing circuit 127, respectively, or the same functions may be realized by hardware only, software only, or a combination of hardware and software.

[0111] In the above-described embodiment, the processing circuitry 127 is implemented by a single processor, but the embodiment is not limited to this. For example, the processing circuitry 127 may be configured by combining multiple independent processors, and each processor may execute a program to implement each processing function. Furthermore, each processing function of the processing circuitry 127 may be implemented by being distributed or integrated as appropriate across a single or multiple processing circuits.

[0112] Furthermore, each processing function of the processing circuitry 127 may be realized by a combination of hardware and software, such as circuits. Although an example in which programs corresponding to each processing function are stored in a single storage circuitry 122 has been described here, the embodiment is not limited to this. For example, the programs corresponding to each processing function may be stored in a distributed manner in multiple storage circuits, and the processing circuitry 127 may read and execute each program from each storage circuit.

[0113] Furthermore, the term "processor" used in the description of the above-mentioned embodiments refers to circuits such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an Application Specific Integrated Circuit (ASIC), a programmable logic device (e.g., a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), and a Field Programmable Gate Array (FPGA)).

[0114] Instead of storing the program in a memory circuit, the program may be directly embedded in the processor circuit. In this case, the processor realizes its function by reading and executing the program embedded in the circuit. Furthermore, 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.

[0115] Here, the program executed by the processor is provided by being pre-installed in a ROM (Read Only Memory), a storage circuit, etc. Note that this program may also be provided by being recorded in a computer-readable, non-transitory storage medium such as a CD (Compact Disk)-ROM, a FD (Flexible Disk), a CD-R (Recordable), or a DVD (Digital Versatile Disk) in a format that can be installed or executed by these devices.

[0116] This program may also be provided or distributed by being stored on a computer connected to a network such as the Internet and downloaded via the network. For example, this program may be composed of modules including each of the processing functions described above. In actual hardware, a CPU reads and executes the program from a storage medium such as a ROM, whereby each module is loaded into a main memory and generated on the main memory.

[0117] Furthermore, in the above-described embodiment and modified examples, the components of each device shown in the drawings are functional concepts and do not necessarily have to be physically configured as shown in the drawings. In other words, the specific form of distribution or integration of each device is not limited to that shown in the drawings, and all or part of them can be functionally or physically distributed or integrated in any unit depending on various loads, usage conditions, etc.

[0118] Furthermore, all or any part of the processing functions performed by each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware using wired logic.

[0119] Furthermore, among the processes described in the above-mentioned embodiments and modifications, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using a known method.In addition, the information including the processing procedures, control procedures, specific names, various data and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified.

[0120] The various data handled in this specification are typically digital data.

[0121] According to at least one of the embodiments described above, important information for determining the next medical procedure to be performed on a subject can be efficiently recorded and shared.

[0122] Although several embodiments 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, modifications, and combinations of embodiments 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 and its equivalents as defined in the claims. [Explanation of symbols]

[0123] 100 Medical Information Processing System 120 Medical information processing device 122 Memory circuit 122a Medical rounds record DB 127 Processing Circuit 127a Reception function 127b 1st acquisition function 127c 2nd acquisition function 127d generation function 127e Extraction Function 127f Specific Functions 127g calculation function 127h Presentation function

Claims

1. an extraction unit that extracts word elements from recorded data that records the contents of statements made by a plurality of speakers regarding the subject; an identifying unit that identifies word attributes of each of the extracted word elements based on a concept indicated by each of the extracted word elements; a calculation unit that calculates the importance of each of the word elements based on at least one of the word attributes, the speaker attributes of the speaker related to the word elements, the time relationship between a time related to the utterance content including the word elements and a reference time, and medical resource information of the medical resource related to the word elements; a presentation unit that presents candidates for medical procedures to be performed on the subject based on the importance of each of the word elements; A medical information processing device comprising:

2. the presentation unit displays, on a display device, information representing a predetermined number of the medical treatment candidates in descending order of importance; The medical information processing device according to claim 1 .

3. the calculation unit calculates the importance only for the word elements of the attribute that may be candidates for the medical treatment. The medical information processing device according to claim 1 .

4. the calculation unit calculates the importance for all of the word elements, the presentation unit presents candidates for the medical treatment derived from the word element of the attribute that cannot be a candidate for the medical treatment; The medical information processing device according to claim 1 .

5. the calculation unit calculates the importance of the word element by taking into account the frequency of appearance of the word element and the number of times that candidates for medical procedures related to the word element have been presented. The medical information processing device according to claim 4 .

6. the presentation unit identifies a predicate element for the word element of the attribute that cannot be a candidate for the medical action from a dependency relationship of words, and presents the candidate for the medical action derived from the word element and a preceding descriptive word element; The medical information processing device according to claim 4 .

7. the presenting unit presents candidates for the medical treatment derived from the word elements of the attribute that cannot be candidates for the medical treatment based on a co-occurrence relationship of the word elements; The medical information processing device according to claim 4 .

8. the calculation unit sets a weight for calculating the importance level based on at least one of the word attribute, the speaker attribute, the temporal relationship, and the medical resource information, and calculates the importance level using the weight. The medical information processing device according to any one of claims 1 to 7.

9. the calculation unit sets the weight so that the weight increases as the speaker attribute has uttered the word element related to the presented medical treatment candidate more frequently. The medical information processing device according to claim 8 .

10. the calculation unit sets the weight so that the weight is smaller for the word element whose temporal relationship is earlier than the present time, and the weight is larger for the word element whose temporal relationship is later than the present time as the word element is closer to the present time. The medical information processing device according to claim 8 .

11. the calculation unit sets the weight for the word element that is the basis of the medical action candidate to be performed when a specific condition is satisfied, in accordance with the current achievement status of the condition; The medical information processing device according to claim 10.

12. the medical resource information includes information indicating the amount of required medical resources set for each of the word elements; the calculation unit compares the current amount of medical resources with the amount of medical resources required identified from the word elements, and sets the weight for a word element for which the difference between the current amount of medical resources and the amount of medical resources required identified from the word elements is less than a threshold value so that the weight is smaller than that for a word element for which the difference is equal to or greater than the threshold value; The medical information processing device according to claim 8 .

13. The medical resources include human resources; When calculating the importance, the calculation unit analyzes the record data, identifies the person who will perform the medical procedure related to the word element and the time when the procedure will be performed, and compares the amount of medical resources at the time corresponding to the time of the procedure with the amount of medical resources required identified from the word element based on the identified person who will perform the procedure and the time when the procedure will be performed, and sets the weight for word elements for which the difference between the amount of medical resources at the time corresponding to the time of the procedure and the amount of medical resources required identified from the word element is less than a threshold value so that the weight is smaller than that for word elements for which the difference is equal to or greater than the threshold value. The medical information processing device according to claim 8 .

14. a notification unit that, when at least one of the person who will perform the activity and the time when the activity will be performed cannot be identified from the recorded data, issues a notification to the user to prompt the user to input information corresponding to the unidentified item of the person who will perform the activity and the time when the activity will be performed; a receiving unit that receives input of information corresponding to the unspecified item; Further provided with The calculation unit identifies at least one of the person who will perform the activity and the time when the activity will be performed, based on the received input. The medical information processing device according to claim 13 .

15. the notification unit issues a notification urging a user to check a schedule when the medical procedure, which shares the medical resources, is scheduled to be performed on a subject other than the subject during the time period corresponding to the identified implementation time. The medical information processing device according to claim 14.

16. an extraction unit that extracts word elements from recorded data that records the contents of statements made by a plurality of speakers regarding the subject; an identifying unit that identifies word attributes of each of the extracted word elements based on a concept indicated by each of the extracted word elements; a calculation unit that calculates the importance of each of the word elements based on at least one of the word attributes, the speaker attributes of the speaker related to the word elements, the time relationship between a time related to the utterance content including the word elements and a reference time, and medical resource information of the medical resource related to the word elements; a presentation unit that presents candidates for medical procedures to be performed on the subject based on the importance of each of the word elements; A medical information processing system comprising:

17. A medical information processing method by a medical information processing device, comprising: an extraction step of extracting word elements from recorded data recording the contents of statements made by a plurality of speakers regarding the subject; an identifying step of identifying word attributes of each of the extracted word elements based on a concept indicated by each of the extracted word elements; a calculation step of calculating the importance of each of the word elements based on at least one of the word attributes, the speaker attributes of the speaker related to the word elements, the time relationship between a time related to the utterance content including the word elements and a reference time, and medical resource information of the medical resource related to the word elements; a presentation step of presenting candidates for medical procedures to be performed on the subject based on the importance of each of the word elements; A medical information processing method including:

Images