JP2023157873A - Medical information processing apparatus and program - Google Patents

Abstract

To support quick decision making in team medical care.SOLUTION: A medical information processing apparatus according to an embodiment has a message acquisition unit, a role acquisition unit, a progress acquisition unit, a candidate specification unit, and a display control unit. The message acquisition unit acquires, in a channel opened in a chat application for discussing a target medical care, a message history that is the history of messages already input by a plurality of members of the channel. The role acquisition unit acquires role information representing a role of a target member of the plurality of members. The progress acquisition unit acquires progress information representing the current progress of the target medical care. The candidate specification unit specifies one or more candidates related to the messages input by the target member based on the message history, the role information and/or the progress information. The display control unit displays the one or more candidates on an input screen related to the channel displayed for the target member.SELECTED DRAWING: Figure 2

Description

Embodiments disclosed in this specification and drawings relate to a medical information processing device and a program.

For diseases such as stroke and pulmonary embolism, international guidelines recommend team medical care in which a care team consisting of multiple departments is formed to make medical decisions and provide treatment. To support team medical care, applications are provided to share medical status and make decisions within the care team. Some applications have also appeared that have a many-to-many message exchange (chat) function. However, it has been pointed out that inputting character strings for chatting may be cumbersome depending on the medical situation, the user's personality, etc. In diseases where every minute counts, such as stroke or pulmonary embolism, the complexity of inputting information can be a factor that prevents the use of chat.

Japanese Patent Application Publication No. 2020-149131 Japanese Patent Application Publication No. 2015-219719 International Publication No. 2020/050109

One of the problems that the embodiments disclosed in this specification and drawings are intended to solve is to support rapid decision-making in team medical care. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problems. Problems corresponding to the effects of each configuration shown in the embodiments described later can also be positioned as other problems.

The medical information processing device according to the embodiment includes a message acquisition section, a role acquisition section, a progress acquisition section, a candidate identification section, and a display control section. The message acquisition unit acquires a message history that is a history of messages already input by a plurality of members of a channel opened in a chat application for discussing the target medical treatment. The role acquisition unit acquires role information representing a role of a target member among the plurality of members. The progress acquisition unit acquires progress information representing the current progress of the target medical treatment. The candidate identifying unit identifies one or more candidates regarding the message input by the target member, based on the message history, the role information, and/or the progress information. The display control unit displays the one or more candidates on an input screen regarding the channel of the target member.

FIG. 1 is a diagram showing an example of a network configuration of a medical information system according to a first embodiment. FIG. 2 is a diagram showing a configuration example of the medical information processing device according to the first embodiment. FIG. 3 is a diagram showing an example of input and output of the first model. FIG. 4 is a diagram illustrating an example of a processing procedure of a chat application by the medical information processing device according to the first embodiment. FIG. 5 is a diagram illustrating an example of message candidate identification processing using the second model and the third model. FIG. 6 is a diagram illustrating an example of transition of the input screen of the chat application according to the first embodiment. FIG. 7 is a diagram illustrating an example of transition of the input screen of the chat application according to the second embodiment. FIG. 8 is a diagram illustrating an example of transition of the input screen of the chat application according to the third embodiment. FIG. 9 is a diagram illustrating an example of transition of the input screen of the chat application according to the fourth embodiment. FIG. 10 is a diagram illustrating an example of transition of the input screen of the chat application according to the fifth embodiment. FIG. 11 is a diagram illustrating a configuration example of a medical information processing apparatus according to the second embodiment. FIG. 12 is a diagram illustrating an example of a processing procedure of a chat application by the medical information processing device according to the second embodiment. FIG. 13 is a diagram illustrating the process of obtaining medical information using push notifications. FIG. 14 is a diagram illustrating the process of obtaining medical information through periodic polling. FIG. 15 is a diagram illustrating the medical information acquisition process in stepwise processing. FIG. 16 is a diagram illustrating a display example of the progress status regarding acquisition of medical information.

Hereinafter, embodiments of a medical information processing device and a program will be described in detail with reference to the drawings.

(First embodiment)
FIG. 1 is a diagram showing an example of a network configuration of a medical information system 100 according to the first embodiment. The medical information system 100 is a computer network system for sharing target medical treatment among members of a care team. Target medical treatment means the medical treatment targeted by the care team. The medical treatment according to this embodiment is a general term for medical examination, diagnosis, and treatment. As shown in FIG. 1, the medical information system 100 includes an in-hospital information system 1, a medical information processing device 2, and N mobile terminals 3-n (n is a subscript representing each mobile terminal. n≦1≦N, N≧1.). The in-hospital information system 1, the medical information processing device 2, and the N mobile terminals 3-n are connected via a network so that they can communicate information with each other according to any standard such as DICOM (digital imaging and communications in medicine). There is. Hereinafter, when the mobile terminals 3-1, 3-2, . . . , 3-N are not particularly distinguished, they will be collectively referred to as the mobile terminal 3.

The hospital information system 1 is a computer network system installed in a medical facility such as a hospital. The in-hospital information system 1 manages a variety of information (hereinafter referred to as medical information) related to medical treatment performed at the medical facility. The hospital information system 1 includes hospital information systems (HIS), picture archiving and communication systems (PACS), electronic medical recording systems (EMR), and radiology information systems (RIS). It is composed of various systems such as radiology information systems).

The medical information processing device 2 is a computer that executes an application (hereinafter referred to as a medical treatment sharing application) for sharing target medical treatment among team members forming a care team. The medical information processing device 2 receives medical information regarding the target medical treatment from the in-hospital information system 1. The medical information processing device 2 executes the medical sharing application according to various commands from the mobile terminal 3. The medical information processing device 2 provides the input screen data of the medical treatment sharing application to the mobile terminal 3. By the medical treatment sharing application, medical information regarding the target medical treatment is shared with team members forming the care team through the mobile terminal 3. The medical information processing device 2 can also execute an application (hereinafter referred to as a chat application) for exchanging messages between team members forming a care team, which is incorporated into a medical treatment sharing application.

The mobile terminal 3 is a computer carried by each team member making up the care team. The mobile terminal 3 is assumed to be a smartphone, but it may be any portable computer such as a smart watch, a notebook computer, a tablet terminal, or the like, as long as it is capable of installing a medical sharing application and a chat application. The mobile terminal 3 includes a processor, an input device, a display device, and a communication device. The mobile terminal 3 functions as an input/output interface for the medical information processing device 2. As an example, the mobile terminal 3 transmits various commands to the medical information processing device 2, and displays an input screen provided from the medical information processing device 2 in response to the commands.

FIG. 2 is a diagram showing an example of the configuration of the medical information processing device 2. As shown in FIG. As shown in FIG. 2, the medical information processing device 2 includes a processing circuit 21, a storage device 22, and a communication device 23. The processing circuit 21, the storage device 22, and the communication device 23 are connected to each other via a bus so that they can input and output signals.

The processing circuit 21 includes a processor such as a CPU (Central Processing Unit). The processing circuit 21 implements a chat function 211, a message acquisition function 212, a role acquisition function 213, a progress acquisition function 214, a candidate identification function 215, an order placement function 216, a display control function 217, etc. by executing the medical care sharing application. do. Note that each of the functions 211 to 217 is not limited to being implemented by a single processing circuit. A processing circuit may be configured by combining a plurality of independent processors, and each of the functions 211 to 217 may be realized by each processor executing a medical information processing program. Furthermore, the functions 211 to 217 may be modularized programs that each constitute a medical information processing program. These applications and/or programs are stored in the storage device 22. Note that the medical care sharing application and the chat application are included in the medical information processing program.

By realizing the chat function 211, the processing circuit 21 exchanges messages between a plurality of members of a channel for discussing the target medical treatment via the chat application. A channel is an area in a chat app where messages are exchanged. Within the chat app, multiple channels have been opened for multiple topics. A plurality of members are registered in each channel, and the registered members can exchange messages in the channel. Message exchange is synonymous with chat.

By implementing the message acquisition function 212, the processing circuit 21 acquires a message history that is a history of messages already input by a plurality of members of the channel. The message history is stored in the storage device 22.

By realizing the role acquisition function 213, the processing circuit 21 acquires role information representing the role of the target member among the plurality of members. The target member means a member to whom a message candidate, which will be described later, is presented. A role means the type and authority of a member. More specifically, the role means the affiliation or specialty of the member, and examples thereof include a neurologist, a neurointerventionalist, an emergency physician, and a neuroradiologist. The role information may be acquired from the in-hospital information system 1, for example. As another example, the role information may be stored in or retrieved from the storage device 22.

By implementing the progress acquisition function 214, the processing circuit 21 acquires progress information representing the current progress of the target medical treatment. The current progress of the target medical treatment is represented by the status (stage) at which the target medical treatment is currently located in the process of examination, diagnosis, and treatment. The current progress of the target treatment is not limited to being represented by the three statuses of diagnosis, diagnosis, and treatment, but may be represented by statuses that are subdivided or derived from each of the above stages. As an example, the statuses that belong to a medical examination include the occurrence of an image acquisition order related to image diagnosis, the start of examination image acquisition, the completion of examination image acquisition, the completion of automatic image analysis, the occurrence of a clinical examination order, the start of a clinical examination, and the There is a completion of inspection, etc. Examples of statuses belonging to treatment include occurrence of a medication order and completion of medication. The progress information may be acquired from the in-hospital information system 1, as an example. As another example, the progress information may be stored in or retrieved from the storage device 22.

By realizing the candidate specifying function 215, the processing circuit 21 can perform a process based on the message history acquired by the message acquisition function 212, the role information acquired by the role acquisition function 213, and/or the progress information acquired by the progress acquisition function 214. , one or more candidates (hereinafter referred to as message candidates) regarding the message input by the target member are identified. Each of the one or more message candidates includes a first message candidate that corresponds to a first message that functions as a character string representing a conversation, and a second message candidate that corresponds to a second message that has the function of ordering a medical procedure. A third message candidate corresponding to a third message having the function of making a message candidate or information inquiry. As an example, the processing circuit 21 uses a model (hereinafter referred to as a first model) to determine whether the target member is input based on the role information acquired by the role acquisition function 213 and/or the progress information acquired by the progress acquisition function 214. Identify message candidates.

FIG. 3 is a diagram showing an example of input and output of the first model 31. As shown in FIG. 3, the first model 31 is a model that represents the relationship between message history, role information, and progress information, and message candidates corresponding to the message history, role information, and progress information. The first model 31 may be an LUT (look up table) or a database, or may be a machine learning model such as a support vector machine or a deep learning model. The first model 31 is based on a set of training samples that are a combination of input samples including message history, role information, and progress information, and output samples including message candidates corresponding to the message history, role information, and progress information. It is possible to generate It is assumed that the first model 31 according to this embodiment is a deep learning model that outputs a multidimensional vector whose elements are likelihood values for all message candidates. The likelihood value corresponds to the recommendation level. For example, the message candidate corresponding to the maximum likelihood value among all message candidates is output as the first candidate, and the message candidate corresponding to the second largest likelihood value is output as the second candidate. The number of candidates to be output can be arbitrarily set.

The input to the first model 31 is not limited to three types: message history, role information, and progress information. The first model 31 may be configured such that at least one type of message history, role information, and progress information can be input as needed. The first model 31 may be composed of one model, or may be composed of a plurality of models obtained by dividing the process from input to output. The first model 31 is stored in the storage device 22.

By implementing the ordering function 216, when the second message candidate is selected via the mobile terminal 3, the processing circuit 21 places an order for the medical procedure corresponding to the candidate. As an example, the processing circuit 21 issues the order to the HIS or RIS of the hospital information system 1.

By realizing the display control function 217, the processing circuit 21 displays various information on the mobile terminal 3. For example, the processing circuit 21 displays an input screen for a medical sharing application or a chat application on the mobile terminal 3. Specifically, the processing circuit 21 displays the message candidates identified by the candidate identification function 215 on the input screen of the chat application.

The storage device 22 is a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), an SSD (Solid State Drive), an integrated circuit storage device, etc. that store various information. In addition to the above-mentioned storage devices, the storage device 22 also includes a drive for reading and writing various information between portable storage media such as CDs (Compact Discs), DVDs (Digital Versatile Discs), and flash memories, and semiconductor memory devices. It may be a device. Furthermore, the storage device 22 may be located in another computer connected via a network.

The communication device 23 is an interface for transmitting and receiving various information with other computers such as the in-hospital information system 1 and the mobile terminal 3 included in the medical information system 100.

Hereinafter, an example of processing of a chat application by the medical information processing device 2 according to the present embodiment will be described.

FIG. 4 is a diagram illustrating an example of a chat application processing procedure by the medical information processing device 2 according to the first embodiment. It is assumed that at the start of step SA1, the processing circuit 21 implements the chat function 211 and is executing a chat application for the target channel opened for the target medical treatment. Multiple members are registered in the target channel. It is assumed that the process shown in FIG. 4 is performed for each of the plurality of members. Here, the member to be processed is referred to as a target member.

As shown in FIG. 4, the processing circuit 21 receives a trigger by implementing the chat function 211 (step SA1). A trigger means an opportunity to present a message candidate. For example, triggers include occurrence of an image acquisition order, start of examination image acquisition, completion of examination image acquisition, completion of automatic image analysis, generation of a clinical examination order, start of a clinical examination, end of a clinical examination, generation of a medication order. , switching of progress status such as completion of medication. As another example, inputting a message in a chat app may be set as a trigger.

When step SA1 is performed, the processing circuit 21 acquires the role information of the target member by realizing the role acquisition function 213 (step SA2). When step SA2 is performed, the processing circuit 21 acquires progress information of the target medical treatment by implementing the progress acquisition function 214 (step SA3). When step SA3 is performed, the processing circuit 21 acquires the message history regarding the target channel by realizing the message acquisition function 212 (step SA4). The order of steps SA2, SA3, and SA4 is not limited to this, and can be changed as appropriate. Furthermore, steps SA2, SA3, and SA4 may be partially or entirely executed in parallel.

When step SA4 is performed, the processing circuit 21 realizes the candidate specifying function 215, based on the role information acquired in step SA2, the progress information acquired in step SA3, and/or the message history acquired in step SA4. , one or more message candidates are identified (step SA5). In step SA5, the processing circuit 21 inputs role information, progress information, and/or message history into the first model to identify message candidates.

The combinations of role information, progress information, and message history are enormous. As an example, the processing circuit 21 can narrow down the message candidates according to the role information of the target member. In this case, the first model includes the second model and the third model.

FIG. 5 is a diagram illustrating an example of message candidate identification processing using the second model 51 and the third model 52. The second model 51 is a model that expresses the relationship between each member's role information and the type of message that the member corresponding to the role information is permitted to input. The second model 51 receives role information and outputs a message type corresponding to the role information. The second model 51 may be an LUT or a database, or may be a machine learning model such as a support vector machine or a deep learning model. The second model 51 is generated based on a set of training samples that are a combination of input samples containing role information and output samples containing message types corresponding to the role information. The second model 51 is stored in the storage device 22.

As shown in FIG. 5, the third model 52 is a model representing the relationship between a message type, progress information, and message history, and a message candidate corresponding to the message type, progress information, and message history. The message history may be the plurality of messages input to the target channel, or may be the current status of the target medical treatment estimated from the plurality of messages. The third model 52 inputs the message type, progress information, and message history, and outputs message candidates. The third model 52 may be an LUT or a database, or may be a machine learning model such as a support vector machine or a deep learning model. The third model 52 is generated based on a set of training samples, which is a combination of input samples including message types, progress information, and message history, and output samples including message candidates. It is assumed that the third model 52 is a deep learning model that outputs a multidimensional vector whose elements are likelihood values for all message candidates. The third model 52 is stored in the storage device 22.

Message candidates are identified using the second model 51 and the third model 52. Specifically, first, the processing circuit 21 identifies the message type that the target member can input based on the target member's role information and the second model 51. The processing circuit 21 then identifies one or more message candidates based on the message type, progress information, message history, and third model 52. According to this processing procedure, the types of message candidates are narrowed down based on the role information of the target member, and message candidates are identified from the narrowed down message candidates based on progress information and message history.

When step SA5 is performed, the processing circuit 21 displays the message candidates identified in step SA5 in the candidate display field by realizing the display control function 217 (step SA6). In step SA6, the processing circuit 21 displays message candidates in a candidate display field included in the input screen of the chat application displayed on the mobile terminal 3 of the target member. At this time, the processing circuit 21 displays the identified one or more message candidates with a visual effect corresponding to the function of the message candidate. Specifically, message candidates that function as character strings representing conversations, message candidates that have the function of ordering medical treatment, and message candidates that have the function of making information inquiries are Displayed with different visual effects.

When step SA6 is performed, the processing circuit 21 determines whether the message candidate displayed in step SA6 has been selected by realizing the display control function 217 (step SA7).

If it is determined that a message candidate is selected (step SA7: YES), the processing circuit 21 displays a message corresponding to the selected message candidate in the message display column in the input screen (step SA8). In step SA8, the processing circuit 21 displays a message in a message display column included in the input screen of the chat application displayed on the mobile terminal 3 of the target member.

If it is determined that no message candidate has been selected (step SA7: NO), the processing circuit 21 displays the manually input message in the message display column (step SA9).

When step SA8 or step SA9 is performed, the processing circuit 21 ends the process shown in FIG. 4. Thereafter, the processing circuit 21 waits for the input of a trigger, and upon receiving the trigger, executes the processing from steps SA1 to SA8 or SA9 again.

Note that the role information in step SA2, the progress information in step SA3, and the message history acquisition in step SA4 do not have to be performed every time a trigger is accepted in step SA1. If the processing circuit 21 has already acquired this information, it is sufficient to use the acquired information, and steps SA2 to SA4 do not need to be executed.

Next, several embodiments of processing related to a chat application will be described.
(Example 1)
FIG. 6 is a diagram showing a transition example of the input screen I1 of the chat application according to the first embodiment. The input screen I1 is displayed on the mobile terminal 3 by the display control function 217 of the processing circuit 21. As shown in the left diagram of FIG. 6, the input screen I1 includes a message display field R1, a candidate display field R2, and a manual input field R3. Further, the name "xxx" of the care team of the target medical treatment may be displayed at the top of the input screen I1.

In the message display field R1, chat messages input by members registered in the target channel are displayed in chronological order. Each message is composed of a combination of an icon representing the sender of the message and a display frame in which a character string representing the specific content of the message is displayed. In addition to the character string representing the specific content of the message, the display frame displays the full name of the sender of the message, role information of the sender, and the date and time of sending the message. For example, the initials of the sender of the message may be written on the icon. “Bot” is displayed on the icon shown in the left diagram of FIG. "Bot" indicates that the sender of the message is a robot built into the chat app. The robot is realized by a chat function 211 by the processing circuit 21. For example, a message related to the current progress of the target medical treatment, such as a message informing the completion of medical image capturing or the end of automatic image analysis performed by the in-hospital information system 1, is automatically input. In the case of the left diagram of FIG. 6, a message "CTA scan arrival" is displayed, which is related to the completion of imaging of a contrast-enhanced CT image and notifies the arrival of a contrast-enhanced CT image to the in-hospital information system 1.

One or more message candidates specified by the candidate specifying function 215 are displayed in the candidate display column R2. The message candidates are displayed as GUI (Graphical User Interface) buttons B11, B12, B13, B14, and B15 on which character strings representing the message candidates are written. When one message candidate is selected via the mobile terminal 3, a message corresponding to the selected message candidate is displayed in the message display column R1. For example, when the message candidate "left M1" B11 is selected, the message "left M1" MS1 corresponding to the message candidate is displayed in the message display column R1. When a message candidate is selected, the candidate display field R2 may be deleted from the input screen I1 in order to improve the visibility of the message display field R1.

The manual input field R3 is a GUI component for each member to manually input a message via the mobile terminal 3. For example, if there is no suitable message among the message candidates displayed in the candidate display field R2, a message is manually input via the manual input field R3. The manually inputted message is displayed in the message display field R1.

It is assumed that the message candidates according to the first embodiment are candidates related to messages that function as character strings representing conversation sentences. By realizing the candidate identification function 215, the processing circuit 21 according to the first embodiment generates characters representing the image diagnosis result as a message candidate when the role information of the target member is a radiologist and the progress information is that the image analysis has been completed. Displays a GUI button (hereinafter referred to as a chat button) for inputting columns.

In the example shown in FIG. 6, the name of the target member is "LLL GGG" and the abbreviation is "LG". The role information of the target member is a radiologist (RAD), and the progress information is that the imaging of a contrast-enhanced CT image has been completed. When the imaging of the contrast-enhanced CT image is completed, the in-hospital information system 1 performs automatic image analysis of the contrast-enhanced CT image. Here, since the patient is suspected of having a cerebral infarction, it is assumed that the contrast-enhanced CT image is analyzed for the presence or absence of a cerebral infarction. The fact that a patient is suspected of having a cerebral infarction is managed and stored by the in-hospital information system 1 as medical information such as electronic medical records and test orders. As a result of image analysis, it is assumed that the patient is suffering from an acute cerebral infarction due to large vessel occlusion (LVO), in which occlusion occurs in the M1 blood vessel in the left hemisphere. The results of the image analysis are managed and stored by the in-hospital information system 1 as medical information such as electronic medical records.

In such a situation, the processing circuit 21 sends message candidates to image diagnosis related to the contrast-enhanced CT image based on the target member's role information "radiologist" and progress information "imaging completion of contrast-enhanced CT image." Narrow down to strings that represent the results. Further, the processing circuit 21 uses medical information indicating that the patient is suspected of having a cerebral infarction and that the LVO is in the M1 blood vessel in the left hemisphere, and selects, as the first candidate, that LVO is suspected in the M1 blood vessel in the left hemisphere. The character string "left M1" representing the image diagnosis result is identified as a message candidate. Note that the processing circuit 21 may specify, as preliminary message candidates, various blood vessels that are anatomically in an inclusive or proximate relationship with the M1 blood vessel in the left hemisphere, based on the medical information. In the example of FIG. 6, "left" meaning the blood vessel of the left brain, the character string "M2" meaning the M2 blood vessel, and the character string "ICA" meaning the cerebral carotid artery are specified as preliminary message candidates. Furthermore, the character string "No LVO", which means that there is no suspicion of LVO, may be specified as a preliminary message candidate.

Chat buttons B11, B12, B13, B14, and B15 representing these message candidates are displayed in the display column R2. Message candidates are displayed from left to right in order of recommendation. In addition, the chat button B11 of the first candidate "left M1" is decorated with shading, and the chat buttons B12, B13, B14, and B15 of the other message candidates "left", "M2", "ICA", and "No LVO" are decorated with shading. Displayed without decoration. By displaying message candidates with different visual effects depending on the recommendation level in this way, it becomes possible to support each member in selecting a message candidate.

As shown in the right diagram of FIG. 6, when the chat button B11 of the message candidate "left M1" is selected, the processing circuit 21 sends the message "left M1" MS1 corresponding to the message candidate "left M1" to the target member. The message "Dr. LLL GGG" is displayed in the message display field R1. The message "left M1" MS1 may be displayed with a visual effect corresponding to the message type being a character string, for example, in black fine print.

According to the first embodiment, candidates for character strings that the target member may input next are displayed as message candidates depending on the role of the target member and the progress of the target medical treatment. It becomes possible to reduce the input effort.

(Example 2)
The message candidates according to the second embodiment are candidates related to messages having the function of ordering medical treatment. The medical practice according to the second embodiment means a medical practice managed by the hospital information system 1, such as surgery, medication, taking a medical image, analyzing a medical image, and interpreting a medical image. By realizing the candidate identification function 215, the processing circuit 21 according to the second embodiment, when the role information of the target member is the doctor in charge of the suspected symptom in the target medical treatment, provides medical care for the symptom as a message candidate. A GUI button (hereinafter referred to as "order button") for placing an order for an action is displayed.

FIG. 7 is a diagram showing a transition example of the input screen I2 of the chat application according to the second embodiment. The layout of the input screen I2 is similar to the layout of the input screen I1 in FIG. Input screen I2 in FIG. 7 represents a continuation of input screen I1 in FIG. It is assumed that the target member according to the second embodiment has a name of "PPP DDD", an abbreviation of "PD", and a role information of "neurologist (NER)". As shown in FIG. 7, the character string "left M1" representing the image diagnosis result is input into the message display field R1 by the member "LG" who is a radiologist, and the current information regarding the target treatment is input by the target member "PD". A string representing the situation is entered. Specifically, the following information is input: that the NIHSS score is 15, that the onset time, that is, the last known well (LKW), is 2 hours, and that the target patient has an allergy to penicillin. These multiple messages are stored in the storage device 22 as a message history. An onset time of 2 hours means that t-PA dosing is clinically possible.

In such a situation, the processing circuit 21 uses the second model to narrow down the message candidates to medical treatment orders related to neurology based on the target member's role information "neurophysician." Furthermore, the processing circuit 21 analyzes the message history and determines that the neurologist has diagnosed LVO. The processing circuit 21 also analyzes the message history and identifies that the onset time is 2 hours. Then, using the third model, the processing circuit 21 uses the message type ``Order for medical treatment related to neurology'', the progress information ``Completion of imaging of contrast CT image'', and the message history ``The neurologist diagnosed LVO.'' Based on the message "The onset time is 2 hours", the t-PA medication order "TPA order" is specified as the first candidate message. As another candidate, the processing circuit 21 may specify an order for surgery for LVO, "intervention order," as a preliminary message candidate, as an order for medical treatment other than t-PA medication.

Order buttons B21 and B22 representing these message candidates are displayed in the candidate display field R2. Message candidates B21 and B22 are displayed from left to right in order of recommendation level. The first candidate "TPA order" B21 is decorated with shading, and the other message candidate "intervention order" B22 is displayed without shading. Further, in order to distinguish the type of message candidate, specifically whether it is a character string or an order, the message candidates are visually distinguished and displayed. For example, in the example of FIG. 7, order buttons B21 and B22 are surrounded by a thick frame. As shown in FIG. 6, chat buttons B11, B12, B13, B14, and B15 are surrounded by thin lines. The target member can clearly understand at a glance that the displayed message candidate is an order by visually observing the visual effect of the GUI button.

As shown in the right diagram of FIG. 7, when the order button B21 of the message candidate "TPA order" is selected, the processing circuit 21 sends the message "TPA order" MS2 corresponding to the message candidate "TPA order" to the target member. The message "Dr. PPP DDD" is displayed in the message display field R1. The message "TPA order" MS2 is preferably displayed in a format corresponding to the message type being an order, for example, in bold. Furthermore, when the message candidate "TPA order" is selected, the processing circuit 21 implements the ordering function 216 to automatically place an order for t-PA medication to the in-hospital information system 1, which is the ordering destination. Note that the processing circuit 21 may display an order button for prompting the target member to give a final instruction for placing an order for t-PA medication. In this case, the processing circuit 21 issues an order for t-PA medication upon pressing the order button.

As another example, the processing circuit 21 may display the person who approves the t-PA medication order on the input screen, along with an order button. When the order button is pressed, the processing circuit 21 displays a GUI button on the display screen of the approver's mobile terminal 3 that indicates whether or not to approve the order, for example, a GUI button that indicates approval (hereinafter referred to as approval). button) and a GUI button indicating disapproval (hereinafter referred to as a "reject button"). When the approval button is pressed, the processing circuit 21 automatically places an order for t-PA medication to the in-hospital information system 1, which is the order destination. When the reject button is pressed, the processing circuit 21 may discard the t-PA medication order, and display a text to the effect that the order has been discarded by the approver on the target member's input screen.

According to the second embodiment, candidates for orders that the target member is likely to order next are displayed as message candidates, so that it is possible to reduce the effort required by the target member to place an order. Furthermore, since a message corresponding to the order candidate is displayed in the message display field, it becomes possible to easily share among members that an order has been placed.

In the above embodiment, an order for t-PA medication is given as an example of the order, but the order is not limited to this, and may also be an order for surgery, an order for medical image capturing, an order for medical image interpretation, etc.

(Example 3)
The message candidates according to the third embodiment are candidates related to messages that have a function of inquiring about onset time. If the role information of the target member is the doctor in charge of the suspected symptom in the target medical treatment, and there is no message regarding the onset time of the symptom in the message history, the processing circuit 21 according to the third embodiment selects the onset time as a message candidate. A GUI button for inquiring about the time (hereinafter referred to as onset time inquiry button) is displayed.

FIG. 8 is a diagram showing a transition example of the input screen I3 of the chat application according to the third embodiment. The layout of the input screen I3 is similar to the layout of the input screen I1 in FIG. Input screen I3 in FIG. 8 represents a continuation of input screen I1 in FIG. As shown in FIG. 8, a character string "left M1" representing an image diagnosis result is input into the display field R1 by a member "LG" who is a radiologist. It is assumed that the target member according to the third embodiment is "PD" who is a neurologist. Since the target member has been diagnosed with LVO by a radiologist, it is necessary to determine the onset time and decide whether to administer t-PA medication.

In such a situation, the processing circuit 21 uses the second model to narrow down the message candidates to medical treatment orders related to neurology based on the target member's role information "neurophysician." Furthermore, the processing circuit 21 analyzes the message history and determines that the neurologist has diagnosed LVO. The processing circuit 21 also analyzes the message history of the target channel and determines that there is no message regarding the onset time. Then, using the third model, the processing circuit 21 uses the message type ``Order for medical treatment related to neurology'', the progress information ``Completion of imaging of contrast CT image'', and the message history ``The neurologist diagnosed LVO.'' ” and “There is no message regarding the onset time.” As the first candidate, a message candidate “LKW?” that inquires about the onset time is identified.

An onset time inquiry button B31 representing a message candidate "LKW?" is displayed in the display field R2. In order to distinguish the type of message candidate, specifically, whether it is a character string, an order, or an inquiry, the message candidates are visually distinguished and displayed. For example, in the example of FIG. 8, the onset time inquiry button B31 is surrounded by a dotted line frame. This makes it possible to understand at a glance that the displayed message candidate is an inquiry.

As shown in the right diagram of FIG. 8, when the onset time inquiry button B31 of the message candidate "LKW?" is selected, the processing circuit 21 sends the message "LKW?" MS31 corresponding to the message candidate "LKW?" Displayed in the display field R1 as a message for the target member "Dr. PPP DDD". The message "LKW?" MS31 is preferably displayed with the same visual effect as the character string.

Triggered by the message "LKW?" being displayed, the processing circuit 21 analyzes the medical information stored in the in-hospital information system 1 and extracts or predicts the onset time of the target patient. When the onset time is extracted or predicted, the processing circuit 21 displays a message MS32 representing the onset time as a message from the sender "Bot" in the message display column R2. For example, if the onset time is 2 hours, the message "2 hours ago" MS32 is displayed.

Note that if the onset time is not extracted or predicted from the hospital information system 1, the processing circuit 21 may display a message requesting other members to input the onset time in the display field R1. The member who saw the message enters a character string representing the onset time as a message via the mobile terminal 3.

According to the third embodiment, GUI button candidates for inquiring about information that the target member may want to know are displayed as message candidates, making it possible to reduce the effort required by the target member to input the inquiry text. Become. Further, when the GUI button is pressed, a response is automatically specified or predicted from the hospital information system 1 and displayed in the message display field, so it is possible to reduce the effort required to obtain a response.

(Example 4)
The message candidates according to the fourth embodiment are candidates related to messages having a function of inquiring about medical treatment content information regarding the contents of the target medical treatment. When the target member is a member newly invited to the target channel, the processing circuit 21 according to the fourth embodiment uses a GUI button (hereinafter referred to as "medical content inquiry") for inquiring medical content information related to the target medical treatment as a message candidate. button).

FIG. 9 is a diagram showing a transition example of the input screen I4 of the chat application according to the fourth embodiment. The layout of the input screen I4 is similar to the layout of the input screen I1 in FIG. The target member is a neurologist "PD". It is assumed that the target member has been newly invited to the target channel regarding the target medical treatment, and has been asked for his or her opinion regarding the target treatment by the neurointerventionalist "RH." It is assumed that a relatively long time has elapsed since the channel was opened for the target channel, and many messages have been input into the message display field R1 of the input screen I4.

When the target member is newly invited to the target channel and the message history of the target channel satisfies the medical treatment content inquiry condition, the processing circuit 21 determines to display the medical treatment content inquiry button. The medical care content inquiry condition means a situation in which the message history is estimated to include many messages. Specifically, the medical care content inquiry conditions are set such that the number of messages included in the message history is equal to or greater than a threshold value, and the elapsed time from the opening of the target channel is equal to or greater than a predetermined time.

When the medical care content inquiry condition is satisfied, the processing circuit 21 displays a medical care content inquiry button B41 written as "Summary?" in the candidate display field R2, as shown in the left diagram of FIG. Since the medical care content inquiry button B41 is a message candidate for inquiry, it is surrounded by a dotted line frame.

As shown in the right diagram of FIG. 9, when the medical care details inquiry button B41 for the message candidate "Summary?" is selected, the processing circuit 21 sends the message "Summary" MS41 corresponding to the message candidate "Summary?" It is displayed in the message display field R1 as a message of the member "Dr. PPP DDD". The message "Summary" MS41 is preferably displayed with the same visual effect as the character string.

When the message "Summary" MS41 is displayed, the processing circuit 21 reads the message history of the target channel stored in the storage device 22, and creates a summary of the read message history. The method of creating the summary sentence is not particularly limited. As an example, the processing circuit 21 predicts content that the target member may want to know in the current situation based on the target member's role information, target medical treatment progress information, and/or message history, and sends a message regarding the predicted content. Extract messages from the message history and create a summary sentence that combines the extracted messages. In addition to the extracted message, the summary text preferably includes the transmission date and time of the message.

When the summary text is created, the processing circuit 21 displays a message MS42 representing the summary text as a message from the sender "Bot" in the message display column R1. In the case of the above-mentioned specific situation related to Example 4, the role information of the target member is a neurologist, the progress information is the completion of surgery, and the message history includes a report on the surgical results, so the target member may want to know in the current situation. Certain contents are predicted to be the contents of image diagnosis results and medical procedures. Therefore, as a summary sentence, for example, "Left M1 (03:14 pm), t-PA (03:21 pm), intervention (04:28 pm)" is created and displayed.

According to the fourth embodiment, since the medical treatment details inquiry button is displayed for the target member newly invited to the channel, it is possible to reduce the effort required for the target member to understand the contents of the message history. Furthermore, for members other than the target member, it is possible to reduce the effort of explaining the contents of the message history to the target member.

(Example 5)
The message candidate according to the fifth embodiment is a message candidate having a function of inquiring about the contents of a task assigned to a target member. When the task assigned to the target member in the target channel satisfies the warning condition without being completed, the processing circuit 21 according to the fifth embodiment creates a button (hereinafter referred to as task inquiry) for inquiring about the contents of the task as a message candidate. button).

FIG. 10 is a diagram showing a transition example of the input screen I5 of the chat application according to the fifth embodiment. The layout of the input screen I5 is similar to the layout of the input screen I1 in FIG. It is assumed that the situation in which the target member according to the fifth embodiment is placed is the same as that in the fourth embodiment. That is, it is assumed that the target member is a neurologist "PD" and has been newly invited to the target channel related to the target medical treatment. It is assumed that the target member has been asked for his or her opinion regarding the target medical treatment by a neurointerventionalist "RH." It is assumed that a relatively long time has elapsed since the channel was opened for the target channel, and many messages have been input in the message display field R1 of the input screen I5. Therefore, as described in the fourth embodiment, the medical treatment content inquiry button B52 is displayed in the candidate display field R2.

The processing circuit 21 according to the fifth embodiment determines to display a task inquiry button when the task assigned to the target member in the target channel satisfies the warning condition without being completed. The warning condition means a situation in which a task assigned to a target member in a target channel is estimated to be incomplete. Specifically, the warning conditions include that the task assigned to the target member has not been completed, that the task has remained uncompleted for a certain period of time, that the number of uncompleted tasks has exceeded a threshold, etc. Set.

The processing circuit 21 determines that a task has been assigned to the target member and that the task is incomplete based on the role information of the target member, the progress information of the target medical treatment, and/or the message history of the target channel. For example, if the message history includes a direct message requesting a task to the target member, such as "PPP DDD, please do XX", the processing circuit 21 determines that the task has been assigned to the target member. Is possible. Alternatively, the processing circuit 21 estimates the current situation of the target member based on the role information of the target member, the progress information of the target medical treatment, and/or the message history of the target channel, and responds to the current situation. A specific task may be identified and it may be determined that the task is assigned to the target member.

The task assigned to the target member is incomplete, for example, if there is a direct message in the message history, such as "Not finished yet", which means the task is incomplete. Alternatively, if there is no direct message in the message history indicating that the task has been completed, such as "XX completed," it is possible to determine that the task is not completed. Alternatively, the processing circuit 21 estimates the current situation of the target member based on the role information of the target member, the progress information of the target medical treatment, and/or the message history of the target channel, and based on the current situation. If it is estimated that the target member has not completed the task, it may be determined that the task has not been completed.

When the warning condition is satisfied, the processing circuit 21 displays a task inquiry button B51 labeled "Task list?" in the candidate display field R2, as shown in the left diagram of FIG. Since the task inquiry button B51 is an inquiry message candidate, it is surrounded by a dotted line frame.

As shown in the right diagram of FIG. 10, when the task inquiry button B51 of the message candidate "Task list?" ) MS5 is displayed in the message display field R1. In the task list MS5, combinations of requesters of uncompleted tasks and the contents of the tasks are arranged in a table format. For example, in the example of FIG. 10, it can be seen that the target member "PD" has received two tasks, the task "aaaaa" and the task "bbbbb", from the requester "LG". By checking the task list MS5, the target member can easily understand the uncompleted tasks assigned to him/her. Note that the task list MS5 is not limited to being displayed in the message display field R1, and the window in which the task list MS5 is displayed may be superimposed on the input screen I5.

According to the fifth embodiment, since the task inquiry button is displayed for the target member, it is possible to reduce the effort and time required for the target member to understand uncompleted tasks. Furthermore, since the target members can easily understand the task, it is expected that the task will be carried out smoothly.

(Modified example)
Some modifications of this embodiment will be described below.

(Modification 1)
If the situation requires the opinion of another expert, the processing circuit 21 may display a GUI button (hereinafter referred to as an invitation button) for inviting the expert to the target channel on the input screen. The specific processing procedure is as follows. First, the processing circuit 21 estimates the current status of the target medical treatment based on the progress information of the target medical treatment and the message history of the target channel. Based on the current situation and the role information of each member registered in the target channel, the processing circuit 21 determines whether there is a role missing in the target channel in light of the current situation. If there are missing roles in the target channel, it means that the situation requires the opinion of another expert.

If it is determined that there is a missing role in the target channel, the processing circuit 21 identifies the person corresponding to the role (invitee) from the personnel management ledger managed in the hospital information system 1, and An invitation button for inviting an invitee is displayed. When the invite button is pressed via the mobile terminal 3, the processing circuit 21 sends a notification inviting the invitee to the target channel through the invitee's chat. Displayed on the input screen of the application. According to the first modification, it becomes possible to easily invite experts to the channel.

(Modification 2)
In Modified Example 1, when the invitee is invited to the target channel, but there is no response from the invitee, the processing circuit 21 selects a GUI button (hereinafter referred to as "telephone button") for making a call to the invitee on the input screen. may be displayed. Specifically, the processing circuit 21 identifies the invitee's telephone number from the personnel management ledger managed in the hospital information system 1, and displays a telephone button to call the invitee together with the telephone number. . When the telephone button is pressed via the mobile terminal 3, the mobile terminal 3 can make a telephone call with the mobile terminal 3 of the invitee. According to the second modification, it becomes possible to easily call an expert.

(Modification 3)
If the situation requires the opinion of another expert, the processing circuit 21 may display a telephone button on the input screen for calling the expert. The method for estimating a situation in which another expert's opinion is required is the same as in the first modification. If it is determined that there is a missing role in the target channel, the processing circuit 21 identifies the person (invitee) corresponding to the role from the personnel management ledger etc. managed in the hospital information system 1, A telephone button for calling an invitee is displayed together with a telephone number. When the telephone button is pressed via the mobile terminal 3, the mobile terminal 3 can make a telephone call with the mobile terminal 3 of the invitee. According to the third modification, it becomes possible to easily call an expert.

(Modification 4)
When an examination image arrives at the in-hospital information system 1 when the chat application is executed, the processing circuit 21 displays a GUI button (hereinafter referred to as the first transition) for transitioning to the examination image display screen on the radiologist's input screen. button) may be displayed. For example, when a CT test image arrives at the in-hospital information system 1, a notification to the effect that the CT test image has arrived is sent from the in-hospital information system 1 to the medical information processing device 2. When the processing circuit 21 receives the notification, it specifies the radiologist member from among the members registered in the target channel regarding the CT examination image. Then, the processing circuit 21 displays the first transition button on the input screen of the radiologist member. When the first transition button is pressed via the radiologist member's mobile terminal 3, the CT examination image is displayed on the radiologist member's mobile terminal 3. According to Modification 4, it becomes possible to check the inspection image quickly and easily.

(Modification 5)
When the automatic image analysis of the examination image is completed when the chat application is executed, the processing circuit 21 displays a GUI button (hereinafter referred to as "GUI button") for transitioning to the display screen of the analysis results of the automatic image analysis on the radiologist's input screen. A second transition button) may also be displayed. For example, when a CT test image arrives at the in-hospital information system 1, the in-hospital information system 1 or the like subjects the CT test image to automatic image analysis. When the automatic image analysis is completed, the in-hospital information system 1 sends a notification to the medical information processing device 2 that the automatic image analysis has been completed. When the processing circuit 21 receives the notification, it specifies the radiologist member from among the members registered in the target channel regarding the CT examination image. Then, the processing circuit 21 displays a second transition button on the input screen of the radiologist member. When the second transition button is pressed via the radiologist member's mobile terminal 3, the analysis result of the CT examination image is displayed on the radiologist member's mobile terminal 3. According to the fifth modification, it becomes possible to quickly and easily confirm the analysis results of automatic image analysis.

(Modification 6)
When running a chat app, important character strings related to diagnosis that should be recorded in the electronic medical record may be input as a message. When an important character string related to diagnosis is input as a message, the processing circuit 21 displays a GUI button (hereinafter referred to as a paste button) on the input screen for instructing to paste the message into the electronic medical record. good. Whether or not a character string is important for diagnosis can be determined by, for example, a technology applying machine learning such as word2vec. When the paste button is pressed via the mobile terminal 3, the processing circuit 21 pastes a message corresponding to an important character string related to the diagnosis to the corresponding item in the electronic medical record related to the target treatment. According to the sixth modification, it is possible to reduce the effort required to input information into the electronic medical record.

Note that displaying the paste button is not essential. For example, when a message displayed in the message display column is selected via the mobile terminal 3, the processing circuit 21 may paste the selected message on the electronic medical record. The message selection method is not particularly limited, but long-pressing, double-clicking, etc. are intuitive and suitable.

(Second embodiment)
The medical information processing device according to the second embodiment identifies message candidates using medical information related to the target medical treatment in addition to the message history, role information, and/or progress information. A medical information processing device according to a second embodiment will be described below. In the following description, components having substantially the same functions as those in the first embodiment are denoted by the same reference numerals, and will be described repeatedly only when necessary.

FIG. 11 is a diagram showing an example of the configuration of the medical information processing device 2 according to the second embodiment. As shown in FIG. 11, the processing circuit 21 of the medical information processing device 2 includes a chat function 211, a message acquisition function 212, a role acquisition function 213, a progress acquisition function 214, a candidate identification function 215, an order placement function 216, and a display control function. In addition to 217, a medical information acquisition function 218 is realized. Each of the functions 211 to 218 is not limited to being implemented by a single processing circuit. A processing circuit may be constructed by combining a plurality of independent processors, and the functions 211 to 218 may be realized by each processor executing a medical information processing program. Furthermore, the functions 211 to 218 may be modularized programs that each constitute a medical information processing program. These applications and/or programs are stored in the storage device 22. Note that the medical care sharing application and the chat application are included in the medical information processing program.

By realizing the medical information acquisition function 218, the processing circuit 21 acquires medical information regarding the target medical treatment from the in-hospital information system 1. Medical information is generated in various computer systems that make up the in-hospital information system 1. Specifically, the hospital information system 1 includes an electronic medical record system, an examination reservation system, a PACS, an image processing system, a reporting system, a vital monitoring system, an electrocardiogram system, and/or a medication information system. The electronic medical record system generates an electronic medical record that includes basic patient information, medical history, current medical status, etc. as medical information. The test reservation system issues orders for various tests such as image tests, blood tests, and medication as medical information. The PACS performs an image test according to the order of the image test and generates a medical image as medical information. The image processing system analyzes the presence or absence of image findings based on medical images and generates image analysis results as medical information. The reporting system generates an interpretation report of a medical image by a radiologist as medical information. The vital monitoring system generates vital information such as a patient's heart rate, respiratory rate, blood pressure, and oxygen saturation as medical information. The electrocardiogram system generates a patient's electrocardiogram or the results of the electrocardiogram as medical information. Note that each of the above systems is configured by one computer or two or more computer network systems that can communicate with each other.

By implementing candidate identification function 215, processing circuitry 21 identifies one or more message candidates based on medical information in addition to message history, role information, and/or progress information.

By realizing the display control function 217, the processing circuit 21 displays various information on the mobile terminal 3. For example, the processing circuit 21 displays the message candidates identified by the candidate identification function 215 on the input screen of the chat application. Furthermore, the processing circuit 21 may display the medical information acquired by the medical information acquisition function 218.

Hereinafter, an example of processing of a chat application by the medical information processing device 2 according to the present embodiment will be described.

FIG. 12 is a diagram illustrating an example of a chat application processing procedure by the medical information processing device 2 according to the second embodiment. It is assumed that at the start of step SB1, the processing circuit 21 implements the chat function 211 and is executing a chat application for the target channel opened for the target medical treatment. Multiple members are registered in the target channel. It is assumed that the process shown in FIG. 12 is performed for each of the plurality of members.

As shown in FIG. 12, the processing circuit 21 receives a trigger by implementing the chat function 211 (step SB1). When step SB1 is performed, the processing circuit 21 acquires the role information of the target member by realizing the role acquisition function 213 (step SB2). Steps SB1 and SB2 are similar to steps SA1 and SA2, respectively.

When step SB2 is performed, the processing circuit 21 acquires the progress information of the target medical treatment by realizing the progress acquisition function 214 (step SB3), and acquires the message history regarding the target channel by realizing the message acquisition function 212 (step SB3). SB4), and by realizing the medical information acquisition function 218, medical information of the target medical treatment is acquired (step SB5). The order of steps SB3, SB4, and SB5 is not limited to this, and some or all of them may be executed serially.

When steps SB3, SB4, and SB5 are performed, the processing circuit 21 implements the candidate identification function 215 to process the role information acquired in step SB2, the progress information acquired in step SB3, the message history acquired in step SB4, and the One or more message candidates are identified based on the medical information acquired in step SB5 (step SB6). When step SB6 is performed, the processing circuit 21 displays the message candidates identified in step SB6 in the candidate display field by realizing the display control function 217 (step SB7). According to the second embodiment, message candidates are further identified using medical information, making it possible to identify and display message candidates that are more appropriate for the current situation.

As an example, similar to the first embodiment, the processing circuit 21 inputs role information, progress information, and/or message history into the first model to identify message candidates, and selects medical information from among the identified message candidates. Extract message candidates related to , and display the extracted message candidates. Below, clinical examples will be explained for each type of medical information.

(Example 1: Medical information in electronic medical record system)
Basic patient information is used in various ways in the medical treatment of patients suspected of having a stroke. For example, patients with a history of high blood pressure, diabetes, or heart disease are at higher risk of developing a stroke. Or, if you are taking medicines that thin the blood, it may be difficult to perform surgeries that involve bleeding. When the input screen is discussing the diagnosis of a patient suspected of having a stroke, whether surgery is required, etc., message candidates could be displayed based on information obtained from the electronic medical record. For example, if there is a medical history that corresponds to the risk of stroke, the processing circuit 21 may display a message candidate such as "High blood pressure and diabetes. Risk of stroke is high."

(Example 2: Medical information in test reservation system)
For example, only when the necessity of a CT examination is being discussed on the input screen and the medical information processing device 2 receives information from the examination reservation system that no CT examination has been ordered, the processing circuit 21 Options for issuing a CT examination order may be displayed on the input screen.

(Specific example 3: PACS medical information)
Once medical images such as CT images taken with an imaging device are registered in PACS, they can be referenced at various locations within the hospital. For example, when the medical information processing device 2 acquires information from the PACS that a head CT image of a target patient has been newly received, a doctor who performs an image diagnosis is naturally motivated to view the image. Therefore, it is preferable that the processing circuit 21 displays the option of displaying an image on the input screen. Furthermore, there may be a situation in which the user wants to check only images suspected of being abnormal. In this case, the processing circuit 21 has the option of displaying the image only when it receives information from the PACS that a new image has been received, and also receives a result from the image processing system indicating that an abnormality is suspected in the analysis results. may be displayed on the input screen.

(Specific example 4: Medical information of image processing system)
For example, the presence or absence of cerebral hemorrhage and the presence or absence of cerebral infarction are factors that determine the next medical action. For example, when the medical information processing device 2 receives analysis result information from the image processing system, and the analysis result indicates that there is no cerebral hemorrhage or that cerebral infarction is suspected, a drug called t-PA that dissolves blood clots is often administered. . Therefore, in this case, the processing circuit 21 may display the option of t-PA arrival on the input screen.

(Example 5: Medical information in reporting system)
The results of an image diagnosis performed by a radiologist on a well-equipped medical image display workstation are more accurate than those obtained by a neurologist or the like viewing the images on a mobile device. When the medical information processing device 2 receives information from the reporting system that an image diagnosis report of the target patient has been registered, the processing circuit 21 may display options for displaying the image diagnosis results on the input screen. In addition, if the chat history includes the results of a neurologist checking images on a mobile device, the processing circuit 21 further receives a summary of the image diagnosis report from the reporting system, and combines the summary results and images. An option to display a report may be displayed on the input screen only when the results of the diagnostic report contradict each other.

(Example 6: Medical information of vital monitoring system)
Vital data can be a turning point in the flow of stroke treatment. For example, in patients with high blood pressure, high blood pressure may have caused cerebral hemorrhage, in which case t-PA administration would further worsen the cerebral hemorrhage. Therefore, for example, when the medical information processing device 2 receives the latest blood pressure information from the vital monitoring system, the received blood pressure information indicates hypertension, and information that t-PA has not been administered is received from the medication information system, the processing circuit 21 may perform processing such that, for example, the administration option is not displayed on the input screen even if t-PA has not been administered.

(Specific example 7: Medical information of electrocardiogram system)
Heart diseases such as atrial fibrillation and myocardial infarction can cause stroke, and electrocardiogram tests are sometimes performed on patients suspected of having a stroke. For example, when the medical information processing device 2 receives from the electrocardiogram system the completion of an electrocardiogram test of a target patient and information about the presence or absence of abnormalities in the results of the electrocardiogram test, the processing circuit 21 receives the results of the electrocardiogram test. When indicating that there is an abnormality, a button to contact a heart specialist may be displayed on the input screen.

(Example 8: Medical information of medication information system)
t-PA is one of the typical drugs used for patients suspected of cerebral infarction, and is a drug that dissolves blood clots that cause cerebral infarction. However, guidelines stipulate that this drug be administered within three hours of the onset of cerebral infarction, and there are also contraindications such as a history of intracranial hemorrhage, major surgery within 14 days, or serious trauma outside the head. For example, even if the medical information processing device 2 receives information from the medication information system that t-PA has not been administered, a message indicating the patient's medical history information from the electronic medical record or a chat message stating that the patient has experienced contraindications If exchange is being performed, the processing circuit 21 may perform a process of not displaying the t-PA administration option on the input screen.

When step SB7 is performed, the processing circuit 21 determines whether the message candidate displayed in step SB7 has been selected by realizing the display control function 217 (step SB8). If it is determined that a message candidate is selected (step SB8: YES), the processing circuit 21 displays a message corresponding to the selected message candidate in the message display column in the input screen (step SB9). If it is determined that no message candidate has been selected (step SB8: NO), the processing circuit 21 displays the manually input message in the message display column (step SB10).

When step SB9 or step SB10 is performed, the processing circuit 21 ends the process shown in FIG. 12. Thereafter, the processing circuit 21 waits for the input of a trigger, and upon receiving the trigger, executes the processing from steps SB1 to SB9 or SB10 again.

In the above embodiment, it is assumed that the medical information acquisition process is executed every time a trigger related to presentation of a message candidate occurs. However, as the types and sources of medical information increase, there is concern that the degree of delay in the acquisition and analysis of medical information will also increase, leading to a reduction in the real-time performance of message candidates and presentation of medical information. . The processing circuit 21 acquires medical information by push notification, periodic polling, or stepwise processing to ensure real-time performance. In the case of a push notification, the processing circuit 21 receives the medical information sent by the hospital information system 1 as a push notification when the medical information is generated in the hospital information system 1, and caches the received medical information in the storage device 22. . In the case of periodic polling, the processing circuit 21 polls the in-hospital information system 1 regarding acquisition of medical information when a trigger regarding presentation of message candidates occurs. In the case of step-by-step processing, the processing circuit 21 displays on the input screen the acquisition status of medical information regarding systems in the hospital information system 1 for which medical information has been acquired, systems in the process of acquiring medical information, and/or systems for which medical information has not been acquired. Note that the measures for ensuring real-time performance described above are not limited to acquiring medical information, but can also be applied to acquiring role information and/or progress information.

FIG. 13 is a diagram illustrating the process of obtaining medical information using push notifications. As shown in FIG. 13, medical information is communicated between the in-hospital information system 1 and the medical information processing device 2. The hospital information system 1 described above includes a variety of computer systems, but for the sake of simplicity, one computer system from which medical information is obtained is illustrated as one block. When new medical information related to the target medical treatment occurs in the hospital information system 1, the hospital information system 1 spontaneously sends a push notification of the medical information to the medical information processing device 2. The medical information processing device 2 receives the push-notified medical information and caches it in the storage device 22. Cached medical information is used to identify subsequent message candidates.

According to the push notification method, medical information is transmitted to the medical information processing device 2 only when new medical information is generated in the in-hospital information system 1. Therefore, compared to the case where medical information acquisition processing is performed every time a trigger occurs, it is possible to reduce the amount of communication data with various in-hospital information systems 1. Therefore, the push notification method can be a means to ensure real-time performance.

FIG. 14 is a diagram illustrating the process of obtaining medical information through periodic polling. Similarly to FIG. 13, in FIG. 14, the in-hospital information system 1 from which medical information is obtained is illustrated as one block. When a trigger occurs, the medical information processing device 2 first inquires of the in-hospital information system 1 whether there is any update or addition to the medical information previously provided to the device 2 (polling) (S1). The in-hospital information system 1 confirms the content of the inquiry, and if there is no update or addition, it replies to the medical information processing device 2 to that effect. In this case, the medical information processing device 2 uses the previously acquired medical information as is for subsequent message candidate identification processing. On the other hand, if there is an update or addition, the in-hospital information system 1 transmits the medical information to the medical information processing device 2 (S2). The medical information processing device 2 receives the medical information and uses the received medical information for subsequent message candidate identification processing and the like.

According to the periodic polling method, compared to the push notification method, processing for checking whether medical information has been updated or added in the hospital information system 1 is added, but this is limited to when there is an update or addition. Since communication of medical information occurs, it is possible to reduce the amount of data communicated with various in-hospital information systems 1. Therefore, periodic polling can also be a means of ensuring real-time performance.

FIG. 15 is a diagram illustrating the medical information acquisition process in stepwise processing. In FIG. 15, the in-hospital information system 1 includes a PACS 1-1, an electronic medical record system 1-2, and a vital monitoring system 1-3. At the stage when the medical information processing device 2 acquires medical information from the PACS 1-1, the electronic medical record system 1-2, and the vital monitoring system 1-3, a device with a slow response (usually slow due to hardware performance, etc.), There is a possibility that a situation such as one where the load on a certain device is temporarily high and the response is slow may occur. In order to cope with such a situation, the medical information processing device 2 performs medical information acquisition processing (SB5) and message candidate identification processing for each of the systems 1-1, 1-2, and 1-3 that make up the hospital information system 1. The processing (SB6) and message candidate display processing (SB7) are parallelized.

As shown in FIG. 15, as an example, the medical information processing device 2 has completed receiving the contrast CT image from the PACS 1-1, has not yet received the electronic medical record from the electronic medical record system 1-2, and the vital monitoring system Assume that the vital values from 1-3 are being received. In this case, the processing circuit 21 acquires the medical information "contrast CT image" and the progress information "contrast CT image received", identifies message candidates based on these information, and Show message suggestions. At this time, the processing circuit 21 displays the progress status regarding the acquisition of medical information from each hospital information system 1 on the input screen of the chat application.

FIG. 16 is a diagram illustrating a display example of the progress status regarding acquisition of medical information. The input screen I6 shown in FIG. 16 corresponds to the input screen I1 shown in FIG. As shown in FIG. 16, the input screen I6 includes a progress display field R4 in addition to a message display field R1, a candidate display field R2, and a manual input field R3. The display contents of the message display field R1, candidate display field R2, and manual input field R3 are the same as the display contents shown in FIG. 6. The display contents of the message display field R1 and the candidate display field R2 are based on the reception of the contrast-enhanced CT image from the PACS 1-1.

In the progress display field R4, the progress of obtaining medical information for each in-hospital information system 1 is displayed. For example, if medical information is waiting to be obtained from the electronic medical record system 1-2 and the vital monitoring system 1-3, a message to that effect, such as "Waiting for response from electronic medical record, vital system", may be displayed. Note that the display contents of the progress display column R4 are not limited to this. For example, information regarding the in-hospital information system 1 from which medical information has been obtained, such as "Information obtained from PACS" may be displayed. Further, detailed information on waiting for acquisition of medical information, such as whether medical information is being received or not yet received, may be displayed.

In order to improve the accuracy of message candidates, the processing circuit 21 may use patient-specific information as medical information. Patient-specific information includes basic patient information, medical history, drug usage history, and/or test results. The patient's basic information includes height, weight, gender, age, presence of allergies, etc. The following describes how patient-specific information is used. The usage situations can be divided into confirming the diagnosis, determining the treatment policy, and predicting the prognosis.

(Usage situation: Confirmation of diagnosis)
For example, a patient's medical history and drug history may be reviewed to determine the cause of a stroke. Stroke can be caused by an embolus or rupture within a blood vessel. Therefore, if the patient has had risk factors such as diabetes or high blood pressure in the past, it is said that the cause is often an embolus in the blood vessel. Additionally, when using drugs that cause blood coagulation abnormalities, blood clots are often the cause, so it is important to check the history of drug use.

(Usage situation: Deciding treatment policy)
For example, if a patient has complications such as high blood pressure or diabetes, the dosage and type of medication may need to be adjusted. Therefore, it is necessary to decide on a treatment policy, taking into account the patient's medical history, drug usage history, presence of allergies, etc. For example, if a patient is using anticoagulants and bleeding occurs, the drug must be discontinued. Therefore, it is necessary to adjust treatment by taking into account the patient's medical history, drug history, test results, etc.

(Usage scene: Prediction of prognosis)
For example, elderly patients or patients with comorbidities are likely to have a poor prognosis. Therefore, it is necessary to predict the prognosis by considering the patient's age, medical history, history of drug use, test results, etc.

According to the several embodiments described above, the medical information processing device 2 according to the present embodiment includes the processing circuit 21. The processing circuit 21 acquires a message history that is a history of messages already input by a plurality of members of the target channel in a target channel opened in a chat application for discussing the target medical treatment. The processing circuit 21 acquires role information representing the role of the target member among the plurality of members. The processing circuit 21 acquires progress information representing the current progress of the target medical treatment. The processing circuit 21 identifies one or more message candidates related to the message input by the target member based on the message history, role information, and/or progress information. The processing circuit 21 displays one or more message candidates on the input screen regarding the target channel of the target member.

According to the above configuration, since message candidates are displayed on the input screen, it becomes possible for the target member to reduce the time and effort required to think about the content of the message and input the message. Faster decision-making among team members. This makes it possible to provide efficient team medical care using a chat app even for urgent diseases such as stroke and pulmonary embolism.

According to at least one embodiment described above, it is possible to support quick decision-making in team medical care.

The term "processor" used in the above description refers to, for example, a CPU, GPU, Application Specific Integrated Circuit (ASIC), or programmable logic device (for example, Simple Programmable Logic Device). This refers to circuits such as SPLDs, complex programmable logic devices (CPLDs), and field programmable gate arrays (FPGAs). A processor realizes its functions by reading and executing a program stored in a memory circuit. Note that instead of storing the program in the memory circuit, the program may be directly incorporated into the circuit of the processor. In this case, the processor realizes its functions by reading and executing a program built into the circuit. On the other hand, when the processor is, for example, an ASIC, instead of storing the program in a storage circuit, the function is directly incorporated into the processor's circuitry as a logic circuit. Note that each processor of this embodiment is not limited to being configured as a single circuit for each processor, but may also be configured as a single processor by combining multiple independent circuits to realize its functions. good. Furthermore, multiple components in FIGS. 1, 2, and 11 may be integrated into one processor to implement its functions.

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

1 Hospital information system 2 Medical information processing device 3 Mobile terminal 21 Processing circuit 22 Storage device 23 Communication device 100 Medical information system 211 Chat function 212 Message acquisition function 213 Role acquisition function 214 Progress acquisition function 215 Candidate identification function 216 Order placement function 217 Display Control function 218 Medical information acquisition function

Claims (20)

a message acquisition unit that acquires a message history that is a history of messages already input by a plurality of members of the channel in a channel opened in a chat application for discussing the target medical treatment;
a role acquisition unit that acquires role information representing a role of a target member among the plurality of members;
a progress acquisition unit that acquires progress information representing the current progress of the target medical treatment;
a candidate identification unit that identifies one or more candidates regarding a message input by the target member based on the message history, the role information, and/or the progress information;
a display control unit that displays the one or more candidates on an input screen regarding the channel of the target member;
A medical information processing device comprising: Each of the one or more candidates includes a first candidate regarding a first message that functions as a character string representing a conversation, a second candidate regarding a second message having the function of ordering medical treatment, or an information inquiry. The medical information processing device according to claim 1, which is a third candidate for a third message having a function of performing. The medical information processing apparatus according to claim 2, wherein the display control unit displays the one or more candidates with a visual effect corresponding to a function of the candidate. The display control unit, when one candidate is selected from the one or more candidates, displays a message corresponding to the one candidate in a message display field of the input screen. Item 1. Medical information processing device according to item 1. Each of the one or more candidates includes a first candidate regarding a message that functions as a character string representing a conversational sentence, a second candidate regarding a message having the function of ordering medical treatment, or a function of making an information inquiry. The third candidate regarding the message,
The display control unit displays a message corresponding to the one candidate with a visual effect corresponding to a function of the one candidate.
The medical information processing device according to claim 4. When the role information of the target member is a radiologist and the progress information is that image analysis has been completed, the candidate specifying unit inputs a character string representing an image diagnosis result as the one or more candidates. The medical information processing device according to claim 1, wherein the medical information processing device displays a button. If the role information of the target member is the doctor in charge of the suspected symptom in the target medical treatment, the candidate specifying unit selects a button for ordering medical treatment for the symptom as the one or more candidates. The medical information processing device according to claim 1, wherein the medical information processing device displays:. 8. The medical information processing apparatus according to claim 7, further comprising an order placing unit that, when the button is pressed, places an order corresponding to the button to a system of a destination of the order. When the role information of the target member is a doctor in charge of the suspected symptom in the target medical treatment, and there is no message regarding the onset time of the symptom in the message history, the candidate identification unit identifies the one or more The medical information processing device according to claim 1, wherein a button for inquiring about the onset time is displayed as a candidate. When the target member is a member newly invited to the channel, the candidate specifying unit displays a button for inquiring medical treatment content information regarding the content of the target medical treatment as the one or more candidates. Item 1. Medical information processing device according to item 1. The candidate specifying unit displays a button for inquiring about the contents of the task as the one or more candidates when the task assigned to the target member in the channel satisfies a warning condition without being completed. The medical information processing device according to claim 1. The medical information processing apparatus according to any one of claims 7 to 11, wherein the display control unit displays a message corresponding to the button on the input screen when the button is pressed. further comprising a medical information acquisition unit that acquires medical information regarding the target medical treatment from an in-hospital information system,
The candidate identifying unit further identifies the one or more candidates based on the medical information.
The medical information processing device according to claim 1. The medical information processing apparatus according to claim 13, wherein the medical information acquisition unit acquires the medical information from the in-hospital information system when a trigger related to presentation of the one or more candidates occurs. The medical information acquisition unit receives the medical information push notification from the hospital information system upon generation of the medical information in the hospital information system, and caches the received medical information in a storage device. The medical information processing device according to claim 13. 14. The medical information processing apparatus according to claim 13, wherein the medical information acquisition unit polls the in-hospital information system regarding acquisition of the medical information when a trigger regarding presentation of the one or more candidates occurs. The display control unit displays, on the input screen, the acquisition status of the medical information regarding a system in which the medical information has been acquired, a system in which the medical information is acquired, and/or a system in which the medical information has not been acquired among the in-hospital information systems. 14. The medical information processing device according to 13. 14. The medical information processing apparatus according to claim 13, wherein the medical information includes basic patient information, medical history, drug use history, and/or test results regarding the target medical treatment. 14. The medical information processing apparatus according to claim 13, wherein the hospital information system includes an electronic medical record system, an examination reservation system, a PACS, an image processing system, a reporting system, a vital monitoring system, an electrocardiogram system, and/or a medication information system. to the computer,
A function for acquiring message history, which is a history of messages already input by multiple members of a channel opened in a chat application for discussing target medical treatment,
a function of acquiring role information representing a role of a target member among the plurality of members;
a function for acquiring progress information representing the current progress of the target medical treatment;
A function for specifying one or more candidates regarding a message input by the target member based on the message history, the role information, and/or the progress information;
a function of displaying the one or more candidates on an input screen regarding the channel of the target member;
A medical information processing program that realizes this.