JP2023106968A - Program, method for processing information, information processor, and method for generating model - Google Patents

Abstract

To provide a program that can present the state of a patient after the patient took an administered medicine.SOLUTION: A computer acquires the type of an injection, the amount of the dosage, and the rate of the dosage for a patient according to a program. The computer outputs state information of a patient to whom an injection was administered by inputting the acquired type of the injection, the amount of the dosage, and the rate of the dosage into a learning model which learned to output state information showing the state of the patient to which the injection was administered when the type of the injection, the amount of the dosage, and the rate of the dosage are input.SELECTED DRAWING: Figure 1

Description

The present invention relates to a program, an information processing method, an information processing apparatus, and a model generation method.

2. Description of the Related Art A system has been proposed that assists a doctor in diagnosing a patient's condition in a medical institution or the like. For example, Patent Document 1 discloses a system for estimating a possible disease diagnosis or a patient's condition based on clinical information including patient examination records, nursing records, test results, medication records, medical images, and the like. there is The system disclosed in Patent Document 1 presumes and presents a patient's diagnosis based on the contents of a drug prescription and an injection order given to the patient, and also compares the diagnosis and the drug prescription. and the content of the injection order, outputting a warning when the ordered drug is contraindicated for the disease condition of the diagnosis name.

Japanese Unexamined Patent Application Publication No. 2007-18460

Medical institutions handle a wide variety of drugs, and the dosage and administration rate differ depending on the drug, so doctors need to prescribe drugs accurately, and nurses need to administer drugs accurately according to the prescription. There is If the type, dose, or rate of administration of the drug is incorrect, the patient may be in a serious condition, so accurate work is required of doctors and nurses, which places a heavy burden on them. In addition, if the type, dosage, or rate of administration of the drug is incorrect, it is necessary to predict the state of the patient after administration and take appropriate measures, but the experience and knowledge of doctors and nurses It is difficult to make good decisions when the The system disclosed in Patent Document 1 estimates the diagnosis from the drug ordered for the patient by associating the diagnosis name or the patient's condition with the name of the drug that can be used or the name of the contraindicated drug. In addition, it only judges whether the ordered drug is contraindicated or not, and cannot present what kind of condition the patient will be in if the wrong drug is administered. .

An object of one aspect is to provide a program or the like capable of presenting a patient's condition after medication.

A program according to one aspect acquires the type, dosage, and administration rate of an injection to be administered to a patient, and when the type, dosage, and administration rate of the injection are input, the patient who received the injection Input the obtained type, dosage and administration rate of the injection into a learning model trained to output the state information indicating the state of the injection, and output the state information related to the patient who received the injection. Make the computer perform the process to

In one aspect, the patient's status after medication can be presented.

It is a block diagram which shows the structural example of an information processing apparatus. It is a schematic diagram which shows the structural example of electronic medical record DB. FIG. 4 is a schematic diagram showing a configuration example of a learning model; FIG. 11 is a flowchart showing an example of a learning model generation processing procedure; FIG. 4 is a flowchart showing an example of a procedure for predicting a state after administration; FIG. 5 is an explanatory diagram showing an example of a screen; FIG. 11 is a flowchart showing another example of a post-medication state prediction processing procedure; FIG. FIG. 11 is an explanatory diagram showing another screen example; FIG. 10 is a schematic diagram showing a configuration example of a learning model according to Embodiment 2; FIG. 10 is a flowchart showing an example of a procedure for predicting a post-medication state according to Embodiment 2. FIG. FIG. 5 is an explanatory diagram showing an example of a screen; FIG. 12 is a flowchart showing an example of a procedure for predicting a post-medication state according to a third embodiment; FIG. FIG. 5 is an explanatory diagram showing an example of a screen; FIG. 10 is an explanatory diagram showing a configuration example of a learning model used for identifying recovery measures;

Hereinafter, the program, the information processing method, the information processing apparatus, and the model generation method of the present disclosure will be described in detail based on the drawings showing the embodiments.

(Embodiment 1)
Based on medication information, including the type of injection (drug name), dosage, and administration rate, doctors and nurses can predict the patient's condition after medication is administered according to the medication information. An information processing device to be presented to a medical worker such as In this embodiment, for example, the state of a patient after administration of an injection used in a method of directly administering a drug into a blood vessel using an infusion pump or a syringe pump is predicted. Injections are not limited to drugs that are directly administered into blood vessels, and may be drugs that are administered intracutaneously, subcutaneously, or into body tissues such as muscles using an injection needle.

FIG. 1 is a block diagram showing a configuration example of an information processing apparatus. The information processing device 10 is a device capable of various information processing and transmission/reception of information, and is, for example, a server computer, a personal computer, or the like. A plurality of information processing apparatuses 10 may be provided to perform distributed processing, or may be realized by a plurality of virtual machines provided in one apparatus. The information processing apparatus 10 is installed and used in, for example, a medical institution. When the information processing device 10 is configured by a server computer, the information processing device 10 may be a local server installed in a medical institution, or may be a cloud server connected for communication via a network such as the Internet. good.

The information processing apparatus 10 performs a process of outputting state information indicating the post-medication state of the patient based on the dosing information regarding the drug (injection) to be administered to the patient and the pre-medication vital data of the patient. . Specifically, as will be described later, the information processing apparatus 10 performs machine learning for learning predetermined training data, receives medication information and vital data, and outputs post-administration state information as a learning model 12M (FIG. 3). ) are prepared in advance. The information processing device 10 inputs the patient's medication information and vital data to the learning model 12M, and acquires post-medication status information from the learning model 12M.

The information processing apparatus 10 includes a control section 11, a storage section 12, a communication section 13, an input section 14, a display section 15, a reading section 16, etc. These sections are interconnected via a bus. The control unit 11 has one or more processors such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), a GPU (Graphics Processing Unit), or an AI chip (AI semiconductor). The control unit 11 appropriately executes the control program 12P stored in the storage unit 12 to perform various information processing, control processing, etc. that the information processing apparatus 10 should perform.

The storage unit 12 includes a RAM (Random Access Memory), flash memory, hard disk, SSD (Solid State Drive), and the like. The storage unit 12 stores in advance a control program 12P (program product) executed by the control unit 11 and various data necessary for executing the control program 12P. The storage unit 12 also temporarily stores data and the like generated when the control unit 11 executes the control program 12P. The storage unit 12 also stores a learning model 12M that has learned training data by, for example, machine learning. The learning model 12M is a learned model that has been trained to output state information indicating the state after medication when the patient's medication information and vital data are input. The learning model 12M is assumed to be used as a program module that constitutes artificial intelligence software. The learning model 12M performs a predetermined calculation on an input value and outputs the calculation result. In the storage unit 12, data such as coefficients of functions and threshold values that define this calculation are stored as the learning model 12M. be done. The storage unit 12 also stores an electronic medical chart DB 12a, which will be described later. The electronic medical record DB 12a may be stored in another storage device connected to the information processing device 10, or may be stored in another storage device with which the information processing device 10 can communicate.

The communication unit 13 is a communication module for connecting to a network such as the Internet or a LAN (Local Area Network) by wired communication or wireless communication, and transmits and receives information to and from another device via the network. The input unit 14 receives an operation input by the user and sends a control signal corresponding to the content of the operation to the control unit 11 . The display unit 15 is a liquid crystal display, an organic EL display, or the like, and displays various information according to instructions from the control unit 11 . The input unit 14 and the display unit 15 may be a touch panel integrally configured.

The reading unit 16 reads information stored in a portable storage medium 1a including CD (Compact Disc)-ROM, DVD (Digital Versatile Disc)-ROM, USB (Universal Serial Bus) memory, SD (Secure Digital) card, etc. read. The control program 12</b>P (program product) and various data stored in the storage unit 12 may be read from the portable storage medium 1 a by the control unit 11 via the reading unit 16 and stored in the storage unit 12 . Also, the control program 12P and various data stored in the storage unit 12 may be downloaded from another device by the control unit 11 via the communication unit 13 and stored in the storage unit 12 .

FIG. 2 is a schematic diagram showing a configuration example of the electronic medical chart DB 12a. The electronic medical chart DB 12a is a database that stores electronic medical chart data (medical records) of patients who use medical institutions. The electronic medical chart DB 12a shown in FIG. 2 includes a patient ID column, a patient information column, a vital data column, a medication information column, and the like. The patient ID column stores identification information (patient ID, patient code) for identifying each patient. The patient ID may be, for example, a patient registration card number issued by a medical institution. The patient information column stores patient information about the patient in association with the patient ID. The patient information includes, for example, attribute information including patient age and sex, diagnosis name, medical history, height and weight, and the like. The vital data column stores vital data measured from the patient in association with the patient ID. Vital data includes, for example, blood pressure, pulse rate, heart rate, respiration rate, electrocardiogram, body temperature, etc. These data are stored in respective columns in association with, for example, measurement date and time. In addition, each data of the vital data may be represented by a graph showing a change over time, and the graph data is, for example, a series of data recorded each time measurement is performed, including a pair of measurement date and time and measured value. be. The medication information column stores medication information related to an injection that has already been administered to a patient or an injection that is scheduled to be administered (medicine), in association with the patient ID. The medication information includes, for example, medication date and time, drug name (medicine type), dosage, administration rate, progress status after medication, and the like. Post-medication progress includes normal (good) and abnormal conditions. An abnormal state is, for example, a state in which vital data (blood pressure, pulse rate, heart rate, respiratory rate, electrocardiogram, body temperature, etc.) deviates from the normal range within a predetermined period of time after the start of medication. Yes, e.g. blood pressure above normal range or below normal range, heart rate above normal range or below normal range, breathing rate above normal range or below normal range , abnormal electrocardiogram waveform, temperature rise above normal range or fall below normal range.

The contents stored in the electronic medical chart DB 12a are not limited to the example shown in FIG. Various information about the patient may be stored, such as comments entered by the patient. The examination information includes X-ray images, ultrasound images, CT (Computed Tomography) images, MRI (Magnetic Resonance Imaging) images, PET (Positron Emission Tomography) ) images.

FIG. 3 is a schematic diagram showing a configuration example of the learning model 12M. A learning model 12M shown in FIG. Calculations are performed to discriminate the post-medication state of the patient, and the result of the calculation (state information indicating the post-medication state) is output. The learning model 12M may be configured such that patient medication information stored in the electronic medical record DB 12a and data included in the electronic medical record data are input. The learning model 12M is composed of a CNN (Convolutional Neural Network), which is a neural network model generated by deep learning, for example. In addition to CNN, the learning model 12M is configured using algorithms such as RNN (Recurrent Neural Network), LSTM (Long Short-Term Memory), Transformer, Decision Tree, Random Forest, SVM (Support Vector Machine), and the like. It may be configured by combining a plurality of algorithms. The information processing apparatus 10 performs machine learning for learning predetermined training data to generate a learning model 12M in advance. The information processing apparatus 10 inputs the patient's medication information and vital data registered in the electronic medical record DB 12a to the learning model 12M, and predicts the patient's post-medication condition based on the output information from the learning model 12M. . The learning model 12M has a plurality of input nodes, each input node is associated with information to be input, and an input associated with each information of medication information and vital data. It is input to the learning model 12M via the node. The vital data input to the learning model 12M may be time-series data measured periodically.

The learning model 12M shown in FIG. 3 has a plurality of output nodes, and each output node is associated with a preset post-medication state. Output the probability that should be determined as Post-medication conditions include normal and abnormal conditions, abnormal conditions such as blood pressure rising above the normal range or falling below the normal range, heart rate increasing above the normal range or normal Includes below range decrease, above normal range increase or below normal respiratory rate, abnormal ECG waveform, etc. In the example shown in FIG. 3, the output node 0 outputs the probability that the post-medication state should be determined to be good, and the output node 1 outputs the possibility that the blood pressure rises above the normal range as the post-medication state. Output node 2 outputs the probability that it should be determined that there is a possibility that the blood pressure will fall below the normal range as a state after administration. The state associated with each output node is not limited to the example shown in FIG. Including conditions that can arise. The output value of each output node is, for example, a value between 0 and 1.0, and the sum of the discrimination probabilities output from each output node is 1.0 (100%). With the above-described configuration, the learning model 12M of this embodiment outputs information (discrimination probability for each state) regarding the post-medication state of the patient when the patient's medication information and vital data are input.

Note that, in the learning model 12M described above, the information processing apparatus 10 predicts, for example, the state associated with the output node that outputs the maximum output value (discrimination probability) among the output values from the output nodes. Specify the post-medication condition that should be administered. Note that the learning model 12M may have one output node that outputs the state with the highest discrimination probability instead of having a plurality of output nodes that output the discrimination probability for each state.

The learning model 12M is generated by subjecting an unlearned learning model to machine learning using training data including medication information and vital data for training, and information indicating the state after medication (correct label). The training data is generated by adding information indicating the post-medication state (progress) to the patient's medication information and pre-medication vital data for patients to whom various injections have been administered. Medication information, vital data, and post-administration status can be obtained from the medication information, vital data, and progress status of patients who have already been administered medication, stored in the electronic medical chart DB 12a.

When training medication information and vital data are input to the learning model 12M, the output value from the output node corresponding to the state indicated by the correct answer label approaches 1.0, and the output value from the other output nodes approaches 1.0. Learn to approach 0.0. In the learning process, the learning model 12M performs calculations based on the input medication information and vital data, and calculates output values from each output node. Then, the learning model 12M calculates a value corresponding to the calculated output value of each output node and the correct label (1 for the output node corresponding to the state indicated by the correct label, and 0 for other output nodes). are compared, and the parameters used in the arithmetic processing are optimized so that each output value approximates the value corresponding to the correct label. The parameters are weights between neurons in the learning model 12M, and the like. The parameter optimization method is not particularly limited, but error backpropagation, steepest descent method, or the like can be used. As a result, when the patient's medication information and vital data are input, the learning model 12M that has learned to predict the patient's condition after medication and output the prediction result is obtained.

Learning of the learning model 12M may be performed by another learning device. A trained learning model 12M generated by learning in another learning device is downloaded from the learning device to the information processing device 10 via, for example, a network or via the portable storage medium 1a, and stored in the storage unit 12. . The learning model 12M is not limited to the configuration shown in FIG. The learning model 12M can be configured to receive various types of information used when predicting the state after administration of the injection. For example, any one or more of each information such as height and weight, medical history, examination information, treatment information, surgery information, medication history, comments entered by doctors etc. contained in electronic medical record data may be input. good. In this case, the learning model 12M is configured to discriminate post-administration states based not only on medication information and vital data, but also on various types of information included in the electronic medical record data, and to output discrimination probabilities for each state. . The electronic medical record data input to the learning model 12M may be time-series data, for example, time-series data (test information) of test results of regularly performed blood tests or urine tests. It may be time-series data (treatment information) on various treatments performed. In this case, for example, the patient's electronic medical record data for the most recent day, electronic medical record data for the most recent week, or the like can be configured to be input to the learning model 12M.

Processing for learning the training data and generating the learning model 12M will be described below. FIG. 4 is a flow chart showing an example of a processing procedure for generating the learning model 12M. The following processing is performed by the control unit 11 of the information processing device 10 according to the control program 12P stored in the storage unit 12, but may be performed by another learning device.

First, the control unit 11 of the information processing device 10 acquires patient information to be used for training data from the electronic medical record DB 12a, generates training data, and uses the generated training data to learn the learning model 12M. For example, the control unit 11 determines whether or not there is a patient whose condition has deteriorated after medication, based on the progress of medication information for each patient stored in the electronic medical record DB 12a (S11). Specifically, the control unit 11 controls vital data (blood pressure, pulse rate, heart rate, respiration rate, electrocardiogram , body temperature, etc.) is outside the normal range. If it is determined that there is a patient whose condition has deteriorated after medication (S11: YES), the control unit 11 retrieves the patient's medication information, pre-medication vital data, and information indicating the post-medication progress from the electronic medical record DB 12a. (S12). Note that the control unit 11 acquires, for example, vital data for a predetermined period of time (several minutes, several tens of minutes) before the start of drug administration as the vital data before drug administration. The control unit 11 generates training data by adding a correct label indicating the acquired progress state to the acquired medication information and vital data, and stores the training data in the storage unit 12 (S13).

The control unit 11 determines whether or not there is a patient (unprocessed patient) for whom training data generation processing has not been executed among the patients whose condition has deteriorated after administration (S14). If it is determined that there is an unprocessed patient (S14: YES), the control unit 11 returns to the process of step S12, and indicates the unprocessed patient's medication information, pre-medication vital data, and post-medication progress status. Information is acquired from the electronic medical chart DB 12a (S12) and stored in the storage unit 12 as training data (S13). The control unit 11 repeats the processing of steps S12 to S14 until it determines that there is no untreated patient. Accordingly, training data used for learning of the learning model 12M can be generated and accumulated based on the information of the patient whose condition has deteriorated after medication.

If it is determined that there is no patient whose condition has deteriorated after administration (S11: NO), the control unit 11 terminates the process without performing the above-described process. Note that training data may be generated for patients whose condition did not worsen after administration. In this case, the control unit 11 acquires from the electronic chart DB 12a the medication information of the patient whose condition did not worsen after medication, the vital data before medication, and the information indicating the progress status (good) after medication, and stores each information. They are stored in the storage unit 12 as training data in association with each other. Patients whose condition did not worsen after medication include patients whose condition did not worsen despite some mistake (medical malpractice) related to medication, in addition to patients who received appropriate medication. For patients in whom some kind of mistake has occurred in medication, medication information, vital data before and after medication, and state information after medication are collected in advance, and based on the collected data, the control unit 11 generates a learning model. A configuration for generating training data used for learning 12M may also be used. Mistakes in medication (medical malpractice) include, for example, mixing up a drug, administering a different dose or rate of administration than prescribed.

When it is determined that there is no unprocessed patient (S14: NO), the control unit 11 performs learning processing of the learning model 12M using each accumulated training data (S15). Here, the control unit 11 inputs medication information and vital data included in the training data to the learning model 12M, and acquires output values from each output node. The control unit 11 sets the output value of each output node, the value corresponding to the correct label (1 for the output node corresponding to the correct state, and 0 for the output node corresponding to the incorrect state), and are compared, and parameters such as weights between neurons in the learning model 12M are optimized using, for example, error backpropagation so that the two approximate each other.

The control unit 11 determines whether or not there is training data (unprocessed data) for which learning processing has not been performed among the training data accumulated in step S13 (S16). If it is determined that there is unprocessed data (S16: YES), the control unit 11 returns to the process of step S15 and repeats the learning process using the unprocessed training data. If it is determined that there is no unprocessed data (S16: NO), the control section 11 terminates the series of processes.

Through the above-described processing, the learning model 12M that has learned to output information indicating the patient's condition after medication when the patient's medication information and vital data is input is generated. The learning model 12M can be further optimized by repeating the learning process using the training data as described above. Also, the learning model 12M that has already been trained can be re-learned by performing the above-described processing, and in this case, the learning model 12M with higher discrimination accuracy can be generated. Further, for example, by learning training data for each medical institution, the learning model 12M corresponding to each medical institution may be generated. Since the type or manufacturer of the injection prescribed to the patient differs slightly depending on the treatment policy of the medical institution, by generating the learning model 12M for each medical institution, the contents of prescriptions that are frequently performed at each medical institution can be calculated. Taking into account, it becomes possible to predict the state after administration.

Processing for predicting the patient's condition after medication and outputting an alert when, for example, a doctor prescribes medication using the learning model 12M generated by the above-described processing will be described below. FIG. 5 is a flow chart showing an example of the post-medication state prediction processing procedure, and FIG. 6 is an explanatory diagram showing an example of a screen. The following processing is performed by the control unit 11 of the information processing device 10 according to the control program 12P stored in the storage unit 12. FIG.

When a doctor places an order for medication for a patient, the doctor inputs the prescription for medication via the electronic chart screen. Therefore, when the control unit 11 of the information processing apparatus 10 is instructed to activate the electronic medical chart via the input unit 14, the electronic medical chart screen for inputting the electronic medical chart data is displayed on the display unit 15 (S21). . FIG. 6A shows an example of an electronic medical chart screen, and the electronic medical chart screen has an input area 15a for inputting medical records and the like, and a menu bar 15b for assisting input of prescription details related to medication. The input area 15a is configured so that arbitrary information can be input. The menu bar 15b has an input field for inputting a drug name, an input field for inputting a dose, an input field for inputting an administration rate, and an input field for instructing an order based on the entered prescription contents. An OK button is provided for Each input field of the menu bar 15b may be configured to allow entry of arbitrary information or numerical values, or may be provided with a pull-down menu for selecting any one of a plurality of options. In addition, the menu bar 15b may be provided with an input field for entering the manufacturer of the medicine. Alternatively, it may be configured to display a list of drug names beginning with numbers.

When a doctor orders medication by entering medication information in the input area 15a, the doctor can enter the medication information using each entry field of the menu bar 15b. The control unit 11 determines whether or not input of medication information has been received via the menu bar 15b (S22), and if it determines that it has not been received (S22: NO), it waits. At this time, the control unit 11 receives input of information other than medication information via the input area 15a, and performs processing for displaying the input information in the input area 15a. Note that the example shown in FIG. 6A is configured such that the medication time can be directly input to the input area 15a.

When the control unit 11 (acquisition unit) determines that input of medication information (drug name, dose, and administration rate) has been received (S22: YES), the entered medication information is displayed in each input field ( S23). When the input of medication information is received, the control unit 11 acquires the vital data of the patient to be administered medication from the electronic chart DB 12a (S24). For example, based on the patient ID (patient identification code) of the patient displayed on the electronic medical record screen, the control unit 11 selects the most recently measured vital data from the vital data stored in association with the patient ID. get. In addition, when the vital data is continuously measured periodically, the control unit 11 acquires the vital data measured in the most recent predetermined time (for example, 1 minute, 10 minutes, etc.). Vital data includes blood pressure, pulse rate, heart rate, respiration rate, electrocardiogram, and the like. In addition to the vital data, the control unit 11 may also acquire various types of information unique to the site of drug administration, such as clinical physical findings of the patient before drug administration.

The control unit 11 inputs the medication information input via the menu bar 15b and the vital data acquired in step S24 to the learning model 12M, and based on the output values from the learning model 12M, the post-medication patient's condition. The state is predicted (S25). For example, the control unit 11 identifies the output node that outputs the maximum output value (discrimination probability) in the learning model 12M, and identifies the state associated with the identified output node as the state of the patient after administration. . Note that the control unit 11 selects a predetermined number (for example, three) of output values (discrimination probabilities) in descending order of the output values from the output nodes, and associates the selected output values with the output nodes that output them. condition may be identified as a candidate post-medication condition to be predicted. In this case, a predetermined number of states can be presented as candidates for the post-medication state. Based on the predicted state of the patient, the control unit 11 determines whether the deterioration of the state has been predicted (S26). For example, when the control unit 11 predicts a state other than good (normal state) as the post-medication state, it determines that the state is predicted to deteriorate.

If it is determined that deterioration of the condition is predicted (S26: YES), the control unit 11 generates alert information for notifying to what kind of condition there is a possibility of deterioration (S27). For example, the control unit 11 generates alert information for notifying the post-administration state predicted in step S25. Specifically, when it is predicted that blood pressure will rise above the normal range as a post-medication condition, the control unit 11 outputs alert information including a message such as "there is a risk of blood pressure rising above the normal range." Generate. Then, the control unit 11 (output unit) outputs an alert by displaying an alert screen as shown in FIG. 6B on the display unit 15 (S28). This allows the doctor to understand that the entered medication information is inappropriate. In the example shown in FIG. 6B, alert information is displayed in the input area 15a, and information input to the input area 15a is temporarily prohibited. As a result, it is possible to issue a warning when it is predicted that the patient's condition may deteriorate after administration of medication information input via the menu bar 15b. Note that if the information processing apparatus 10 has an audio output unit such as a speaker, the control unit 11 may be configured to output an alert by outputting the alert information audibly in addition to displaying the alert information. .

When the alert is output, the doctor determines that the medication information input via the menu bar 15b is inappropriate and corrects the medication information. The control unit 11 receives input of medication information through the menu bar 15b, here, an instruction to change the input medication information (specifically, changed medication information). Therefore, after outputting the alert, the control unit 11 returns to the process of step S22, and determines whether or not the input (change instruction) of the medication information is received via the menu bar 15b (S22). If the control unit 11 determines that an instruction to change the medication information has been received (S22: YES), the control unit 11 performs the processing of steps S23 to S25, based on the post-medication state predicted based on the changed medication information. is predicted to deteriorate (S26).

The control unit 11 repeats the processing of steps S22 to S28 until it is predicted that the condition after administration will not deteriorate. When the control unit 11 determines that the condition will not deteriorate (S26: NO), that is, when it predicts that the condition after administration is good (normal condition), the OK button on the menu bar 15b is clicked. It is determined whether or not it has been operated (S29). If the controller 11 predicts that the post-medication condition is good, and the alert information is being displayed, the controller 11 terminates the display of the alert information. Moreover, the control unit 11 may display a message indicating that the medication information is appropriate in addition to ending the display of the alert information.

If the control unit 11 determines that the OK button has not been operated (S29: NO), it returns to the process of step S22, inputs medical records (electronic medical record data) through the input area 15a, and operates the menu bar 15b. The process of receiving input of used medication information is repeated. When determining that the OK button has been operated (S29: YES), the control unit 11 displays the medication information input using the menu bar 15b in the input area 15a and adds it to the electronic medical record data (S30).

The control unit 11 determines whether or not to end the electronic medical record input process described above (S31). For example, when a close button (not shown) provided on the electronic medical chart screen is operated, the control unit 11 accepts an electronic medical chart end instruction from the doctor and determines to end the above-described processing. When the control unit 11 determines not to end the above-described processing (S31: NO), it returns to the processing of step S22 and repeats the processing of steps S22 to S30. If it is determined to end the above-described process (S31: YES), the control unit 11 stores the information (electronic medical record data) input to the input area 15a in the electronic medical record DB 12a (S32), and performs a series of processes. finish.

With the above-described processing, when a doctor prescribes a drug, the patient's condition after medication can be predicted, and if it is predicted that the condition may deteriorate, an alert can be issued. Therefore, when a doctor prescribes a drug, it is possible to propose a review of the medication content by an alert, and the doctor corrects the medication content, thereby suppressing the occurrence of an order with an incorrect medication content. In a situation where a wide variety of drugs are used in medical institutions and new drugs are being sold one after another, even inexperienced doctors and doctors with little knowledge of drugs can prescribe drugs with appropriate dosages. Become. This makes it possible to avoid drug prescription errors and reduce the risk of medical malpractice. In addition, in this embodiment, the learning model 12M is used to predict the post-medication patient condition from the patient's medication information and vital data (or electronic medical record data). When using , it is possible to predict the state after administration more accurately.

Next, a description will be given of the process of predicting the patient's condition after administration and outputting an alert using the learning model 12M described above when administering a drug to a patient in the medical field. FIG. 7 is a flowchart showing another example of the post-medication state prediction processing procedure, and FIG. 8 is an explanatory diagram showing another screen example. In the process shown in FIG. 7, steps S41 to S44 are added instead of steps S21 to S23, steps S45 to S46 are added after step S28, and steps S29 to S32 are replaced by step S47. is added. Description of the same steps as in FIG. 5 will be omitted.

In the medical field, for example, when administering medication to a patient, a nurse inputs medication information (drug name, dosage, administration rate) about the medication to be administered via a prediction screen as shown in FIG. 8A. , confirms the prediction result of the patient's condition change due to this medication. Medication to a patient is performed in, for example, an operating room, an ICU (intensive care unit), a hospital room, a treatment room, or the like. For the sake of simplicity of explanation, a configuration in which the information processing device 10 is provided in a room where medication is administered will be described below as an example. A terminal having an input unit and a display unit may be provided in the room where medication is administered, and the terminal and the information processing apparatus 10 in the medical institution may be connected for communication. In this case, the medication information input via the terminal is transmitted to the information processing device 10, the state of the patient after medication predicted by the information processing device 10 based on the medication information is transmitted to the terminal, and the display unit of the terminal is displayed. presented to the nurse via

In the process shown in FIG. 7, the control unit 11 displays a prediction screen as shown in FIG. 8A on the display unit 15 (S41). The screen shown in FIG. 8A includes an input field for patient ID, input fields for drug name, dosage and administration rate as medication information, and an OK button for instructing prediction of post-medication status based on the entered medication information. button. Each entry field may be configured to allow entry of arbitrary information or numerical values, or may be provided with a pull-down menu for selecting an arbitrary one from a plurality of options.

The control unit 11 acquires the patient ID input via the input unit 14, for example, and displays it in the input field for the patient ID (S42). For example, when a code reader such as a barcode reader or a QR code (registered trademark) reader is connected to the information processing apparatus 10, the control unit 11 may read the patient ID by the code reader and display it in the input field. good. For example, when a patient wears a wristband printed with a patient ID code, the patient ID may be read from the wristband code by a code reader. The patient ID may be read from the Also, the control unit 11 (acquisition unit) acquires the medication information input via the input unit 14, for example, and displays it in each input field (S43). When a sticker printed with a code in which drug information including a drug name is coded is attached to the drug package, the control unit 11 reads and inputs the drug name from the code of the drug package using a code reader. can be displayed in the column. Further, when an RFID (Radio Frequency Identifier) tag storing drug information including a drug name is attached to the drug, and a reader for reading the stored information of the RFID tag is connected to the information processing apparatus 10, the control unit 11 may read the drug name from the RFID tag with a reader. Also, when administering medication using an infusion pump or a syringe pump, the drug name, dose, and administration rate are set in the pump and displayed on the display section provided in the pump. Therefore, the nurse can enter the drug name, dosage, and administration rate displayed on the pump in the input fields of the prediction screen. Further, when the information processing device 10 is configured to be able to communicate with the pump, the control unit 11 transmits and receives information to and from the pump, thereby controlling the drug name, dosage, and dosage set for the pump. The configuration may be such that the speed is obtained from the pump and displayed in the input field. In the case of a configuration in which each piece of information is acquired from the pump, input errors for each piece of information do not occur, and the operation burden on nurses can be reduced. Note that the method of inputting medication information is not limited to the example described above. For example, it may be input by a doctor, or may be input via another terminal capable of communicating with the information processing device 10 .

The control unit 11 determines whether or not the OK button on the prediction screen has been operated (S44), and if it determines that the OK button has not been operated (S44: NO), returns to the process of step S43, and administers medication. Repeat acquisition and display of information. If it is determined that the OK button has been operated (S44: YES), the control section 11 performs steps S24 to S26. The process of S28 is performed. In step S28 in FIG. 7, the control unit 11 displays an alert screen as shown in FIG. 8B on the display unit 15 and outputs an alert (S28). This allows the nurse to grasp that the medication to be administered from now on is not appropriate. The screen shown in FIG. 8B is provided with a return button for returning the display to the prediction screen of FIG. 8A, and the control unit 11 determines whether or not the return button in the alert screen has been operated (S45). . When determining that the return button has not been operated (S45: NO), the control unit 11 continues to display the alert screen, and when determining that the return button has been operated (S45: YES), the prediction screen shown in FIG. 8A is displayed. (S46). The prediction screen here is in a state in which the medication information acquired in step S43 is displayed in each input field.

After the process of step S46, the control unit 11 returns to the process of step S43 to input the medication information, here, a change instruction to the medication information displayed in each input column (specifically, the changed medication information). is acquired and displayed in each input column (S43). When the information processing device 10 is configured to communicate with the pump, a medical worker such as a doctor or a nurse instructed by the doctor sets any of the drug name, dosage, and dosage rate set in the pump. By doing so, the control unit 11 can acquire the medication information whose setting has been changed from the pump and display it in the input field. After that, the control unit 11 repeats the processing of steps S43 to S44, S24 to S28, and S45 to S46 until it is predicted that the condition after administration will not deteriorate.

If the control unit 11 determines that the condition will not worsen (S26: NO), the control unit 11 displays a message indicating that the condition is not expected to worsen with respect to the medication based on the medication information input on the prediction screen. It is displayed on the display unit 15 (S47), and the series of processing ends. For example, the control unit 11 displays a message such as "no worsening of condition due to this medication". The nurse then activates the infusion pump or syringe pump to begin administering the drug to the patient.

With the above-described processing, when a nurse or the like administers a drug to a patient, the patient's condition after administration can be predicted, and if it is predicted that the condition may deteriorate, an alert can be sent. can. Therefore, before the nurse or the like administers the drug, it becomes possible to suggest confirmation and review of the medication content by an alert, and it is possible to avoid administering medication with incorrect administration content. As a result, even an inexperienced nurse or a nurse with little knowledge of drugs can administer drugs with appropriate dosage content. Therefore, it is possible to avoid the occurrence of mistakes regarding drug administration and reduce the risk of medical malpractice. Also in the above-described processing, the learning model 12M is used to predict the post-medication patient condition from the patient's medication information and vital data (or electronic medical record data). When using 12M, it becomes possible to predict the state after administration more accurately.

In this embodiment, when a doctor prescribes a drug to a patient, it can be confirmed that there is no wrong prescription content, and before a nurse or the like administers the drug to the patient, it is possible to check that there is no wrong administration content. . Therefore, the risk of medication error (medical malpractice) can be reduced. By suppressing the occurrence of medication errors, patients can receive appropriate medication, and the risk of doctors and nurses becoming perpetrators of medical malpractice is reduced. In addition, when medical malpractice occurs, treatment, treatment, etc. that are not originally necessary are required for recovery, but such treatment, treatment, etc., are no longer necessary because medication errors do not occur. Therefore, the use of medical resources including medical instruments, medicines, and human resources (medical workers such as doctors and nurses) used for treatment and treatment that are not necessary is reduced.

In the present embodiment, a configuration has been described in which the information processing apparatus 10 locally performs the process of predicting the state of the patient after medication using the learning model 12M, but the configuration is not limited to this. For example, the information processing apparatus 10 may be configured by a server computer, and the server computer may perform the process of predicting the state of the patient after administration using the learning model 12M. In this case, the server computer acquires patient medication information and vital data from a terminal device used by a doctor or nurse connected via a network, for example, and outputs the predicted post-medication patient condition to the terminal device. may be configured to display Note that the server computer may store the electronic medical record DB 12a, and in this case, the server computer may acquire only patient medication information from the terminal device. Alternatively, a configuration may be adopted in which a first server that performs post-medication state prediction processing using the learning model 12M and a second server that stores the electronic medical record DB 12a are separately provided. In this case, the first server can be configured to acquire the patient's medication information and vital data from the second server and perform post-administration state prediction processing using the learning model 12M. Even with such a configuration, the same processing as in the present embodiment described above is possible, and the same effects can be obtained.

(Embodiment 2)
When predicting the state after medication from the patient's medication information (drug name, dosage, administration rate) and vital data, out of the medication information and vital data input to the learning model 12M, An information processing apparatus that specifies data with a large degree of contribution and presents the data together with a prediction result will be described. Since the information processing apparatus of this embodiment has the same configuration as the information processing apparatus 10 of Embodiment 1, detailed description of the configuration will be omitted. In addition, in the information processing apparatus 10 of the present embodiment, the learning model stored in the storage unit 12 is slightly different from the learning model 12M of the first embodiment shown in FIG.

FIG. 9 is a schematic diagram showing a configuration example of a learning model according to the second embodiment. The learning model 12Ma of the present embodiment is a so-called explainable AI (Explainable AI, XAI), and the patient's medication information and vital data are input, and the patient's state information after medication (discrimination probability of each state), It is a machine learning model that outputs explanatory data that indicates the grounds for prediction. Also in this embodiment, the learning model 12Ma may be configured to receive medication information and electronic medical record data. The learning model 12Ma uses the SHAP or Attention mechanism or the like to calculate the degree of contribution (SHAP value, Attention, etc.) that should be used as the basis for obtaining the prediction result (model output) for each input data. is configured to The degree of contribution is a value indicating the degree to which each input data of the learning model 12Ma contributed to the output of the predicted state (prediction result). ) is a value that indicates how much attention has been paid to it. Therefore, input data with a high degree of contribution can be regarded as the basis for obtaining the prediction result. A learning model 12Ma shown in FIG. 9 has, in addition to the same configuration as the learning model 12M shown in FIG. . Therefore, the learning model 12Ma is configured to output the degree of contribution (description data) to each input data from each output node for description data. For example, output node 101 outputs the degree of contribution to input data 1 (eg, input data of input node 0), and output node 102 outputs the degree of contribution to input data 2 (eg, input data of input node 1).

The information processing apparatus 10 selects a predetermined number (for example, three) of the output values (contributions) from the output nodes for explanation data in descending order, and sends the selected output values to the output node that outputs the output values. Identify the associated input data as a candidate for the basis of prediction. In this case, a predetermined number of pieces of input data can be specified as candidates for the basis of prediction. Note that the information processing apparatus 10 may specify input data associated with an output node that outputs a contribution (output value) equal to or greater than a predetermined value as candidates for grounds for prediction. The learning model 12Ma of the present embodiment has a plurality of output nodes that output the contribution to each input data, instead of outputting information indicating the input data with the highest contribution (for example, an input node number). A configuration having an output node may also be used.

FIG. 10 is a flowchart showing an example of the post-medication state prediction processing procedure according to the second embodiment, and FIG. 11 is an explanatory diagram showing a screen example. The process shown in FIG. 10 is obtained by adding step S51 between steps S27 and S28 in the process shown in FIG. Description of the same steps as in FIG. 5 will be omitted. The control unit 11 of the information processing apparatus 10 of this embodiment performs the same processes as steps S21 to S27 in FIG. As a result, even in the present embodiment, when medication information is input via the electronic medical record screen, the post-medication condition is predicted based on the medication information and vital data, and an alert is issued when the condition is predicted to deteriorate. Information is generated. In this embodiment, in step S25, the control unit 11 predicts the post-administration state based on the output value from the learning model 12Ma, and specifies the input data that serves as the basis for the prediction result. For example, the control unit 11 selects a predetermined number (for example, three) of contributions from the explanation data (contributions of each input data) output from the learning model 12Ma in descending order, and outputs the selected contributions. The input data associated with the output node is specified as a basis candidate for the prediction result.

After the process of step S27, the control unit 11 generates prediction basis information for notifying the input data that is the basis of the prediction result specified in step S25 (S51). For example, if the administration rate is specified as the input data that is the basis of the prediction result, the control unit 11 generates a message (for example, "candidate cause: administration rate") that notifies the administration rate as the basis of the prediction result. Further, when the administration rate and blood pressure are specified as the input data that serve as the basis for the prediction result, the control unit 11 generates a message (for example, "candidate cause: administration rate/blood pressure ”).

Then, based on the alert information generated in step S27 and the ground information for prediction generated in step S51, the control unit 11 displays an alert screen as shown in FIG. S28). The alert screen shown in FIG. 11A presents, in addition to the contents of the alert screen shown in FIG. 6B, input data that contributes highly to the prediction of the alerted state. In the example shown in FIG. 11A, administration rate and blood pressure are presented as input data with a high degree of contribution to the prediction result. Note that if any of the drug name, dosage, and administration rate included in the medication information is specified as input data that contributes highly to the prediction result, as shown in FIG. The input data entry fields may be highlighted. In addition, in the example shown in FIG. 11A, the administration speed in the menu bar 15b is displayed in reversed black and white. As a result, it is possible to present input data with a high degree of contribution to the prediction result (prediction of the predicted state). , the input field of the menu bar 15b can be emphasized by highlighting.

By the above-described processing, it is presented which of the medication information and vital data has a high degree of contribution to the prediction result (predicted state). Therefore, by checking the state notified by the alert screen and the data that contributes highly to the prediction of this state, the doctor determines whether or not to correct the medication information, and corrects it as necessary. Therefore, it is possible to set appropriate medication information (medication content). By notifying the doctor of the medication information that contributes highly to the prediction result via the menu bar 15b, the doctor can receive information that is highly likely to be corrected among the entered medication information. .

In this embodiment, the learning model 12Ma is not limited to the configuration shown in FIG. can be configured to input various information such as comments input by the . In this case, the learning model 12Ma determines the post-administration state based on not only the medication information and vital data, but also various information included in the electronic medical record data, and each input data for the determination result (prediction result) can be configured to output the contribution of Also in the learning model 12Ma, the electronic medical record data to be input may be time-series data. 12Ma can be configured to be input.

Further, the information processing apparatus 10 of the present embodiment can execute processing similar to that of FIG. 7 when a doctor or a nurse administers medicine to a patient. Note that in the present embodiment, in step S28 in FIG. 7, the control unit 11 displays the alert information in the prediction screen as shown in FIG. 11B instead of displaying the alert screen shown in FIG. 8B. to display the specified input data. Also here, if any of the drug name, dosage, and administration rate included in the medication information is specified as input data that contributes highly to the prediction result, the control unit 11 performs the specified input data as shown in FIG. 11B. The input field for the input data is highlighted. As a result, an alert can be output to present a review of medication content, and it is possible to present which data among the entered medication information has a high degree of contribution to the prediction result.

In this embodiment, the same effects as those of the first embodiment described above can be obtained. Moreover, in this embodiment, it is possible not only to predict the post-medication condition, but also to indicate which input data is the basis for the prediction result. Therefore, the doctor can judge whether the presented input data is appropriate or not, and the doctor or a nurse or the like who receives instructions from the doctor can appropriately modify the medication contents and administer the medication. By doing so, the occurrence of wrong prescriptions and wrong medications can be suppressed. Also in the present embodiment, it is possible to apply the modified examples appropriately described in the first embodiment.

(Embodiment 3)
In the above-described Embodiments 1 and 2, when a doctor prescribes a drug or before a nurse administers a drug to a patient, the patient's condition after drug administration is predicted to confirm whether the drug content is appropriate. It is a configuration that In the present embodiment, an information processing apparatus that predicts the patient's future condition and presents it to doctors and nurses in the event that an incorrect medication content is actually administered to a patient at a medical site will be described. Since the information processing apparatus of this embodiment has the same configuration as the information processing apparatus 10 of Embodiment 1, detailed description of the configuration will be omitted.

In the information processing apparatus 10 of this embodiment, the control unit 11 can execute the same processing as the processing shown in FIG. As a result, if a medication error occurs, the patient ID of the patient and the medication information indicating the content of the medication are input to the prediction screen. Physicians and nurses can be notified by an alert screen as shown in 8B. In the information processing apparatus 10 of the present embodiment, when the control unit 11 predicts that the post-medication condition will worsen, the control unit 11 performs the predicted post-medication condition as well as measures to recover from this condition (drug administration). administration, etc.) is specified and notified.

FIG. 12 is a flow chart showing an example of the post-medication state prediction processing procedure of the third embodiment, and FIG. 13 is an explanatory diagram showing a screen example. The process shown in FIG. 12 is obtained by adding step S61 between steps S27 and S28 and deleting steps S45 and S46 in the process shown in FIG. Description of the same steps as in FIG. 7 will be omitted.

In the information processing apparatus 10 of the present embodiment, when the control unit 11 determines that the state is expected to deteriorate (S26: YES), it generates alert information (S27), and further, when the state deteriorates, Recovery information regarding recovery measures to be taken (recovery measures) is generated (S61). For example, the control unit 11 may specify a recovery action for recovering from a deteriorated state on a rule basis. Specifically, it is possible to use a database in which patient vital data (blood pressure, pulse rate, heart rate, respiratory rate, electrocardiogram, body temperature, etc.) are registered in association with countermeasures for stabilizing the vital data. can. In this case, the control unit 11 identifies a corresponding action corresponding to the predicted state from the DB, and generates recovery information for notifying the identified action. For example, when the control unit 11 specifies the recovery treatment of "medicine A, dosage B, administration speed C", the control unit 11 may provide a Generate recovery information including messages. Furthermore, if this recovery action is a action against an increase in blood pressure, the control unit 11 generates recovery information including a message such as "blood pressure stabilizes when drug A is administered at dose B at administration rate C." may

Note that the process of identifying the treatment to be performed based on the patient's vital data may be performed using a learning model learned by machine learning. For example, it is possible to use a learning model learned to output information indicating a treatment to be performed when vital data is input using algorithms such as CNN, RNN, and LSTM. In this case, the control unit 11 can, for example, input the vital data acquired in step S24 to the learning model, and specify the treatment (recovery treatment) to be performed on the patient based on the output information from the learning model.

FIG. 14 is an explanatory diagram of a configuration example of a learning model used for specifying recovery measures. The learning model shown in FIG. 14 receives the patient's vital data (blood pressure, pulse rate, heart rate, respiratory rate, electrocardiogram, body temperature, etc.), and determines the treatment to be performed on the patient based on the input data. It is trained to perform calculations and output the results of calculations. In addition to the vital data, the learning model shown in FIG. 14 also includes any or more of information such as examination information, treatment information, surgery information, history of taking medicines, comments entered by doctors, etc. included in the electronic medical record data. may be configured to be input. The learning model shown in FIG. 14 has a plurality of output nodes. For example, the output node 0 indicates the probability that the medication treatment of "drug A1, dose B1, administration rate C1" should be performed. Output node 1 outputs the probability that it should be determined that the medication treatment of "drug A2, dose B2, administration rate C2" should be performed, and output node 2 outputs "drug A3, dose B3 , administration rate C3'' is output. The output value of each output node is, for example, a value between 0 and 1.0, and the sum of the discrimination probabilities output from each output node is 1.0. The content of treatment assigned to each output node is not limited to medication, and various treatments performed at medical institutions can be used.

The learning model shown in FIG. 14 uses training data containing vital data for training and information (correct label) indicating the treatment to be performed on the patient of this vital data, and the unlearned learning model is machine-processed. Generated by learning. In the learning model, when vital data for training is input, the output value from the output node corresponding to the action indicated by the correct label approaches 1.0, and the output value from the other output nodes approaches 0.0. Learn to get closer. In the learning process, the learning model performs calculations based on the input vital data and calculates output values from each output node. Then, the learning model compares the calculated output value of each output node with the value (0 or 1) corresponding to the correct label, and performs arithmetic processing so that each output value approximates the value corresponding to the respective correct label. Optimize the parameters used for Again, parameters to be optimized are weights between neurons in the learning model, and the optimization method is not particularly limited. As a result, a learning model is obtained that is trained to determine the treatment to be performed on a patient when the patient's vital data is input. Training of the learning model may be performed on other learning devices. A learned learning model generated by learning in another learning device is downloaded from the learning device to the information processing device 10 via, for example, a network or via the portable storage medium 1a, and stored in the storage unit 12.

After generating the alert information and the recovery information, the control unit 11 displays an alert screen as shown in FIG. 13 on the display unit 15 and outputs an alert (S28). In this embodiment, an alert screen such as that shown in FIG. 13 can notify a patient's condition that may worsen after medication, and can propose advice on measures to recover from this condition. Therefore, doctors, nurses, and the like can predict the worsening of the condition after drug administration, and can grasp and prepare recovery measures to be implemented when the condition worsens. Also here, when the information processing apparatus 10 has an audio output unit, the control unit 11 may be configured to output the alert information and the recovery information by audio.

Also in this embodiment, the learning model 12M shown in FIG. 3 is used when predicting the state after administration in step S25 in FIG. In addition, the configuration may be such that vital data after administration is input. In this case, for example, when an incorrect medication is administered, the patient's subsequent condition can be predicted by considering the vital data before and after the medication before the patient's condition worsens.

In this embodiment, the same effects as those of the above-described embodiments can be obtained. In addition, in this embodiment, when an erroneous medication is administered, it is possible to predict the worsening of the condition after medication, and to propose recovery measures when the worsening is predicted. Mistakes related to medication may occur in situations where the degree of urgency is increasing, such as when a patient suddenly changes, or in situations where a small number of doctors and nurses are responding, such as at night. . In such a situation, it is difficult for a doctor to make a quick and appropriate judgment, and it is conceivable that there are no doctors with a lot of experience or knowledge. Therefore, by using the information processing apparatus 10 of the present embodiment, even in an urgent situation, and even by doctors and nurses with little experience and knowledge, it is possible to predict the condition after medication and to prevent the condition from worsening. It is possible to envision recovery actions to be taken in the event of a failure. Therefore, according to this embodiment, it is possible to reduce the burden on the doctor in an emergency situation or in a situation where the support of a doctor with abundant experience and knowledge cannot be received.

The configuration of the present embodiment can be applied to the information processing apparatuses 10 of the first and second embodiments, and similar effects can be obtained even when applied to the information processing apparatuses 10 of the first and second embodiments. Also in this embodiment, it is possible to apply the modified examples appropriately described in each of the above-described embodiments.

The embodiments disclosed this time are illustrative in all respects and should not be considered restrictive. The scope of the present invention is indicated by the scope of the claims rather than the meaning described above, and is intended to include all modifications within the scope and meaning equivalent to the scope of the claims.

REFERENCE SIGNS LIST 10 information processing device 11 control unit 12 storage unit 13 communication unit 14 input unit 15 display unit 12a electronic chart DB
12M learning model 12Ma learning model

Claims (10)

Acquiring the type, dosage and administration rate of the injection to be administered to the patient,
A learning model that has been trained to output state information indicating the state of a patient who has received the injection when the type, dosage, and administration rate of the injection are input, the acquired type of the injection, A program for causing a computer to execute a process of inputting a dose and an administration rate and outputting status information relating to a patient who has received the injection. obtaining electronic medical record data of the patient;
The obtained type of injection is stored in a learning model that outputs state information indicating the state of the patient who received the injection when the type, dosage, and administration rate of the injection, and electronic medical record data are input. , dosage and administration rate, and electronic medical chart data, and outputting status information about the patient. obtaining an identification code assigned to the patient;
3. The computer according to claim 2, causing the computer to acquire the electronic medical chart data of the patient from a storage unit that stores the electronic medical chart data of the patient in association with the identification code based on the acquired identification code. program. 4. The program according to any one of claims 1 to 3, wherein the state information related to the patient includes information indicating a normal state and information indicating an abnormal state. when the state information indicating the abnormal state is output, outputting alert information indicating the abnormal state;
5. The program according to claim 4, which causes the computer to execute a process of receiving a change instruction for the acquired type, dosage and administration rate of the injection. obtaining from the learning model information corresponding to the degree of contribution of each data input to the learning model to the output of the state information when the state information indicating the abnormal state is output;
6. The program according to claim 4 or 5, causing the computer to execute a process of outputting data having a high degree of contribution to the output of the state information among the data input to the learning model based on the acquired information. Acquiring electronic medical record data of the patient when outputting the state information indicating the abnormal state,
Entering the acquired electronic medical record data into a learning model trained to output the type of injection, dosage and administration rate when the patient's electronic medical record data is input, and outputting the administration rate. Acquiring the type, dosage and administration rate of the injection to be administered to the patient,
A learning model that has been trained to output state information indicating the state of a patient who has received the injection when the type, dosage, and administration rate of the injection are input, the acquired type of the injection, An information processing method in which a computer executes a process of inputting a dose and an administration rate and outputting status information relating to a patient to whom the injection is administered. an acquisition unit that acquires the type, dosage and administration rate of the injection to be administered to the patient;
A learning model that has been trained to output state information indicating the state of a patient who has received the injection when the type, dosage, and administration rate of the injection are input, the acquired type of the injection, An information processing apparatus comprising: an output unit for inputting a dose and an administration rate and outputting state information relating to a patient to whom the injection has been administered. Acquiring training data including the type, dosage, and administration rate of an injection to be administered to a patient, and state information indicating the state of the patient after administration of the injection;
Using the acquired training data, a learning model is generated that outputs state information indicating the state of a patient to whom the injection is administered when the type, dosage, and administration rate of the injection are input. A computer-implemented model generation method.

Images