JP2023098761A - Information processing method, learning model generation method, information collection method, computer program, and information processing device - Google Patents

Abstract

To provide an information processing method, a learning model generation method, an information collection method, a computer program, and an information processing device with which it is possible to assist in administering medicine to patients.SOLUTION: An information processing method pertaining to the present embodiment includes an information processing device acquiring the medical record data of a patient, inputting the acquired medical record data to a learning model having been trained by machine learning so as to output the type of medicine administered to the patient, dosage, and the speed of administration for the input of the medical record data of the patient, acquiring the type of medicine, dosage, and administration speed outputted by the learning model, and outputting the acquired type of medicine, dosage, and administration speed. The information processing device may acquire the identification information of the patient, and acquire medical record data corresponding to the acquired identification information from a database in which the medical record data of a plurality of patients are stored.SELECTED DRAWING: Figure 1

Description

TECHNICAL FIELD The present invention relates to an information processing method, a learning model generation method, an information collection method, a computer program, and an information processing apparatus for assisting drug administration to a patient.

In Patent Document 1, an infusion pump is connected to a needle and an infusion container via an infusion tube, and the infusion solution flowing from the infusion container is discharged from the infusion pump by adjusting the flow rate and delivered to the needle. and a flow sensor for measuring the flow rate of the infusion solution, and controlling at least one of the voltage and frequency of the drive pulse applied to the piezoelectric element of the infusion pump so that the flow rate measured by the flow sensor becomes the set flow rate. Infusion volume control devices have been proposed that are capable of maintaining an infusion volume at a predetermined set value.

JP 2011-177411 A

By using the infusion volume adjusting device described in Patent Document 1, it can be expected that the infusion solution can be administered to the patient at an appropriate flow rate. However, it is a person such as a doctor or a nurse who sets the flow rate and the like for this device, and by making appropriate settings by a person, an appropriate infusion solution can be administered to the patient.

The present invention has been made in view of such circumstances, and aims to provide an information processing method, a learning model generation method, an information collection method, which can be expected to support the administration of drugs to patients. An object of the present invention is to provide a computer program and an information processing device.

In an information processing method according to one embodiment, an information processing device acquires patient chart data, and outputs the type, dosage, and administration rate of a drug to be administered to the patient in response to input of the patient chart data. Input the acquired medical record data into a learning model that has undergone machine learning, acquire the type of drug, dosage and administration rate output by the learning model, and obtain the type, dosage and administration rate of the drug to output

According to one embodiment, it can be expected to assist in the administration of drugs to patients.

1 is a schematic diagram for explaining an overview of an information processing system according to an embodiment; FIG. 1 is a block diagram showing the configuration of a server device according to an embodiment; FIG. 2 is a block diagram showing the configuration of a terminal device according to this embodiment; FIG. FIG. 2 is a schematic diagram for explaining the configuration of a drug prediction model according to the present embodiment; FIG. FIG. 10 is a flow chart showing the procedure of drug prediction model generation processing performed by the server device according to the present embodiment; FIG. FIG. 4 is a schematic diagram showing a display example of drug administration information; 4 is a flow chart showing a procedure of drug prediction processing performed by the server device according to the present embodiment; 4 is a flow chart showing a procedure of drug prediction processing performed by the terminal device according to the present embodiment; 4 is a flow chart showing a procedure of information collection processing regarding vital information by the server device according to the present embodiment. FIG. 11 is a schematic diagram showing a display example of an inquiry screen for progress information; 6 is a flow chart showing the procedure of feedback information acquisition processing performed by the server device according to the present embodiment.

A specific example of an information processing system according to an embodiment of the present invention will be described below with reference to the drawings. The present invention is not limited to these examples, but is indicated by the scope of the claims, and is intended to include all modifications within the scope and meaning equivalent to the scope of the claims.

<System configuration>
FIG. 1 is a schematic diagram for explaining an outline of an information processing system according to this embodiment. In the information processing system according to this embodiment, a drug is administered to a patient 101 who needs to be administered a drug at a medical institution or the like using a pump 5 such as a syringe pump or a micropump. The type, dose, and rate of medicine to be administered by the pump 5 are set by the doctor 102 or the like in charge of the patient 101 . The information processing system according to the present embodiment is a system that assists the doctor 102 in determining the type of medicine, the dosage, the speed of administration, etc. in a situation where the medicine is administered using the pump 5 as described above. In this embodiment, the doctor 102 determines the type, dose, and rate of administration of the drug, and the doctor 102 administers the drug to the patient 101. However, the present invention is not limited to this. . For example, the doctor 102 may determine the type, dosage, and administration rate of the drug, and based on the decision of the doctor 102, a person other than the doctor 102, such as a nurse, may administer the drug to the patient. .

In the information processing system according to the present embodiment, the terminal device 3 used by the doctor 102 can read the two-dimensional code 110 attached to, for example, the patient's 101 clothes, bed, hospital room, medical record printed on a paper medium, or the like. , the patient ID of the patient 101 can be obtained. The terminal device 3 that has acquired the patient ID transmits the acquired patient ID to the server device 1 . The server device 1 according to the present embodiment performs a process of predicting the type, dosage, administration rate, etc. of a drug suitable for administration to the patient 101 using so-called AI (Artificial Intelligence). The server device 1 receives the patient ID transmitted from the terminal device 3, reads the medical chart data corresponding to the received patient ID from the medical chart DB (database) 7, and selects a drug suitable for the patient 101 based on the read medical chart data. Predict the type, dosage, administration rate, etc. The server device 1 transmits prediction results such as the type of medicine, dosage, and administration speed to the terminal device 3 of the doctor 102 as medicine administration information. The terminal device 3 displays the medicine administration information transmitted from the server device 1 . The doctor 102 refers to the drug type, dosage, administration rate, etc. displayed on the terminal device 3, and determines the final drug type, dosage, administration rate, etc. for the patient 101, including his/her own judgment. , the pump 5 is operated to administer the drug to the patient 101 .

Further, in the information processing system according to the present embodiment, vital information such as heart rate, respiration rate, blood pressure, and body temperature of the patient 101 to whom the drug is administered is measured by the measuring instrument 6, and the measured vital information is transmits to the server device 1. The server device 1 stores, as information on the result of administration of the drug to the patient 101, medical record data of the patient 101 used for prediction of the type, dosage, administration rate, etc. of the drug, type of drug based on this medical chart data, administration It collects drug administration information, which is the result of predicting the amount and administration rate, etc., and vital information of the patient 101 after administering the drug, and records and accumulates it in the drug administration DB 8 . Based on the vital information of the patient 101, the server device 1 determines whether or not the type, dose, and administration rate of the drug administered to the patient 101 are appropriate, and records the determination result in the drug administration DB 8. You may These pieces of information recorded in the drug administration DB 8 are used to improve the accuracy of AI that predicts the type of drug, dose, and speed of administration.

Furthermore, in the information processing system according to the present embodiment, the server device 1 acquires feedback information from the doctor 102 regarding the administration of the drug to the patient 101 via the terminal device 3, and records the acquired information in the drug administration DB 8. You may For example, when the doctor 102 corrects the predicted values for the drug type, dosage, and administration rate predicted by the server device 1 and uses the pump 5, the corrected predicted values from the doctor 102 can be acquired by the server device 1 as feedback information. Further, for example, when the pump 5 is used according to the type, dosage, and speed of medicine predicted by the server device 1, the doctor 102 who examined the condition of the patient 101 thereafter The server device 1 can acquire the result of determining whether or not it was appropriate, or an event such as a change in the state of the patient 101 as feedback information. These pieces of information acquired by the server device 1 from the doctor 102 and recorded in the drug administration DB 8 can be used to improve the accuracy of AI that predicts the type of drug, dosage, administration speed, and the like.

<Device configuration>
FIG. 2 is a block diagram showing the configuration of the server device 1 according to this embodiment. The server device 1 according to the present embodiment includes a processing unit 11, a storage unit (storage) 12, a communication unit (transceiver) 13, and the like. In this embodiment, the explanation is given assuming that the processing is performed by one server device, but the processing may be performed by a plurality of server devices in a distributed manner.

The processing unit 11 includes an arithmetic processing unit such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), a GPU (Graphics Processing Unit) or a quantum processor, a ROM (Read Only Memory) and a RAM (Random Access Memory). It is configured using The processing unit 11 reads and executes the server program 12a stored in the storage unit 12 to generate a learning model based on the data acquired from the terminal device 3, and to prepare a drug using the generated learning model. Various processing such as processing for predicting the type, dose, and rate of administration are performed.

The storage unit 12 is configured using, for example, a large-capacity storage device such as a hard disk. The storage unit 12 stores various programs executed by the processing unit 11 and various data required for processing by the processing unit 11 . In the present embodiment, the storage unit 12 stores a server program 12a executed by the processing unit 11 and a drug prediction model 12b, which is a learning model generated by machine learning processing. Further, the storage unit 12 is provided with the above-described chart DB 7 and drug administration DB 8 . However, the chart DB 7 and the medicine administration DB 8 may be provided in an external storage device instead of the storage unit 12 provided in the server device 1 .

In this embodiment, the server program (computer program, program product) 12a is provided in a form recorded in a recording medium 99 such as a memory card or an optical disc, and the server device 1 reads the server program 12a from the recording medium 99 and stores it. Store in section 12 . However, the server program 12a may be written in the storage unit 12 during the manufacturing stage of the server device 1, for example. Further, for example, the server program 12a may be delivered by another remote server device or the like, and the server device 1 may acquire the program through communication. For example, the server program 12 a may be recorded in the recording medium 99 and read by a writing device and written in the storage unit 12 of the server device 1 . The server program 12 a may be provided in the form of distribution via a network, or may be provided in the form of being recorded on the recording medium 99 .

The drug prediction model 12b stored in the storage unit 12 contains configuration information indicating the configuration of the drug prediction model 12b, and internal parameter values of the drug prediction model 12b determined by machine learning processing. and other information may be included. Details of the drug prediction model 12b will be described later.

The chart DB 7 is a database that stores chart data including various information about the patient 101 created by the doctor 102 . The medical chart DB 7 stores medical chart data in association with a patient ID, which is identification information uniquely assigned to each patient. Each medical record data may include various information such as the patient's 101 age, weight, primary disease, test values of various examinations, and treatment history.

The drug administration DB 8 is a database that stores information on drugs administered to the patient 101 . The data stored in the drug administration DB 8 in this embodiment is used for machine learning (relearning) processing for improving the prediction accuracy of the drug prediction model 12b. In the present embodiment, the drug administration DB 8 stores, in association with the patient ID, medical chart data when the drug was administered to the patient 101, drug type and dosage predicted by the drug prediction model 12b based on the medical chart data. and drug administration information such as administration rate, vital information measured by the measuring device 6 in the patient 101 who administered the drug, and feedback information from the doctor 102 in charge of administering the drug.

The communication unit 13 communicates with various devices via a network N including a mobile phone communication network, a wireless LAN (Local Area Network), the Internet, and the like. In this embodiment, the communication unit 13 communicates with the terminal device 3 and the measuring instrument 6 via the network N. FIG. The communication unit 13 transmits the data given from the processing unit 11 to other devices, and gives the data received from the other devices to the processing unit 11 .

Note that the storage unit 12 may be an external storage device connected to the server device 1 . The server device 1 may be a multicomputer including a plurality of computers, or may be a virtual machine virtually constructed by software. The server device 1 is not limited to the above configuration, and may include, for example, a reading unit that reads information stored in a portable storage medium, an input unit that receives operation inputs, or a display unit that displays images. .

In the server device 1 according to the present embodiment, the server program 12a stored in the storage unit 12 is read out and executed by the processing unit 11, so that the patient ID acquisition unit 11a, the chart data acquisition unit 11b, and the drug prediction processing are performed. A unit 11c, a drug administration information output unit 11d, a vital information acquisition unit 11e, a feedback information acquisition unit 11f, a learning processing unit 11g, and the like are implemented in the processing unit 11 as software functional units. In this figure, as functional units of the processing unit 11, functional units related to prediction processing using the drug prediction model 12b and generation processing of the drug prediction model 12b are illustrated, and functional units related to other processing are not shown. omitted.

The patient ID acquisition unit 11a performs processing for acquiring the patient ID assigned to the patient 101 to whom the drug is to be administered. In the present embodiment, the patient ID is acquired by reading the two-dimensional code 110 using the terminal device 3 by the doctor 102 and is transmitted from the terminal device 3 to the server device 1 . The patient ID acquisition unit 11 a acquires the patient ID of the target patient 101 by receiving the patient ID transmitted from the terminal device 3 by the communication unit 13 .

The medical record data acquisition unit 11b performs a process of acquiring the medical chart data of the patient 101 to whom the drug is administered from the medical chart DB7. The chart data acquisition unit 11b reads the chart data of the target patient 101 from the chart DB 7 based on the patient ID acquired from the terminal device 3 by the patient ID acquisition unit 11a.

The drug prediction processing unit 11c uses a learned drug prediction model 12b that has undergone machine learning in advance to perform processing for predicting the type, dosage, and administration rate of the drug to be administered to the patient 101 . The drug prediction processing unit 11c inputs various information included in the medical record data of the patient 101 acquired by the medical record data acquisition unit 11b to the drug prediction model 12b, and calculates the type, dosage, and administration of the drug output by the drug prediction model 12b. Predictions are made by obtaining velocity information.

The drug administration information output unit 11d transmits (outputs) to the terminal device 3, as drug administration information, the prediction result of the type, dose, and administration rate of the drug to be administered to the patient 101 by the drug prediction processing unit 11c. A process for displaying (outputting) drug administration information on the terminal device 3 is performed. Further, the medicine administration information output unit 11d outputs the medicine administration information to the medicine administration DB 8 to store the medicine administration information in the medicine administration DB 8 .

The vital information acquisition unit 11 e acquires vital information such as heart rate, respiration rate, blood pressure, and body temperature of the patient 101 measured by the measuring device 6 by communicating with the measuring device 6 through the communication unit 13 . The vital information acquisition unit 11e stores the vital information of the patient 101 acquired from the measuring instrument 6 in the drug administration DB 8 together with information such as the patient ID and the measurement time. Note that the measuring device 6 continuously measures the heart rate, respiration rate, blood pressure, body temperature, etc. of the patient 101 and periodically transmits the measurement results to the server device 1 . The vital information acquisition unit 11e acquires the measurement results periodically transmitted by the measuring device 6 as vital information, and stores the vital information in the drug administration DB 8 each time it is acquired.

Also, the vital information acquisition unit 11e may determine a change in the condition of the patient 101 to whom the drug is administered based on the acquired vital information. For example, if one or more numerical values included in the vital state exceed a threshold indicating abnormality, or the vital information acquisition unit 11e calculates the amount of change in one or more numerical values included in the vital state for a certain period of time, When the amount of change in any of the numerical values exceeds the threshold, it can be determined that an abnormality has occurred in the patient 101, and the abnormality can be notified to the terminal device 3 or the like. Further, for example, the vital information acquisition unit 11e determines that the patient 101 has recovered when one or more values included in the vital state return from abnormal values to normal values, and notifies the terminal device 3 or the like of the recovery of the patient 101. can do. When performing these determinations, the vital information acquisition unit 11e stores the determination results in the medicine administration DB 8 together with the vital information.

The feedback information acquisition unit 11f performs processing for acquiring feedback information from the doctor 102 who has administered the medicine to the patient 101 . In the present embodiment, the feedback information acquisition unit 11f obtains a drug from the values corrected by the doctor 102 for the drug type, dosage, and administration rate predicted by the drug prediction model 12b, and the result of administering the drug to the patient 101. information such as the judgment by the doctor 102 as to whether the type, dosage and administration rate of the medicine were appropriate, and the diagnosis result by the doctor 102 of the change in the condition of the patient 101 due to the administration of the medicine (for example, recovered, worsened, etc.) Get it as feedback information. These pieces of feedback information are input to the terminal device 3 by the doctor 102 , and the terminal device 3 transmits the input feedback information to the server device 1 . The feedback information acquisition unit 11f acquires the feedback information transmitted from the terminal device 3 by receiving the feedback information by the communication unit 13, and associates the acquired feedback information with the patient ID of the target patient 101 and stores it in the drug administration DB 8. Remember.

The learning processing unit 11g performs processing for generating the drug prediction model 12b in advance by performing machine learning processing using learning data (teacher data) prepared in advance. In the present embodiment, the learning processing unit 11g generates the drug prediction model 12b by performing so-called supervised machine learning. The drug prediction model 12b according to the present embodiment is a learning model that outputs the drug type, dosage and administration rate in response to the input of medical record data. For example, a neural network learning model is used for the drug prediction model 12b, and the learning processing unit 11g determines internal parameters of the neural network by performing supervised learning using learning data, and stores the determined parameters in the storage unit. 12 as a drug prediction model 12b. The processing of supervised learning of the learning model is an existing technology, so a detailed description is omitted. Model learning process can be performed.

Further, in order to improve the prediction accuracy of the drug prediction model 12b, the learning processing unit 11g may further perform machine learning processing on the learned drug prediction model 12b using the information accumulated in the drug administration DB 8. good. In this case, the learning processing unit 11g may, for example, convert the information stored in the drug administration DB 8 as appropriate to generate learning data. learning data may be generated based on the information stored in , and the learning processing unit 11g may re-learn the drug prediction model 12b using the generated learning data.

In the present embodiment, the server device 1 performs the process of generating the drug prediction model 12b by machine learning, but the invention is not limited to this, and a device other than the server device 1 generates the drug prediction model 12b. may be performed, and information such as parameters of the drug prediction model 12b resulting from the generation may be transmitted to the server device 1 . In the present embodiment, the drug prediction model 12b is a neural network learning model, but it is not limited to this, and various learning models such as SVM (Support Vector Machine) or decision tree can be employed. Also, although the learning processing unit 11g generates the drug prediction model 12b by supervised learning, it is not limited to this. The learning processing unit 11g may generate the drug prediction model 12b based on a reinforcement learning technique, for example, by giving a reward according to the appropriateness of the administered drug. Further, for example, the learning processing unit 11g may generate the drug prediction model 12b by ensemble learning in which a plurality of types of learning models are combined.

FIG. 3 is a block diagram showing the configuration of the terminal device 3 according to this embodiment. The terminal device 3 according to the present embodiment includes a processing unit 31, a storage unit (storage) 32, a communication unit (transceiver) 33, a display unit (display) 34, an operation unit 35, a camera 36, and the like. . The terminal device 3 is a device used by a doctor 102 who administers medicine to the patient 101, and can be configured using an information processing device such as a smart phone, a tablet terminal device, or a personal computer, for example.

The processing unit 31 is configured using an arithmetic processing unit such as a CPU or MPU, a ROM, and the like. The processing unit 31 reads out and executes the program 32a stored in the storage unit 32, thereby acquiring the patient ID assigned to the patient 101, and acquiring and displaying the prediction result of drug administration by the server device 1. processing, and various processing such as processing for receiving input of feedback information regarding drug administration from the doctor 102 .

The storage unit 32 is configured using, for example, a nonvolatile memory device such as a flash memory or a storage device such as a hard disk. The storage unit 32 stores various programs executed by the processing unit 31 and various data required for processing by the processing unit 31 . In the present embodiment, the storage unit 32 stores a program 32a executed by the processing unit 31. FIG. In this embodiment, the program 32a is distributed by a remote server device or the like, and the terminal device 3 acquires it through communication and stores it in the storage unit 32. FIG. However, the program 32a may be written in the storage unit 32 during the manufacturing stage of the terminal device 3, for example. For example, the program 32a may be stored in the storage unit 32 after the terminal device 3 reads the program 32a recorded in the recording medium 98 such as a memory card or an optical disk. For example, the program 32 a may be recorded in the recording medium 98 and read by a writing device and written in the storage unit 32 of the terminal device 3 . The program 32 a may be provided in the form of distribution via a network, or may be provided in the form of being recorded on the recording medium 98 .

The communication unit 33 communicates with various devices via a network N including a mobile phone communication network, a wireless LAN, the Internet, and the like. In the present embodiment, the communication unit 33 communicates with the server device 1 via the network N. FIG. The communication unit 33 transmits data received from the processing unit 31 to other devices, and provides the processing unit 31 with data received from other devices.

The display unit 34 is configured using a liquid crystal display or the like, and displays various images, characters, etc. based on the processing of the processing unit 31 . The operation unit 35 receives a user's operation and notifies the processing unit 31 of the received operation. For example, the operation unit 35 receives a user's operation using an input device such as mechanical buttons or a touch panel provided on the surface of the display unit 34 . Further, for example, the operation unit 35 may be an input device such as a mouse and a keyboard, and these input devices may be detachable from the terminal device 3 .

The camera 36 is configured using an imaging device such as a CCD (Charge Coupled Device) image sensor or a CMOS (Complementary Metal Oxide Semiconductor) image sensor. The camera 36 is arranged, for example, at an appropriate location on the housing of the terminal device 3 . The camera 36 provides the processing unit 31 with data of an image (moving image) captured by the imaging device. Note that the camera 36 may be built in the terminal device 3 or may be detachable from the terminal device 3 .

In the terminal device 3 according to the present embodiment, the program 32a stored in the storage unit 32 is read out and executed by the processing unit 31, so that the patient ID acquisition unit 31a, the drug administration information acquisition unit 31b, and the feedback information reception unit 31c and the like are implemented in the processing unit 31 as software functional units.

The patient ID acquisition unit 31a performs processing for acquiring the patient ID assigned to the patient 101 to whom the drug is administered. In this embodiment, the two-dimensional code 110 of the patient ID is printed, for example, on the patient's 101 clothes, bed, hospital room, or medical record printed on a paper medium. 36 is used to capture the two-dimensional code 110 to acquire the patient ID. The patient ID acquisition unit 31a acquires an image of the two-dimensional code 110 captured by the camera 36 and converts the two-dimensional code 110 shown in the acquired image into a patient ID, thereby acquiring the patient ID. The patient ID acquisition unit 31 a transmits the acquired patient ID to the server device 1 .

In the present embodiment, the two-dimensional code 110 is used to acquire the patient ID, but the invention is not limited to this. may be obtained in any manner. For example, the doctor 102 may use the operation unit 35 of the terminal device 3 to directly input the patient ID of numbers and letters, or the patient ID stored as data in an IC (Integrated Circuit) tag or the like. may be acquired by the terminal device 3 through short-range wireless communication.

The drug administration information acquisition unit 31b performs processing for acquiring drug administration information including information such as the type of drug predicted by the server device 1, the dose, and the administration rate. When the patient ID acquisition unit 31a transmits the patient ID, the server device 1 predicts the type, dosage, and administration rate of the drug to be administered to the patient 101 of this patient ID, and outputs the prediction result as drug administration information to the terminal. Send to device 3. The medicine administration information acquisition unit 31 b acquires the medicine administration information transmitted by the server device 1 by receiving the information at the communication unit 13 and displays the acquired medicine administration information on the display unit 34 . The doctor 102 determines the type, dosage and administration rate of the drug to be administered to the patient 101 based on the type, dosage and administration rate of the drug displayed on the terminal device 3 . Based on this determination, the doctor 102 or a nurse or the like can administer the drug to the patient 101 by setting the drug in the pump 5, setting the dosage and administration rate, and operating the pump 5. . However, the type, dosage and administration rate of the drug to be administered to the patient 101 need not be exactly the same as the information displayed on the terminal device 3, and can be modified or changed by the doctor 102 as appropriate.

The feedback information receiving unit 31c performs processing for receiving input of feedback information regarding drug administration from the doctor 102 who has determined the type, dose, and administration rate of the drug to be administered to the patient 101 . For example, the feedback information receiving unit 31c displays the type, dosage, and administration speed of the medicine based on the medicine administration information acquired from the server device 1, and the doctor 102 Inquires whether or not the correction has been made, and if it has been corrected, accepts the input of the corrected drug type, dosage or administration rate. Further, for example, the feedback information receiving unit 31c inquires of the doctor 102 whether or not the type, dose, and administration rate of the drug administered to the patient 101 were appropriate after a predetermined time has elapsed since the drug was administered, and whether or not the drug was appropriate. If the answer is inappropriate, it accepts the input of the kind of drug, dosage and administration rate that the doctor 102 considers appropriate. Further, for example, the feedback information reception unit 31c inquires whether the condition of the patient 101 has recovered or deteriorated after a predetermined time has passed since the administration of the medicine, and receives input of the answer. The feedback information reception unit 31c receives these inputs by the doctor 102 and transmits the input information to the server device 1 as feedback information.

<Drug prediction model generation processing>
FIG. 4 is a schematic diagram for explaining the configuration of drug prediction model 12b according to the present embodiment. The drug prediction model 12b according to the present embodiment receives as input the chart data of the patient 101 to whom the drug is administered, and is generated in advance by machine learning so as to output the type, dosage, and administration rate of the drug to be administered to this patient 101. It is a learning model that In the present embodiment, the drug prediction model 12b may employ, for example, a neural network learning model.

The medical chart data input to the drug prediction model 12b may include various information such as the patient's 101 age, weight, primary disease, test values, and treatment history. The chart data includes, for example, information input by the doctor 102 when examining the patient 101, results of various examinations performed on the patient 101, information on surgery performed on the patient 101, or data on the patient 101. Information may be included, such as information about medications previously administered to the patient. A plurality of pieces of information included in medical record data are input to the drug prediction model 12b. If the information contained in the medical record data is numerical information, this numerical information may be directly input to the drug prediction model 12b. 12b.

If the information contained in the chart data is information such as character strings instead of numerical information, preprocessing is performed to appropriately convert this information into numerical information, and the converted numerical information is input to the drug prediction model 12b. Learning models such as RNN (Recurrent Neural Network), LSTM (Long Short Term Memory), or BERT (Bidirectional Encoder Representations from Transformers) that perform natural language processing can be used for conversion from character string information to numerical information. . Alternatively, the drug prediction model 12b may be a learning model that performs natural language processing, and character string information may be directly input to the drug prediction model 12b.

The chart data may also include images such as X-rays or CT (Computed Tomography). The images included in the chart data are pre-converted into numerical information such as feature amount vectors of other dimensions, and input to the drug prediction model 12b. A learning model that handles images, such as a CNN (Convolutional Neural Network), can be used for conversion from images to numerical information. Alternatively, the drug prediction model 12b may be a learning model that handles images, and an image may be directly input to the drug prediction model 12b.

The drug prediction model 12b according to the present embodiment outputs information on the type, dosage and administration rate of the drug to be administered to the patient 101 of the input chart data. In the example of FIG. 4, one drug prediction model 12b is shown as outputting three types of information of drug type, dosage and administration rate, but it is not limited to this. The drug prediction model 12b is divided into three learning models, for example, a learning model that outputs the type of drug in response to the input of medical record data, a learning model that outputs the dosage, and a learning model that outputs the administration rate. may Further, in this case, the type of drug may be input together with the chart data to the learning model that outputs the dose and the learning model that outputs the administration rate.

The drug prediction model 12b outputs, for example, numerical values corresponding to the number of drug types, and each numerical value indicates the likelihood (certainty) of administering the corresponding drug. In the example of FIG. 4, they are shown as output nodes of "medicine A", "medicine B", "medicine C", and so on. The drug corresponding to the maximum value among the plurality of numerical values output by the drug prediction model 12b is the type of drug to be administered to the patient 101 of the input chart data.

The drug prediction model 12b also outputs, for example, the numerical value of the dose of the drug and the numerical value of the administration rate. In the example of FIG. 4, they are shown as output nodes for "Dose" and "Dose Rate". In this example, the drug prediction model 12b is assumed to output the dose and the rate of administration of the drug as one value, respectively, but the present invention is not limited to this. It may be configured to output the dose and administration rate for each type of drug, such as the dose and rate of administration, the dose and rate of administration of drug C, and so on.

Generation of drug prediction model 12b according to the present embodiment is performed by so-called supervised machine learning using learning data prepared in advance. The designer or administrator of the information processing system according to the present embodiment, based on the past medical record data and drug administration data accumulated in the medical institution, the medical record data of the patient who administered the drug, Data that associates the type, dosage, and administration rate of an appropriate drug administered to the server device 1 is generated as learning data (teacher data) and stored in the storage unit 12 of the server device 1 .

FIG. 5 is a flow chart showing the procedure of generation processing of the drug prediction model 12b performed by the server device 1 according to the present embodiment. The learning processing unit 11g of the processing unit 11 of the server device 1 reads learning data pre-stored in the storage unit 12 (step S1). The learning processing unit 11g acquires one or a plurality of data from among a plurality of data (data in which the medical record data is associated with the drug type, dosage, and administration speed) included in the read learning data ( step S2). The learning processing unit 11g inputs the chart data included in the data acquired in step S2 to the drug prediction model 12b (step S3). In the learning processing unit 11g, the drug prediction model 12b outputs the information on the drug type, dosage, and administration rate included in the data acquired from the learning data in step S2, and the input of the medical record data in step S3. Errors from the information on the drug type, dosage and administration rate are calculated (step S4). The learning processing unit 11g updates the parameters of the drug prediction model 12b based on the error calculated in step S4 (step S5).

The learning processing unit 11g determines whether or not the processing of steps S2 to S5 has been completed for all the data included in the learning data read out in step S1 (step S6). If the processing has not been completed for all the data (S6: NO), the learning processing unit 11g returns the processing to step S2, acquires another data from the learning data, and repeats the same processing. When the processing has been completed for all data (S6: YES), the learning processing unit 11g ends the generation processing of the drug prediction model 12b. In the illustrated procedure, the learning processing unit 11g performs machine learning using the learning data once and ends the processing. can be repeated.

<Drug prediction processing>
In the information processing system according to the present embodiment, the drug prediction model 12b generated by the machine learning described above is stored in advance in the storage unit 12 of the server device 1, and the server device 1 uses the drug prediction model 12b to predict the patient 101. Predicting the type, dose, and administration rate of the drug to be administered to the When administering a drug to a patient 101 using the pump 5 , for example, a doctor 102 in charge of the patient 101 uses his own terminal device 3 to acquire the patient ID of the patient 101 . In this embodiment, the patient ID is acquired by reading the two-dimensional code 110 with the terminal device 3 . The terminal device 3 converts the read two-dimensional code 110 into a patient ID and transmits the patient ID to the server device 1 .

The server device 1 acquires the patient ID transmitted by the terminal device 3, and reads the chart data corresponding to this patient ID from the chart DB7. The server device 1 appropriately preprocesses a plurality of pieces of information contained in the read medical chart data, thereby converting the medical chart data into data to be input to the drug prediction model 12b. The server device 1 inputs the input data converted from the medical record data to the drug prediction model 12b, and acquires the drug type, dosage and administration rate output by the drug prediction model 12b. The server device 1 transmits to the terminal device 3 the drug type, dosage and administration rate predicted by the drug prediction model 12b as drug administration information.

The terminal device 3 receives the medicine administration information transmitted by the server device 1 and performs processing for displaying the received information on the display unit 34 . FIG. 6 is a schematic diagram showing a display example of drug administration information. On the notification screen of the medicine administration information displayed by the terminal device 3 based on the medicine administration information received from the server device 1, for example, a title character string of "medicine administration information" is displayed at the top, and the title character string is displayed below the title character string. , information related to the target patient 101, such as "patient ID" and "patient name", is displayed. In this example, the "patient ID" is "A1234" and the "patient name" is "XXXX". The information about the patient 101 displayed by the terminal device 3 on this screen is, for example, information included in the medical chart data of the patient 101. The server device 1 acquires these information necessary for display from the medical chart data, and displays drug administration information. together with the terminal device 3. Any information may be displayed as the information about the patient 101 .

In addition, the terminal device 3 displays the prediction results of the drug prediction model 12b below the information about the patient 101, namely, "drug type", "dosage" and "administration rate". In this example, the "medicine type" is "medicine A", the "administration amount" is "5.8", and the "administration rate" is "0.4". Based on this information displayed on the terminal device 3, the doctor 102 sets the specified drug type (medicine A) to the pump 5 by the specified dose (5.8), The pump 5 can be activated with the dose rate setting set to the designated dose rate (0.4). Operations such as setting the medicine to the pump 5 and setting the administration rate may not be directly performed by the doctor 102 but may be performed by a nurse or the like upon receiving an instruction from the doctor 102 .

In this embodiment, the doctor 102 has the final right to decide the type, dosage and administration rate of the drug to be administered to the patient 101. Drug administration to the patient 101 may be performed at different drug types, doses, or administration rates. When the doctor 102 makes such modifications to the drug type, dosage, and administration rate, the terminal device 3 collects feedback information about what kind of corrections have been made. The notification screen is provided with an input box for receiving the input of the correction value.

In the illustrated example, the information of "medicine type", "dosage amount" and "administration rate" are displayed vertically on the lower left side of the notification screen of the drug administration information. , three input boxes corresponding to "medicine type", "dosage amount", and "administration rate" are vertically arranged and displayed on the lower right side of the drug administration information notification screen. The input box corresponding to the "medicine type" is a so-called pull-down box, pull-down menu, drop-down list, or the like, and the doctor 102 can select one of a plurality of pre-registered drugs. is possible. The doctor 102 directly inputs numerical values into the input boxes corresponding to "dosage" and "administration rate".

When the doctor 102 inputs the modified values of the drug type, dosage and administration rate, the terminal device 3 transmits the input values to the server device 1 as feedback information. The server device 1 receives the feedback information transmitted from the terminal device 3 and stores it in the medicine administration DB 8 . As a result, for example, for the medical chart data of the patient 101, it is assumed that the modified values input by the doctor 102 are suitable instead of the drug type, dosage, and administration rate predicted by the drug prediction model 12b. and the drug type, dosage amount, and administration rate modified by the doctor 102 are associated with each other to generate learning data, which can be used for re-learning the drug prediction model 12b.

FIG. 7 is a flow chart showing the procedure of drug prediction processing performed by the server device 1 according to the present embodiment. The patient ID acquisition unit 11a of the processing unit 11 of the server device 1 according to this embodiment determines whether or not the patient ID of the patient 101 to whom the drug is administered has been received from the terminal device 3 (step S21). If the patient ID has not been received, the patient ID acquisition unit 11a waits until the patient ID is received from the terminal device 3 .

When the patient ID is received (S21: YES), the chart data acquisition unit 11b of the processing unit 11 reads the chart data corresponding to the patient ID acquired in step S21 from the chart DB 7 (step S22). The drug prediction processing unit 11c of the processing unit 11 inputs the medical record data of the patient 101 read in step S22 to the drug prediction model 12b (step S23), and calculates the type, dosage, and administration of the drug output by the drug prediction model 12b. A speed is acquired (step S24). The drug administration information output unit 11d of the processing unit 11 transmits (outputs) the drug administration information including the drug type, dose, and administration rate acquired in step S24 to the terminal device 3 (step S25). to display drug administration information.

The feedback information acquisition unit 11f of the processing unit 11 determines whether or not the type, dose, and administration speed of the drug have been corrected in the terminal device 3 displaying the drug administration information (step S26). If the correction has been made (S26: YES), the feedback information acquisition unit 11f acquires the corrected information on the drug type, dosage, and administration rate from the terminal device 3, and uses the acquired information as feedback information for drug administration. Store in DB 8 (step S27), and terminate the process. If no correction has been made (S26: NO), the processing unit 11 terminates the process.

FIG. 8 is a flow chart showing the procedure of drug prediction processing performed by the terminal device 3 according to the present embodiment. The patient ID acquisition unit 31a of the processing unit 31 of the terminal device 3 according to the present embodiment acquires the two-dimensional code 110 of the patient 101 by photographing with the camera 36 (step S41). The patient ID acquisition unit 31a acquires the patient ID by converting the acquired two-dimensional code 110 into the patient ID (step S42). The patient ID obtaining unit 31a transmits the obtained patient ID to the server device 1 (step S43).

The drug administration information acquisition unit 31b of the processing unit 31 determines whether or not the drug administration information transmitted by the server device 1 in response to the transmission of the patient ID has been received (step S44). If the medicine administration information has not been received ( S<b>44 : NO), the medicine administration information acquisition unit 31 b waits until the medicine administration information is received from the server device 1 . When the medicine administration information is received (S44: YES), the medicine administration information acquisition unit 31b displays the acquired medicine administration information on the display unit 34 (step S45).

The feedback information reception unit 31c of the processing unit 31 determines whether or not the doctor 102 has corrected the type, dosage, and administration speed of the drug administration information displayed on the display unit 34 (step S46). . If the drug type, dosage and administration rate have been corrected (S46: YES), the feedback information reception unit 31c acquires the corrected drug type, dosage and administration rate, and sends this information as feedback information. to the server device 1 (step S47), and the process ends. If the drug type, dosage and administration rate have not been corrected (S46: NO), the processing unit 31 terminates the process.

<Information collection processing>
In the information processing system according to the present embodiment, the server device 1 performs an information collection process of collecting information on drug administration to the patient 101 and storing and accumulating the information in the drug administration DB 8 . When the doctor 102 corrects the drug type, dosage and administration rate on the above-described drug administration information notification screen, the process of storing the corrected values in the drug administration DB 8 as feedback information is also part of this information collection process. is one. Note that the server device 1 may store the type, dosage, and administration rate of the drug predicted by the drug prediction model 12b in the drug administration DB 8 when the doctor 102 does not correct them.

In addition to collecting the above information, the server device 1 according to the present embodiment collects (1) vital information of the patient 101 to whom the medicine is administered, Information on suitability and (3) information on the doctor's 102 judgment on the patient 101 to whom the drug was administered (suitability of the drug, state of the patient 101, etc.) are collected and stored in the drug administration DB 8 .

(1) Vital information of a patient administered a drug In the information processing system according to the present embodiment, the heart rate, respiration rate, blood pressure, body temperature, etc. measured by the measuring device 6 for the patient 101 to whom the drug is administered by the pump 5 of vital information is measured. The measuring instrument 6 displays the measured vital information on its own display unit and also transmits the vital information to the server device 1 via the network. The server device 1 receives the vital information transmitted by the measuring instrument 6 and stores the vital information in the drug administration DB 8 in association with information such as the patient ID and date and time.

The measuring instrument 6 performs measurement at a predetermined cycle such as 1 second or 1 minute, and transmits the measurement result to the server device 1 as vital information. Instead of storing in the drug administration DB 8, the vital information may be stored at a frequency such as once every several tens of seconds or once every few minutes. A minimum value or the like may be calculated and these calculation results may be stored in the medicine administration DB 8 .

(2) Information on the appropriateness of the type, dosage, and administration rate of the drug administered to the patient Based on the vital information received from the measuring instrument 6, the server device 1 determines whether the patient 101 to whom the drug was administered Determine the presence or absence of anomalies. For example, in the server device 1, thresholds for determining abnormalities are set in advance for each information such as heart rate, respiration rate, blood pressure, and body temperature included in the vital information, and the vital information received from the measuring instrument 6 The presence or absence of patient 101 is determined by comparing each piece of information contained with a corresponding threshold value. When the server device 1 determines that the patient 101 has an abnormality, the vital information of the patient 101 at this time is associated with information such as a flag indicating that the patient 101 has been determined to be abnormal, and is stored in the medicine administration DB 8 . Note that the threshold for abnormality determination is set in advance by the doctor 102, for example. Any method may be used to determine the presence or absence of abnormality based on vital information.

In addition, the server device 1 according to the present embodiment, for example, when a predetermined time (for example, several tens of minutes, several hours, etc.) has passed without any abnormality since the start of administration of the medicine to the patient 101, the measuring instrument 6 Based on a plurality of pieces of vital information received from and stored in the drug administration DB 8, the suitability of the drug administered to the patient 101 is determined. For example, the server device 1 compares the vital information immediately after the start of drug administration with the vital information after a predetermined period of time has elapsed, and determines whether the vital information of the patient 101 has improved or deteriorated to determine whether the drug is suitable. determine whether or not The server apparatus 1 determines that the patient 101 has improved when the vital information after the lapse of a predetermined period of time has changed from the vital information immediately after the start of administration of the drug by exceeding the threshold in the improvement direction, and determines that the patient 101 has improved. If there is no change exceeding any threshold, it can be determined that there is no change. The server apparatus 1 determines that the drug was suitable when the patient 101 improved, and determined that the drug was not suitable when the patient 101 deteriorated. Whether or not the medicine is suitable when there is no change in vital information can be appropriately determined according to, for example, the condition of the patient 101 or the type of medicine administered.

FIG. 9 is a flow chart showing the procedure of information collection processing on vital information by the server device 1 according to the present embodiment. The vital information acquisition unit 11e of the processing unit 11 of the server device 1 according to the present embodiment receives vital information such as heart rate, respiration rate, blood pressure, body temperature, etc., measured for the patient 101 to whom the drug is administered from the measuring instrument 6. It is determined whether or not (step 61). If the vital information has not been received ( S<b>61 : NO), the vital information acquisition unit 11 e waits until the vital information is received from the measuring instrument 6 . When the vital information is received (S61: YES), the vital information acquisition unit 11e stores the received vital information in the medicine administration DB 8 (step S62).

In addition, the vital information acquisition unit 11e determines whether or not there is an abnormality in the vital information of the patient 101 by comparing one or more measured values included in the received vital state with a predetermined threshold value. (Step S63). If there is an abnormality in the vital information (S63: YES), the vital information acquisition unit 11e notifies the terminal device 3 of the abnormality by, for example, sending a message (step S64). The vital information acquisition unit 11e stores information indicating that an abnormality has occurred due to administration of the medicine in the medicine administration DB 8 (step S65), and terminates the process.

If there is no abnormality in the vital information (S63: NO), the vital information acquisition unit 11e determines whether or not a predetermined time has passed since the medicine was administered to the patient 101 (step S66). If the predetermined time has not elapsed (S66: NO), the vital information acquisition section 11e returns the process to step S61 and repeats the above-described process. If the predetermined time has passed (S66: YES), the vital information acquisition unit 11e determines whether the medicine administered to the patient 101 is appropriate based on a plurality of pieces of vital information stored from the start of medicine administration (step S67). The vital information acquiring unit 11e stores the determination result of the suitability of the medicine as feedback information in the medicine administration DB 8 (step S68), and ends the process.

Based on the information stored in the drug administration DB 8, for example, regarding the administration of a drug determined to be appropriate, the medical record data of the patient 101 when the drug was administered, and the type, dosage, and administration rate of the drug determined to be appropriate. can be generated and used for re-learning of the drug prediction model 12b. It should be noted that the determination results based on the vital information may be stored in the drug administration DB 8 and the vital information itself used for the determination may not be stored in the drug administration DB 8 .

(3) Information about doctor's judgment about the patient to whom the drug is administered The server device 1 according to the present embodiment, for example, waits for a predetermined time (for example, several tens of minutes, several hours, etc.) after starting the administration of the drug to the patient 101. after the elapse of time, inquire of the doctor 102 whether the type, dosage and administration rate of the administered drug were appropriate, and the condition of the patient 101 after administration, etc., and obtain the reply from the doctor 102 as feedback information Store in the drug administration DB 8 . For example, when a predetermined time has passed since the prediction result of the drug prediction model 12b was transmitted to the terminal device 3 as drug administration information, the server device 1 causes the terminal device 3 to display an inquiry screen for progress information. , to obtain feedback information from the physician 102 .

FIG. 10 is a schematic diagram showing a display example of a progress information inquiry screen. The terminal device 3 displays the illustrated inquiry screen on the display unit 34 in response to the instruction from the server device 1 when a predetermined time has passed since the administration of the drug to the patient 101 was started. In the illustrated inquiry screen, for example, a title string of "Progress after drug administration" is displayed at the top. Information such as name is displayed.

The terminal device 3 displays the item name "<administered drug>" below the information about the patient 101 who administered the drug, and the information about the drug administered to this patient 101, "type of drug" and "dosage". and "dosing rate". In addition, when the doctor 102 determines that the drug administered to the patient 101 is not suitable and that the drug needs to be corrected, the terminal device 3 displays an input box for accepting input of information on drug correction on an inquiry screen. set in In the example shown in the figure, for "type of drug", "dosage amount" and "administration rate" of "<administered drug>" provided on the left side of the inquiry screen, "correction required" and a rightward arrow are placed between them. The terminal device 3 displays three input boxes respectively corresponding to "medicine type", "dosage amount" and "administration rate". Below these, a check box for selecting "no correction" is provided to reply that the doctor 102 judged that the administered drug was suitable.

In addition, the terminal device 3 displays the item name of “<Patient's condition>” and three check boxes for accepting selection by the doctor 102 below the item of “<Administered drug>”. The three check boxes are for accepting any selection of "improvement", "no change", or "worse" for the patient's 101 medical condition.

There is also a button labeled "Answer" at the bottom right of the inquiry screen. When a click operation, a touch operation, or the like is performed on the "answer" button, the terminal device 3 transmits information input to the inquiry screen to the server device 1 as feedback information. The server device 1 acquires the feedback information transmitted by the terminal device 3 and stores it in the medicine administration DB 8 .

FIG. 11 is a flow chart showing the procedure of feedback information acquisition processing performed by the server apparatus 1 according to the present embodiment. The feedback information acquisition unit 11f of the processing unit 11 of the server device 1 according to this embodiment determines whether or not a predetermined time has passed since the start of drug administration to the patient 101 (step S81). If the predetermined time has not passed (S81: NO), the feedback information acquiring unit 11f waits until the predetermined time has passed. If the predetermined time has passed (S81: YES), the feedback information acquiring unit 11f gives a display instruction or the like to the terminal device 3 to display a progress information inquiry screen on the display unit 34, so that the doctor 102 can see the progress regarding drug administration. (step S82).

The feedback information acquiring unit 11f determines whether or not there is an answer from the doctor 102 to the progress inquiry based on whether or not the feedback information is transmitted from the terminal device 3 (step S83). If there is no answer (S83: NO), the feedback information acquisition unit 11f waits until an answer is obtained from the terminal device 3. If there is an answer (S83: YES), the feedback information acquisition unit 11f obtains feedback information including information such as the suitability of the drug that the terminal device 3 has received as an answer to the inquiry and changes in the patient's 101 medical condition. The information is stored in the drug administration DB 8 (step S84), and the process ends.

Based on the information stored in the drug administration DB 8, for example, regarding the administration of a drug that the doctor 102 has determined to be appropriate and that the patient's condition has improved, medical record data of the patient 101 when the drug was administered, It is possible to generate learning data in which the type of drug, dosage amount, and administration rate determined to be appropriate by the doctor 102 are associated and used for re-learning of the drug prediction model 12b.

<Summary>
In the information processing system according to the present embodiment having the above configuration, the server device 1 acquires the medical chart data of the patient 101, inputs the medical chart data to the drug prediction model 12b that has undergone machine learning in advance, and the drug prediction model 12b The drug administration information is displayed on the display unit 34 of the terminal device 3 by acquiring the type, dosage, and administration rate of the drug to be output and transmitting the acquired information to the terminal device 3 as the drug administration information. As a result, the information processing system according to the present embodiment can be expected to assist the doctor 120 and the like in administering drugs to the patient 101 .

Further, in the information processing system according to the present embodiment, the server device 1 obtains the patient ID from the terminal device 3 that reads the patient ID of the patient 101 from the two-dimensional code 110, and stores the medical record data of a plurality of patients in the medical record DB 7. , the server device 1 acquires the chart data corresponding to the acquired patient ID. Thereby, the server device 1 can easily acquire medical record data to be input to the drug prediction model 12b.

Further, in the information processing system according to the present embodiment, the server device 1 receives through the terminal device 3 corrections made by the doctor 102 to the type, dose, and administration rate of the drug predicted by the drug prediction model 12b, and the corrected drug information on the type, dose and rate of administration are stored in the drug administration DB 8 as feedback information. In the information processing system according to the present embodiment, the server device 1 acquires the vital information measured by the measuring device 6 for the patient 101 to whom the drug is administered, and stores the acquired vital information in the drug administration DB 8 as feedback information. Further, in the information processing system according to the present embodiment, based on the vital information acquired from the measuring instrument 6, the server device 1 determines the appropriateness of the drug type, dosage and administration rate predicted by the drug prediction model 12b, The determination result is stored in the drug administration DB 8 as feedback information. In addition, in the information processing system according to the present embodiment, information from the doctor 102 regarding the suitability of a drug administered to the patient 101 or an event occurring in response to the administration of the drug (for example, a change in the patient's condition) The server device 1 accepts the input of through the terminal device 3, and stores the accepted information in the drug administration DB 8 as feedback information.

By storing and accumulating such feedback information in the drug administration DB 8, the accumulated information can be used to generate learning data for learning (relearning) of the drug prediction model 12b. The server device 1 can acquire learning data generated based on the feedback information stored in the drug administration DB 8 and re-learn the drug prediction model 12b. As a result, it can be expected to improve the accuracy of the drug type, dose, and administration rate output by the drug prediction model 12b.

Further, in the information processing system according to the present embodiment, the server device 1 acquires and acquires learning data in which the patient's chart data and the type, dosage, and administration rate of the drug to be administered to the patient are associated with each other. The drug prediction model 12b is generated using the obtained learning data. As a result, the information processing system can predict the appropriate drug type, dosage and administration rate to be administered to the patient 101 using the machine learning drug prediction model 12b. It can be expected to support the work of 102 and the like.

Further, in the information processing system according to the present embodiment, the server device 1 stores medical record data of the patient 101, the type, dosage, and administration rate of the drug administered to the patient 101, and the vital information of the patient 101 to whom the drug is administered. collects and stores in the drug administration DB 8. As a result, the information processing system can generate learning data for machine learning of the drug prediction model 12b based on the information stored and accumulated in the drug administration DB 8 .

The embodiments disclosed this time are illustrative in all respects and should be considered not 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.

1 server device 3 terminal device 5 pump 6 measuring instrument 7 chart DB
8 Drug administration database
11 processing unit 11a patient ID acquisition unit 11b chart data acquisition unit 11c drug prediction processing unit 11d drug administration information output unit 11e vital information acquisition unit 11f feedback information acquisition unit 11g learning processing unit 12 storage unit 12a server program 12b drug prediction model 13 communication Part 31 Processing Part 31a Patient ID Acquisition Part 31b Drug Administration Information Acquisition Part 31c Feedback Information Receiving Part 32 Storage Part 32a Program 33 Communication Part 34 Display Part 35 Operation Part 36 Cameras 98, 99 Recording Medium 101 Patient 102 Doctor 110 Two-Dimensional Code N network

Claims (14)

The information processing device
Get patient chart data,
inputting the acquired medical chart data into a learning model that has undergone machine learning so as to output the type, dosage, and administration rate of a drug to be administered to the patient in response to the patient's medical chart data input; Acquire the type, dosage and administration rate of the drug output by
Output the type, dosage and administration rate of the obtained drug,
Information processing methods. The information processing device
obtaining identification information of the patient;
Obtaining medical chart data corresponding to the obtained identification information from a database storing medical chart data of a plurality of patients;
The information processing method according to claim 1 . The information processing device performs processing to display the type, dosage, and administration rate of the drug acquired from the learning model on the display unit;
The information processing method according to claim 1 or 2. The information processing device
Receiving modifications to the drug type, dosage and administration rate obtained from the learning model;
storing modified drug type, dose and rate information;
The information processing method according to any one of claims 1 to 3. The information processing device
Acquiring vital information of the patient who administered the drug based on the drug type, dosage and administration rate obtained from the learning model,
storing the obtained vital information;
The information processing method according to any one of claims 1 to 4. The information processing device
Based on the acquired vital information, determine the appropriateness of the drug type, dosage and administration rate acquired from the learning model,
storing information related to the determination result of suitability;
The information processing method according to claim 5. The information processing device
Receiving input of information regarding the suitability of the drug or events occurring in response to the administration of the drug for the patient to whom the drug was administered based on the drug type, dosage, and administration rate obtained from the learning model. ,
storing the received information;
The information processing method according to any one of claims 1 to 6. The information processing device
Obtain learning data generated based on the stored information,
performing machine learning of the learning model using the acquired learning data;
The information processing method according to any one of claims 4 to 7. The information processing device
Acquiring learning data in which the patient's chart data and the type, dosage, and administration rate of the drug to be administered to the patient are associated,
Using the acquired learning data, machine learning generates a learning model that outputs the type, dosage, and administration rate of the drug to be administered to the patient in response to the input of the patient's medical chart data.
How to generate a learning model. The information processing device
Obtaining medical record data of a patient, the type, dosage and administration rate of the drug administered to the patient, and vital information of the patient who administered the drug based on the type, dosage and administration rate of the drug,
storing in a database the obtained medical chart data, the drug type, dosage and administration rate, and the vital information in association with each other;
How we collect information. to the computer,
Get patient chart data,
inputting the acquired medical chart data into a learning model that has undergone machine learning so as to output the type, dosage, and administration rate of a drug to be administered to the patient in response to the patient's medical chart data input; Acquire the type, dosage and administration rate of the drug output by
A computer program that executes a process of outputting the type, dosage, and administration rate of the obtained drug. to the computer,
Acquiring learning data in which the patient's chart data and the type, dosage, and administration rate of the drug to be administered to the patient are associated,
Using the acquired learning data, executing a process of generating a learning model by machine learning that outputs the type, dosage, and administration rate of the drug to be administered to the patient in response to the input of the patient's medical record data, computer program. a first acquisition unit that acquires patient chart data;
inputting the acquired medical chart data into a learning model that has undergone machine learning so as to output the type, dosage, and administration rate of a drug to be administered to the patient in response to the patient's medical chart data input; A second acquisition unit that acquires the type, dosage and administration rate of the drug output by
An information processing apparatus comprising: an output unit for outputting the type, dosage, and administration rate of the obtained medicine. an acquisition unit that acquires learning data in which the patient's chart data and the type, dosage, and administration rate of the drug to be administered to the patient are associated with each other;
a generating unit that uses the acquired learning data to generate, by machine learning, a learning model that outputs the type, dosage, and administration rate of the drug to be administered to the patient in response to the input of the patient's medical chart data. Information processing equipment.

Images