Learning device, determination device, learned model generation method, and program

Abstract

A learning device of the present invention is provided with: an acquiring means for acquiring biometric information of a patient who may possibly become restless, and biometric information of a non-patient; and a model generating means for using the biometric information of the patient and the biometric information of the non-patient to generate a restlessness determination model for determining whether, on the basis of the biometric information of a subject patient, the subject patient has become restless or has not become restless.

Description

Technical Field

[0001] The present invention relates to a learning device, a determination device, a method for generating a learned model, and a recording medium.

Background Art

[0002] In the fields of medicine and nursing care, patients may fall into an unstable state. When a patient falls into an unstable state, the risks of tube removal, needle removal, removal, falling, or tumbling increase, and the patient may get injured. Therefore, techniques for detecting such an unstable state of a patient in advance are known.

[0003] Patent Document 1 discloses a biological information processing system that determines identification information indicating whether the condition of a target patient has changed compared to a normal state based on the feature amount of the biological information of the target patient to be input, and estimates coping information for the target patient based on the identification information and pre-learned coping prediction parameters.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In order to reduce the risks of tube removal, needle removal, removal, falling, or tumbling of a patient, it is necessary to accurately determine whether the patient is in an unstable state. In order to accurately determine whether the patient is in an unstable state, it is preferable to improve the accuracy of the model for determining the unstable state disclosed in Patent Document 1.

[0006] Therefore, the present invention has been made to solve the above problems, and an object thereof is to provide an apparatus or the like capable of improving the accuracy of a model for determining the state of a patient.

Means for Solving the Problems

[0007] In one aspect of the present invention, a learning device includes an acquisition unit that acquires biometric information of a patient who is a person likely to be in an unstable state and biometric information of a non-patient, and uses the biometric information of the patient and the biometric information of the non-patient to generate an instability determination model that determines whether the target patient is in an unstable state or a non-unstable state based on the biometric information of the target patient.

[0008] Further, a determination device in one aspect of the present invention includes a determination unit that determines whether the target patient is in an unstable state using the biometric information of the target patient and an instability determination model, and the instability determination model is a learned model generated by a learning device including an acquisition unit that acquires biometric information of a patient who is a person likely to be in an unstable state and biometric information of a non-patient, and a model generation unit that generates an instability determination model that determines whether the target patient is in an unstable state or a non-unstable state based on the biometric information of the target patient using the biometric information of the patient and the biometric information of the non-patient.

[0009] Further, a method for generating a learned model in one aspect of the present invention includes a computer acquiring biometric information of a patient who is a person likely to be in an unstable state and biometric information of a non-patient, and generating an instability determination model that determines whether the target patient is in an unstable state or a non-unstable state based on the biometric information of the target patient using the biometric information of the patient and the biometric information of the non-patient.

[0010] Further, a recording medium in one aspect of the present invention stores a program that causes a computer to execute a process of generating an instability determination model that determines whether the target patient is in an unstable state or a non-unstable state based on the biometric information of the target patient using the biometric information of a patient who is a person likely to be in an unstable state and biometric information of a non-patient.

Advantages of the Invention

[0011] According to the present invention, the accuracy of a model for determining the state of a patient can be improved.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Modes for Carrying Out the Invention

[0013] Hereinafter, each embodiment of the present invention will be described with reference to the drawings.

[0014] In each embodiment of the present invention, the instability determination model is a learned model for determining whether a patient is in an unstable state. The unstable state indicates a state in which the patient is restless. The unstable state may include a state in which the patient cannot normally control their mind. Further, the unstable state may include a state caused by the patient's delirium. The unstable state may be caused by the patient's mental or physical factors. It is known that patients in an unstable state often exhibit problem behaviors. That is, patients in an unstable state are likely to exhibit problem behaviors. Therefore, by determining whether a patient is in an unstable state, it is possible to predict whether the patient is likely to exhibit problem behaviors. Here, the patient's problem behavior is, for example, an action that requires some kind of countermeasure by a medical staff who performs a recuperation action on the patient in response to the action. The patient's problem behavior is, for example, getting out of bed, walking alone, wandering, going to another floor of the hospital, removing the railing of the bed, falling out of the bed, fiddling with the drip or tubes, removing the drip or tubes, making strange noises, uttering abusive language, using violence, etc. Note that whether the patient's behavior corresponds to a problem behavior may be determined according to the patient's condition. Here, the patient's condition includes at least one state of cognitive function, physical function, and motor function. In each embodiment of the present invention, the instability determination model may determine whether the patient is exhibiting problem behaviors. Hereinafter, the normal state of the patient, that is, the state that is not an unstable state, will be referred to as a non-unstable state.

[0015] In each embodiment of the present invention, a patient is a person who receives medical treatment by a medical professional. The patient may be a person who may become unstable. That is, the patient may be a person with a probability of becoming unstable equal to or higher than a predetermined probability. Also, for example, a person who meets at least one of the characteristics such as having a specific disease, taking a specific drug, having a reduced cognitive function, having blood loss, or having pain in the body is likely to develop an unstable state. The patient may be a person who meets at least one of the above characteristics. Further, the patient may include at least one of inpatients, discharged patients, outpatient patients, etc. Note that the patient is not limited to this as long as the patient is a person to be judged for an unstable state.

[0016] In each embodiment of the present invention, a non-patient is, as an example, a healthy person. A healthy person is a person who meets at least one of the following conditions, for example, can perform daily living activities by oneself, has no underlying disease, and does not require assistance or care from others. The non-patient may be a person with a low probability of becoming unstable. That is, the non-patient may be a person with a probability of becoming unstable equal to or lower than a predetermined probability. Also, the non-patient may be a person who does not meet the characteristics of a person who is likely to develop the above-described unstable state. Note that a person with a probability of becoming unstable equal to or higher than a first threshold may be defined as a patient, and a person with a probability of becoming unstable lower than the first threshold and equal to or lower than a second threshold may be defined as a non-patient.

[0017] In each embodiment of the present invention, biological information is information that changes along with human life activities. That is, biological information is time-series information indicating changes associated with human life activities. Biological information is, for example, at least one of heart rate, heart rate variability, respiratory rate, blood pressure value, body temperature, skin temperature, blood flow rate, blood oxygen saturation, body movement, etc. Biological information may include other information used for determining an unstable state. Biological information is measured, for example, using at least one sensor worn by the measurement subject. Note that the measurement subject includes patients and non-patients. The sensor is, for example, a heart rate sensor, a respiratory rate sensor, a blood pressure sensor, a body temperature sensor, a blood oxygen saturation sensor, an acceleration sensor, etc. The measurement subject may wear a device equipped with one sensor, or may wear a device equipped with a plurality of sensors. The measurement subject may wear a plurality of devices. The device is mainly a wearable device, and specifically, a smartwatch, a smart band, an activity tracker, a clothing sensor, a wearable heart rate sensor, etc. can be mentioned. Also, biological information may be extracted, for example, from image information acquired by an imaging device (such as a camera) installed in the living room of the measurement subject, the voice of the measurement subject, and sound information in the surrounding environment of the measurement subject. Note that the living room of the measurement subject is, for example, a patient's hospital room, etc.

[0018] <First Embodiment> Hereinafter, the configuration of the learning device 10 in the first embodiment will be described. The learning device 10 in the present embodiment generates an instability determination model.

[0019] FIG. 1 is a block diagram showing the configuration of the learning device 10 in the present embodiment. The learning device 10 shown in FIG. 1 includes an acquisition unit 11 and a model generation unit 12.

[0020] The acquisition unit 11 is an acquisition means for acquiring the biological information of patients and the biological information of non-patients. That is, the acquisition unit 11 acquires the biological information of patients, who are persons who may be in an unstable state, and the biological information of non-patients.

[0021] Biological information is, for example, stored in a storage device (not shown) in association with a measurement target person ID that identifies the measurement target person, in association with time information indicating the time when the biological information was measured. The acquisition unit 11 may acquire the biological information of the measurement target person from the storage device.

[0022] The acquisition unit 11 may acquire the biological information of the measurement target person associated with the time information from sensors and devices connected so as to be communicable with the learning device 10 via a communication network such as wireless or wired. Further, the acquisition unit 11 may acquire one or both of the biological information of the patient and the biological information of the non-patient at a predetermined timing.

[0023] Note that the acquisition unit 11 may acquire state information indicating the state of the measurement target person at the time when the biological information was measured. When the measurement target person is a patient, the state is, for example, an unstable state or a non-unstable state. When the measurement target person is a non-patient, the state is, for example, a resting state or a non-resting state which is a state other than the resting state. The resting state will be described later. The state information may be acquired from a storage device, sensors, or devices, similarly to the biological information described above. Further, the state information may be, for example, chart information indicating information described in the medical record of the measurement target person. The state information may be information determined based on information that can be acquired by sensors or devices. Here, examples of information that can be acquired by sensors or devices include pedometer information, position information, and the like. Further, the acquisition unit 11 may acquire only the biological information used for generating the instability determination model in the model generation unit 12 described later.

[0024] The biological information of non-patients may include biological information of non-patients with different attributes. Here, the attributes are, for example, age group, age, gender, etc. That is, the biological information of non-patients may include, for example, biological information of non-patients of different age groups. Here, the age group is a criterion for appropriately determining age. As an example, the age group is divided by the tens digit of the age. For example, a person aged 20 or older and under 30 belongs to the 20s. Note that the attributes are not limited to these as long as they are attributes that may affect biological information.

[0025] The model generation unit 12 is a model generation means for generating an instability determination model using the biological information of patients and the biological information of non-patients. Here, the instability determination model is a model for determining whether a patient is in an unstable state based on the biological information of the patient. Hereinafter, the patient to be the subject of the instability determination may be referred to as the target patient. The model generation unit 12 uses the biological information acquired by the acquisition unit 11. The model generation unit 12 generates an instability determination model using the biological information of patients and the biological information of non-patients as learning data. The model generation unit 12 performs machine learning using the biological information of patients and the biological information of non-patients as learning data, and generates an instability determination model. Here, the learning data includes learning data corresponding to the unstable state and learning data corresponding to the non-unstable state. The learning data corresponding to the unstable state is the biological information labeled with the unstable state. Also, the learning data corresponding to the non-unstable state is the biological information labeled with the non-unstable state. Note that the biological information labeled with the unstable state may be called a positive example, and the biological information labeled with the non-unstable state may be called a negative example.

[0026] As described above, the instability determination model is a model for determining whether a patient is in an unstable state based on the biological information of the patient. The instability determination model takes the biological information of the patient as input and outputs an instability score. The instability score is a value that serves as an indicator indicating an unstable state or a non-unstable state. The instability score is, for example, a value between 0 and 1. In this case, the closer the instability score is to 1, the higher the possibility of being in an unstable state, and the closer the instability score is to 0, the higher the possibility of being in a non-unstable state. For example, a predetermined value between 0 and 1 is used as a threshold, and an unstable state or a non-unstable state is determined based on the threshold. Also, the instability score may be a value expressed by two values of 0 or 1. In this case, the instability score indicates 1 if it is in an unstable state and 0 if it is in a non-unstable state.

[0027] The model generation unit 12 uses the biological information labeled with the label of the unstable state or the non-unstable state as learning data, and performs learning using, for example, a support vector machine (SVM), a neural network, and other known machine learning methods.

[0028] Note that the labeling of the unstable state or the non-unstable state for the biological information acquired by the acquisition unit 11 may be performed by the model generation unit 12 as described later, or may be performed by other devices or users not shown in the figure.

[0029] The learning data corresponding to the non-unstable state includes the biological information of non-patients. That is, the model generation unit 12 generates an instability determination model using the biological information of non-patients labeled with the non-unstable state label.

[0030] The model generation unit 12 generates an instability determination model using the biological information of patients and the biological information of non-patients in a quiet state. That is, the learning data includes the biological information of non-patients in a quiet state. The quiet state is, for example, a sleep state, a relaxed state, etc. The relaxed state includes, as an example, a state without physical or mental load, a state in which the autonomic nerve is in a parasympathetic-dominant state, and the like. At this time, the model generation unit 12 uses the biological information of non-patients in a quiet state as negative example learning data. That is, the biological information of non-patients in a quiet state is labeled with the non-unstable state label.

[0031] Note that the quiet state may be a state other than the non-quiet state. The non-quiet state includes at least one of a state in which the body is moving, a state in which the head is working, and a state in which the body is receiving a stimulus. The quiet state may be a state that does not fall under the above states.

[0032] In addition, the learning data corresponding to the non-unstable state may include the biological information of non-patients in a quiet state. That is, the model generation unit 12 generates an instability determination model using the biological information of non-patients in a quiet state labeled with the non-unstable state label.

[0033] On the one hand, the learning data corresponding to the unstable state includes the biological information of the patient in the unstable state. That is, the model generation unit 12 generates an instability determination model using the biological information of the patient in the unstable state with the label of the unstable state. In this way, the model generation unit 12 may use only the biological information of the patient in the unstable state among the biological information of the patient. Also, the learning data corresponding to the non-unstable state may include the biological information of the patient in the non-unstable state. That is, the model generation unit 12 may generate an instability determination model using the biological information of the patient in the non-unstable state with the label of the non-unstable state.

[0034] In addition, the model generation unit 12 may be configured to determine one or both of the states of the patient and the non-patient corresponding to the biological information. At this time, the model generation unit 12 may label the biological information with an unstable state or a non-unstable state based on the determined state. When the state information is acquired by the acquisition unit 11, the model generation unit 12 may refer to the state information at the time when the biological information is measured to determine one or both of the unstable state or the non-unstable state of the patient and the resting state of the non-patient.

[0035] As an example, the model generation unit 12 may determine whether the non-patient is in a resting state based on one piece of acquired biological information or a combination of a plurality of pieces of biological information. When determining that the non-patient is in a resting state based on one piece of biological information, the model generation unit 12 determines whether the non-patient is in a resting state based on, for example, information indicating the body movement of the non-patient. In this case, as an example, when the value of the information indicating body movement is lower than a predetermined value determined in advance, the model generation unit 12 determines that the non-patient is in a resting state. Here, the information indicating body movement includes the acceleration acquired by the acceleration sensor. When determining that the non-patient is in a resting state based on a combination of a plurality of pieces of biological information, the model generation unit 12 determines that the non-patient is in a resting state based on, for example, a combination of the results of comparing the values of the plurality of pieces of biological information with predetermined values determined in advance, respectively. For example, when the deep body temperature is lower than a predetermined value such as the normal temperature of the non-patient and the respiratory rate is lower than the predetermined value, the model generation unit 12 determines that the patient is in a sleeping state, that is, a resting state.

[0036] Subsequently, the operation performed by the learning device 10 will be described with reference to FIG. 2. FIG. 2 is a flowchart showing an example of the operation performed by the learning device 10. The operation of the learning device 10 may be performed, for example, when a predetermined number or more of biological information is accumulated in a storage device (not shown).

[0037] The acquisition unit 11 acquires the biological information of the patient and the biological information of the non-patient (step S101).

[0038] The model generation unit 12 generates an instability determination model for determining whether the target patient is in an unstable state based on the biological information of the target patient using the biological information of the patient and the biological information of the non-patient (step S102).

[0039] In the instability determination model, learning is performed using learning data in which the collected biological information of the patient is accurately labeled as to whether it is in an unstable state, which leads to an improvement in accuracy.

[0040] However, the biometric information of patients in an unstable state collected includes biometric information of patients in various states. As patients in various states, for example, even if a patient is in an unstable state, it may not be possible to confirm the unstable state from the patient's behavior, and there may be patients who appear calm. In this case, since the patient appears calm at first glance, there is a possibility that the biometric information is labeled as a non-unstable state even though the patient is actually in an unstable state. Thus, there is a possibility that a non-stable determination model is generated using learning data labeled with biometric information of a patient whose unstable state is unclear. As a result, it may be difficult to generate a highly accurate unstable determination model.

[0041] In the learning device 10 according to the present embodiment, in the model generation unit 12, an unstable determination model is generated using biometric information of patients and biometric information of non-patients. Non-patients have fewer cases where the correspondence between the state and the biometric information is unclear compared to the case of the patients described above. That is, the biometric information of non-patients is relatively easier to determine the state than the biometric information of patients, and is likely to be accurately labeled. Therefore, the learning device 10 can perform learning by adding biometric information accurately labeled to the learning data. Then, the learning device 10 can generate an accurate unstable determination model.

[0042] As an example, in the model generation unit 12, the learning device 10 generates an unstable determination model using the biometric information of patients as learning data corresponding to the unstable state and the biometric information of non-patients as learning data corresponding to the non-unstable state. By using such learning data, the difference between the biometric information included in the learning data corresponding to the unstable state and the biometric information included in the learning data corresponding to the non-unstable state becomes clear. Therefore, the learning device 10 can generate an accurate unstable determination model.

[0043] In the learning device 10 according to this embodiment, the model generation unit 12 generates an instability determination model using the biological information of patients and the biological information in the resting state of non-patients. As an example, a patient is a person with a possibility of being in an unstable state equal to or higher than a predetermined probability, while a non-patient is a person with a possibility of being in an unstable state equal to or lower than a predetermined probability. Therefore, the resting state of a non-patient is highly likely to be a non-unstable state, and the biological information in the resting state of a non-patient is highly likely to be accurately labeled as a non-unstable state. Accordingly, the learning device 10 can perform learning by adding the biological information accurately labeled as a non-unstable state to the learning data. Then, the learning device 10 can generate an instability determination model with high accuracy.

[0044] In the learning device 10 according to this embodiment, the model generation unit 12 generates an instability determination model using the biological information in the unstable state of patients and the biological information in the resting state of non-patients. The biological information in the unstable state of patients is highly likely to be accurately labeled as an unstable state because it is labeled as an unstable state after the patient is actually determined to be in an unstable state. Also, as described above, the biological information in the resting state of non-patients is also highly likely to be accurately labeled as a non-unstable state. Accordingly, the learning device 10 can perform learning using the biological information accurately labeled as an unstable state and a non-unstable state. Then, the learning device 10 can generate an instability determination model with high accuracy.

[0045] In this embodiment, the biological information of non-patients can include the biological information of non-patients with different attributes. The biological information may have different characteristics corresponding to different attributes. For example, the heartbeat variation, which is an example of biological information, tends to decrease as the age or years increase. In the learning device 10 in this embodiment, the model generation unit 12 generates an instability determination model using the biological information of the patient and the above-described biological information of non-patients. Thereby, the learning device 10 can widely learn the biological information of non-patients that may have different characteristics corresponding to the differences in the attributes of patients or non-patients, and generate an instability determination model. Therefore, the learning device 10 can generate an instability determination model with high accuracy.

[0046] <Second Embodiment> Hereinafter, the configuration of the instability determination system 200 in the second embodiment will be described. FIG. 3 is a block diagram showing the configuration of the instability determination system 200 in the second embodiment of the present invention. As shown in FIG. 3, the instability determination system 200 in the second embodiment includes a determination device 220, a biological information acquisition device 230, and a determination result output device 240. The determination device 220, the biological information acquisition device 230, and the determination device 220 and the determination result output device 240 are connected so as to be able to communicate with each other via a communication network such as wireless or wired such as Wi-fi or Bluetooth (registered trademark).

[0047] The determination device 220 determines the unstable state of the target patient using the instability determination model. The determination device 220 includes a target patient information acquisition unit 221, a determination unit 222, and an output unit 223. The determination device 220 is realized, for example, in an information terminal such as a computer provided in a medical institution. The determination device 220 may be realized, for example, on a cloud server.

[0048] The target patient information acquisition unit 221 acquires the biological information of the target patient who is the target for determining the unstable state. The target patient information acquisition unit 221 acquires the biological information of the target patient by receiving the biological information used for determining the unstable state of the target patient, which is acquired by the biological information acquisition device 230 described later.

[0049] The determination unit 222 is a determination means for determining whether or not the target patient is in an unstable state by using the biological information of the target patient and the unstable determination model. Specifically, the determination unit 222 inputs the biological information of the target patient into the unstable determination model to obtain an unstable score. Then, the determination unit 222 determines the unstable state or non-unstable state of the target patient based on the unstable score.

[0050] Here, the unstable determination model is a model generated by the learning device 10 in the first embodiment. That is, the unstable determination model in the present embodiment is a learned model generated in advance using the biological information of patients and the biological information of non-patients. The determination unit 222 acquires the unstable determination model stored in a storage device (not shown) or the like, and determines the unstable state of the target patient.

[0051] The output unit 223 outputs the determination result of the unstable state of the target patient by the determination unit 222. The output unit 223 outputs the determination result to a determination result output device 240 described later. The output unit 223 outputs the determination result in a format that can be output by the determination result output device 240. For example, when the determination result output device 240 includes display means such as a display for outputting the determination result, the output unit 223 functions as a display control unit for controlling the display means. In this way, the output unit 223 functions as a means for controlling the determination result output device 240 according to the format of the determination result output in the determination result output device 240.

[0052] The biological information acquisition device 230 is a device for acquiring the biological information of a patient. The biological information acquisition device 230 is, for example, a wearable device or the like. The biological information acquisition device 230 is a device including at least one sensor that acquires the biological information of the patient by being worn on the patient. The biological information and the sensor are as described above. Further, the biological information acquisition device 230 may be, for example, an imaging device installed in the patient's hospital room, or a device that acquires one or both of the patient's voice and sound information in the patient's surrounding environment. In this case, the biological information acquisition device 230 performs a process of extracting the biological information of the patient based on the acquired image information and sound information.

[0053] The determination result output device 240 outputs the determination result of the unstable state of the target patient acquired from the determination device 220. The determination result output device 240 is, for example, an information terminal such as a computer provided in a medical institution. The determination result output device 240 may be an information terminal such as a tablet terminal or a smartphone held by a medical staff member. The determination result output device 240 includes at least one of, for example, a display means capable of displaying characters and images such as a display, and a sound output means capable of outputting sounds such as a speaker. The determination result output device 240 presents the determination result of the unstable state of the target patient to the medical staff using at least one of the display means, the sound output means, etc.

[0054] In addition to the determination result of the unstable state of the target patient, the determination result output device 240 may also output the biological information of the target patient acquired by the biological information acquisition device 230. In this case, the biological information acquisition device 230 and the determination result output device 240 are connected so as to be able to communicate with each other via a communication network such as wireless or wired as described above.

[0055] Subsequently, the operations performed by the instability determination system 200 will be described with reference to FIG. 4. FIG. 4 is a flowchart showing an example of the operations performed by the instability determination system 200.

[0056] The biological information acquisition device 230 acquires the biological information of the target patient (step S201). Then, the biological information acquisition device 230 transmits the acquired biological information of the target patient to the determination device 220 (step S202).

[0057] The target patient information acquisition unit 221 of the determination device 220 receives the biological information of the target patient from the biological information acquisition device 230 (step S203). The determination unit 222 determines the unstable state of the target patient using the biological information of the target patient and the instability determination model (step S204). The output unit 223 transmits the determination result of the unstable state of the target patient by the determination unit 222 to the determination result output device 240 (step S205).

[0058] The determination result output device 240 receives the determination result of the unstable state of the target patient from the determination device 220 (step S206). Then, the determination result output device 240 outputs the determination result of the unstable state of the target patient to medical staff or the like using at least one of a display means, a sound output means, etc. (step S207).

[0059] In the unstable determination system 200 in the present embodiment, in the determination device 220, the unstable state of the target patient is determined using the biological information of the target patient and the unstable determination model. The unstable determination model is a model generated by the learning device 10 in the first embodiment. By using the unstable determination model, the determination device 220 can accurately determine the unstable state of the target patient. The determination result in which the unstable state of the target patient is accurately determined is output to the determination result output device 240, so that medical staff or the like can efficiently grasp the unstable state of the patient. Thus, the unstable determination system 200 contributes to improving the work efficiency of medical staff or the like.

[0060] The unstable determination system 200 may include the learning device 10 in the first embodiment. That is, the unstable determination system 200 may be a system including a learning device. In this case, the determination device 220 determines the unstable state of the target patient using the unstable determination model generated by the learning device 10. The unstable determination system 200 may have a function of re-learning. By further using the biological information of the target patient acquired by the target patient information acquisition unit 221 and the biological information of non-patients acquired from a storage device (not shown) or the like by the learning device 10, the unstable determination system 200 generates a re-learned unstable determination model. The unstable determination system 200 may perform re-learning when the determination result of the unstable state of the target patient output by the determination device 220 does not reach a predetermined accuracy. Note that the determination device 220 in the unstable determination system 200 may include the configuration provided in the learning device 10.

[0061] <Hardware configuration for realizing each component of the embodiment>

[0062] In each embodiment of the present invention, each component of each device and system represents a functional unit block. Some or all of the components of each device and system are realized by any combination of an information processing device 300 and a program as shown in FIG. 5, for example. The information processing device 300 includes the following configuration as an example. ·CPU (Central Processing Unit) 301 ·ROM (Read Only Memory) 302 ·RAM (Random Access Memory) 303 ·Program 304 loaded into RAM 303 ·Storage device 305 that stores program 304 ·Drive device 307 that reads and writes to recording medium 306 ·Communication interface 308 connected to communication network 309 ·Input / output interface 310 for inputting and outputting data ·Bus 311 connecting each component

[0063] Each component of each device in each embodiment is realized by the CPU 301 acquiring and executing a program 304 that realizes these functions. The program 304 that realizes the functions of each component of each device is stored in the storage device 305 or the RAM 303 in advance, for example, and is read by the CPU 301 as needed. Note that the program 304 may be supplied to the CPU 301 via the communication network 309, or may be stored in the recording medium 306 in advance, and the drive device 307 may read the program and supply it to the CPU 301.

[0064] There are various modifications to the method of realizing each device. For example, each device may be realized by any combination of a separate information processing device 300 and a program for each component. Also, a plurality of components included in each device may be realized by any combination of one information processing device 300 and a program.

[0065] Also, part or all of each component of each device is realized by a general-purpose or dedicated circuit including a processor or the like, or a combination thereof. These may be constituted by a single chip, or may be constituted by a plurality of chips connected via a bus. Part or all of each component of each device may be realized by a combination of the above-described circuit or the like and a program.

[0066] When part or all of each component of each device is realized by a plurality of information processing devices, circuits, etc., the plurality of information processing devices, circuits, etc. may be centrally arranged or may be distributed. For example, the information processing devices, circuits, etc. may be realized in a form in which each is connected via a communication network, such as a client and server system, a cloud computing system, etc.

[0067] In the above description, an example of generating a model for determining a patient's unstable state has been shown. However, the present invention is applicable not only to a model for determining an unstable state, but also to all scenarios for generating a model for determining the state of a subject such as a patient.

[0068] The present invention has been described above with reference to the embodiments, but the present invention is not limited to the above embodiments. Those skilled in the art can make various changes within the scope of the present invention in terms of the configuration and details of the present invention. Also, the configurations of the above-described embodiments may be combined or some of the constituent parts may be replaced.

[0069] Part or all of the above embodiments may be described as follows in the following supplementary notes, but are not limited thereto. (Supplementary Note 1) An acquisition means for acquiring biometric information of a patient who is a person who may be in an unstable state and biometric information of a non-patient; Model generation means for generating an instability determination model for determining whether the target patient is in an unstable state or a non-unstable state based on the biometric information of the target patient using the biometric information of the patient and the biometric information of the non-patient; A learning device comprising: (Supplementary Note 2) The model generation means is the learning device according to appended note 1, which generates the instability determination model using the biological information of the patient and the biological information of the non-patient as learning data. (Appended note 3) The learning device according to appended note 2, wherein the learning data includes the biological information of the non-patient in a resting state. (Appended note 4) The learning device according to appended note 2 or 3, wherein the learning data corresponding to the non-unstable state includes the biological information of the non-patient. (Appended note 5) The learning device according to appended note 3 or 4, wherein the resting state of the non-patient includes a sleeping state. (Appended note 6) The learning device according to any one of appended notes 2 to 5, wherein the learning data corresponding to the unstable state includes the biological information of the patient in the unstable state. (Appended note 7) The learning device according to any one of appended notes 2 to 6, wherein the learning data corresponding to the non-unstable state includes the biological information of the patient in the non-unstable state. (Appended note 8) The learning device according to any one of appended notes 1 to 7, wherein the non-patient is a person whose probability of becoming in the unstable state is equal to or less than a predetermined probability. (Appended note 9) The learning device according to any one of appended notes 1 to 8, wherein the non-patient is at least one of a person who can perform daily life activities by himself / herself, a person without underlying diseases, and a person who does not require assistance or care from others. (Appended note 10) The learning device according to any one of appended notes 1 to 9, wherein the biological information of the non-patient includes the biological information of non-patients of different ages. (Appended note 11) It includes determination means for determining whether the target patient is in the unstable state using the biological information of the target patient and the instability determination model. The determination device, wherein the instability determination model is a learned model generated by the learning device according to any one of appended notes 1 to 10. (Appended note 12) A computer Obtain biological information of a patient who is a person likely to be in an unstable state and biological information of a non-patient, A learned model generation method for generating an instability determination model that uses the biological information of the patient and the biological information of the non-patient to determine whether the target patient is in an unstable state based on the biological information of the target patient. (Appendix 13) Obtain biological information of a patient who is a person likely to be in an unstable state and biological information of a non-patient, A recording medium storing a program for causing a computer to execute a process of generating an instability determination model that uses the biological information of the patient and the biological information of the non-patient to determine whether the target patient is in an unstable state based on the biological information of the target patient.

Explanation of Signs

[0070] 10 Learning device 11 Acquisition unit 12 Model generation unit 200 Instability determination system 220 Determination device 221 Target patient information acquisition unit 222 Determination unit 223 Output unit 230 Biological information acquisition device 240 Determination result output device 300 Information processing device 301 CPU 302 ROM 303 RAM 304 Program 305 Storage device 306 Recording medium 307 Drive device 308 Communication interface 309 Communication network 310 Input / output interface 311 Bus

Claims

1. An acquisition means for acquiring biometric information of a patient who is a person likely to be in an unstable state and biometric information of a non-patient; A model generation means for generating an instability determination model that uses the biometric information of the patient and the biometric information of the non-patient as learning data and determines whether the target patient is in an unstable state or a non-unstable state based on the biometric information of the target patient; A learning device comprising the above.

2. The learning device according to claim 1, wherein the learning data includes biometric information of the non-patient in a resting state.

3. The learning device according to claim 1 or 2, wherein the learning data corresponding to the non-unstable state includes biometric information of the non-patient.

4. The learning device according to claim 2 or 3, wherein the resting state of the non-patient includes a sleeping state.

5. The learning device according to any one of claims 1 to 4, wherein the learning data corresponding to the unstable state includes biometric information of the patient in the unstable state.

6. The learning device according to any one of claims 1 to 5, wherein the learning data corresponding to the non-unstable state includes biometric information of the patient in the non-unstable state.

7. The learning device according to any one of claims 1 to 6, wherein the non-patient is a person with a probability of becoming in an unstable state of a predetermined probability or less.

8. The learning device according to any one of claims 1 to 7, wherein the non-patient is at least one of a person who can perform daily life activities by himself / herself, a person without underlying diseases, and a person who does not require assistance or care from others.

9. The learning device according to any one of claims 1 to 8, wherein the biometric information of the non-patient includes biometric information of non-patients of different ages.

10. Comprising a determination means for determining whether the target patient is in the unstable state by using the biometric information of the target patient and the instability determination model, The determination device, wherein the instability determination model is a learned model generated by the learning device according to any one of claims 1 to 9.

11. A computer, acquires biometric information of a patient who is a person likely to be in an unstable state and biometric information of a non-patient, A learned model generation method for generating an instability determination model that uses the biometric information of the patient and the biometric information of the non-patient as learning data and determines whether the target patient is in an unstable state based on the biometric information of the target patient.

12. Acquires biometric information of a patient who is a person likely to be in an unstable state and biometric information of a non-patient, A program that causes a computer to execute a process of generating an instability determination model that uses the biological information of the patient and the biological information of the non-patient as learning data and determines whether or not the target patient is in an unstable state based on the biological information of the target patient.

Images