JP2019213888A - Watching system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a watching system capable of treating a patient on the basis of a common biological signal in various hospitals, clinics, and various medical spots during home medical treatment, and watching over the patient. SOLUTION: The biometric information of the user is detected, and further the illuminance and noise of the environment are detected. Then, based on the illuminance and noise, determine the accuracy of the biometric information, when the biometric information is accurate, perform an abnormality determination using the biometric information, when the biometric information is not accurate, the The abnormality determination is performed using the biometric information obtained by modifying the biometric information based on the illuminance and noise. At this time, when the desired number of the biometric information of the first user is accumulated in the storage unit, the second reference obtained by adjusting the first reference is calculated by using machine learning, and the second reference of the first user is calculated. The abnormality of the first user is determined based on the biometric information or the corrected biometric information and the second criterion. [Selection diagram] Figure 2

Description

  The present invention relates to a monitoring system that can share a patient's biological signal in hospitals, clinics, nursing homes, home care, and the like.

  When a patient is in a hospital, biological signals such as pulse and respiration are measured on a hospital bed using a hospital biological signal detection device. On the other hand, when a patient leaves the hospital and obtains a similar biological signal at a clinic near his home, a biological signal detection device provided in the clinic is used. In addition, when medical treatment is performed at home, a biological signal is detected using a biological signal detection device equipped at home. However, the detected values of these biosignals are limited to use in hospitals, clinics and homes. In that case, the characteristics or the like unique to each detection device may affect the accuracy of the detection result, and may cause a variation in the diagnosis of a doctor or a nurse.

  In addition, the respiratory pattern for notification of the patient is registered in advance, and when the respiratory pattern detected by the respiratory pattern detecting means matches the respiratory pattern for report, this is communicated to a remote medical management server via a communication line. A medical device and a home medical system for notifying via a telephone have been proposed (Patent Document 1).

Patent No. 4607365

  However, the watching conventional technology described in Patent Document 1 requests an urgent dispatch to an emergency hospital or the like via a server when abnormal breathing occurs during home medical treatment. , Hospitals and neighboring clinics, etc., do not watch patients.

  The present invention has been made in view of such a problem, and in a hospital, a clinic, and various medical sites during home medical treatment, a patient is treated based on a common biosignal and monitored by medical treatment. It is an object of the present invention to provide a patient monitoring system that can perform the monitoring.

The watching system according to the present invention includes:
Detecting the biological information of the user that is continuous data, a detecting unit that detects the illuminance and noise of the environment for detecting the biological information,
A first criterion used to determine an abnormality in biological information, and a storage unit capable of holding biological information for each user,
Based on the illuminance and noise, determine the accuracy of the biological information, when the biological information is accurate, perform an abnormality determination using the biological information, when the biological information is not accurate, An abnormality determination is performed using biological information obtained by correcting information based on the illuminance and the noise, and when a desired number of biological information of the first user is stored in the storage unit, the second determination is performed using machine learning. Calculating a second criterion obtained by adjusting the first criterion; determining an abnormality of the first user based on the first user's biometric information or corrected biometric information and the second criterion; When the desired number of pieces of biological information are not stored in the storage unit, the abnormality of the second user is determined based on the biological information of the second user or the corrected biological information and the first reference. A control unit to
Is provided.

In this watching system,
For example,
The detection unit includes a sheet-shaped bag laid under the user and a piezoelectric sensor installed in the bag, and the piezoelectric sensor detects a change in air pressure in the bag.

In this monitoring system,
For example,
The detection unit includes a load sensor provided on a bed used by the first user or the second user, and detects whether the first user or the second user is in bed or leaves the bed, Calculating the movement of the center of gravity of the first user or the second user on the bed;
The control unit determines whether or not the range of movement of the center of gravity of the first user is abnormal based on the second criterion, and determines whether the range of movement of the first user is abnormal. It is determined whether the range of movement is abnormal for the second user based on the first criterion.

Further, in the present invention,
The detection unit includes a first sensor and a second sensor,
The storage unit holds biological information of the first user acquired by the first sensor,
The control unit calculates the second criterion based on the biological information of the first user acquired by the first sensor,
The control unit can determine an abnormality of the first user based on the biological information of the first user acquired by the second sensor and the second criterion.

  Furthermore, the control unit can notify the outside when it determines that there is an abnormality.

  According to the present invention, a biological signal provided with an ID (identification symbol) linking a patient and a detection device is stored in a server, and the data of the server is analyzed in real time by machine learning and continuously updated. Thus, data suitable for the patient can be stored. Thus, the mutation point suitable for each patient can be stored in the server, so that when the control unit determines that there is an abnormality, the execution unit can issue a suitable alarm signal or the like.

  The biological signal stored in the server is, for example, a respiratory signal, but is not limited thereto. Since various biological signals can be stored in the server, all biological signals necessary for treatment or medical treatment for one patient can be stored. Is stored, and the biological signal can be grasped in association with the detection device. In addition, not only doctors at the hospital (first medical institution), but also doctors at a family clinic near the home (second medical institution) can access the server to obtain the biological information acquired at the first medical institution. The signal, together with its detection device, can be ascertained. Therefore, all medical institutions related to one patient can appropriately make a diagnosis for the patient based on the shared common data, and the diagnostic accuracy and efficiency are greatly improved. In addition, since the same device can be used to diagnose a patient's condition at a hospital, a doctor's office, and at home based on continuous data, it is possible to accurately detect a mutation point in a patient's biological signal.

It is a block diagram showing a watching system concerning an embodiment of the present invention. FIG.

  Hereinafter, embodiments of the present invention will be specifically described with reference to the accompanying drawings. As shown in FIG. 1, at a specific medical institution (first medical institution), various biological signals are acquired by a detection device 1 for a patient, and bibliographic information such as a patient name is input from an input device 2. As the information is input, so-called chart information such as a diagnosis result is also input. Then, the diagnostic information (the chart information) of these patients and the information of the detecting device 1 are input to the ID assigning device 3, and the patient diagnostic information and the detecting device information are linked, and one ID is assigned. Is done. The patient information to which the ID is assigned and also includes the information of the detecting device is input from the ID assigning device 3 to the server 4 and stored. The storage device of the server 4 stores the normal respiratory rate and the like of the patient from the biological signal and stores the data. The data relating to the type of the detecting device when the respiratory rate and the like are detected is also stored in the same ID. Stored in association. The biological signal detected by the detection device 1 includes, in addition to data indicating a physiological state of a patient's human body such as respiratory rate, heart rate, body temperature, blood pressure, SPO2 (arterial oxygen saturation), and body weight, on a bed There are data indicating the behavioral state of the human body such as body movement and falling from the bed (leaving from the bed), and are detection signals for various patient states. All are data that contribute to patient diagnosis. The information detected by the detection device 1 may include environmental signals such as temperature, humidity, illuminance, and noise in addition to the biological signal. This environmental signal, at the time of data analysis whether or not the data of the measured biological signal is accurate, by processing the environment signal diversified, to determine the accuracy of the measured biological signal, Used to correct.

  The control unit 5 reads the data stored in the server 4 and determines whether or not the respiratory rate and the like grasped from the current biological signal are within the normal range of the patient. When the biological signal is out of the normal range of the patient, the control unit 5 outputs an abnormal signal to the execution unit 6, and the execution unit 6 executes a process based on the abnormal signal.

  As shown in FIG. 2, the detecting device 1, the input device 2, and the ID assigning device 3 are provided for each medical facility or each medical facility including a home. That is, the medical equipment 10 including the detecting device 1, the input device 2, and the ID assigning device 3 is provided, for example, in a hospital or the like, and the test data (biological signal) and the diagnosis result (medical chart) acquired by the medical institution are provided. Information) is stored on the server. In a nearby clinic, which is another medical institution (second medical institution), a medical facility 10a including a detecting device 1a, an input device 2a, and an ID assigning device 3a is provided. Similarly, a medical facility such as a home is provided with a medical facility 10b including a detecting device 1b, an input device 2b, and an ID assigning device 3b. Each of the medical facilities 10, 10a, and 10b is provided with an output device 7, 7a, and 7b, respectively. The patient information stored in the server 4 is output via the output devices 7, 7a, and 7b. Is acquired. The diagnosis result (medical record information) may be made known only to each medical institution, and other medical institutions may not be able to usually know the diagnostic result (medical record information).

  As the detection device 1, for example, there is a sensor for a respiration rate. As this respiration rate sensor, there is a sensor in which air is sealed in a sheet-like bag, and a change in pressure of the air is detected by a piezoelectric sensor. This sheet-like sensor is laid under the patient on the bed (between the patient and the bed), and the state in which the air pressure changes according to the patient's breathing is detected by the piezoelectric sensor to detect the patient's respiratory rate. be able to. The normal range of the respiration rate is set in the server 4. In addition, a load sensor is provided on a column or the like that supports the frame of the bed, and it is possible to detect whether the patient is on or off the bed based on the load detection value of the load sensor. . Further, by detecting the load applied to the strut on the four struts and calculating the position of the center of gravity in the bed, the patient on the bed is in the sitting position at the end of the bed, and whether the patient is about to leave the bed. Can be predicted. Further, the movement of the patient on the bed can be detected by the movement of the position of the center of gravity. These normal ranges of the amount of movement of the center of gravity, for example, are set in the server 4. In addition, the behavior of the patient on the bed can be predicted based on these detection results.

  Next, the operation of the watching system configured as described above will be described. For example, a normal range of the respiratory rate is stored in the server 4 as a biological signal. The normal range of the respiration rate stored in the server 4 is, for example, 12 to 20 times / minute as a default reference. Then, at this hospital, every time the respiratory rate of the patient specified by this ID is measured, this data is input to the server 4, and the true normal range of the respiratory rate of this patient is machine-learned. As a default criterion, a normal range of the respiratory rate considered to be standard for all patients and the detection device 1 (respiratory sensor) is set in the server 4, and the measured values of the patient's biosignals usually fall out of this normal range. If it is deviated, the control unit 5 outputs an abnormal signal to the execution unit 6 and instructs the execution unit 6 to issue an alarm or the like. However, this normal range is considered to be a unique and unique range for the patient, and the measured respiratory rate may be abnormal for one patient but not abnormal for another patient. The control unit 5 performs the machine learning of the normal range of the respiratory rate unique and unique to the patient every time a biological signal is input to the server 4. For example, if the detected value of the respiratory rate of this patient is often 10 to 20 times / minute, the range of 10 to 20 times / minute is regarded as a normal range, and the normal range of the respiratory rate of this patient is corrected. I do. That is, each time the respiratory rate biological signal is input to the server 4, the control unit 5 corrects the variation point of the respiratory rate biological signal, that is, the upper limit value and the lower limit value of the normal range. By such machine learning, the control unit 5 corrects the mutation point of the patient specified by the ID, and sets a mutation point unique to the patient in the server 4. In addition, this ID is coupled not only to the patient but also to the detecting device that has acquired the biological signal, so that the appearance of the detected value of the respiratory rate based on the personality of the detecting device can be taken into consideration. . If the detection device is of another type, the patient's respiratory rate itself may not change, but slight variations may occur. For this reason, the ID is associated not only with information about a specific patient, but also with the type of detection device that acquires the biological signal.

  Similarly, when the patient leaves the hospital and is undergoing medical treatment at home or undergoing treatment or examination at his / her local clinic, the detected biosignal and chart information are also transmitted to the server 4. Stored. In this way, based on the ID, the patient's biological signal, the detection device, and the medical record information are stored in the server 4, and any physician of any medical institution can access the server 4 and output it. The information can be extracted by the devices 7, 7a, 7b. Therefore, the doctor can also grasp the biological signal at the time of examination at another medical institution, which contributes to the realization of appropriate treatment.

  In this way, for example, in the case of the respiratory rate, the unique normal range of the patient corrected by the machine learning is set in the server 4. Therefore, when the patient is discharged and is undergoing medical treatment at home, the abnormality occurs in the patient, and the respiratory rate is out of the normal range. Notify outside. The notification of this abnormality may be sent to a hospital or a clinic, in which case, a doctor or the like will visit the patient's home via the hospital or clinic where the abnormality is detected, or The person contacts the patient's home by telephone, grasps the degree of the abnormality, and if necessary, transports the patient to a hospital or clinic by an ambulance or the like. Alternatively, the execution device 6 notifies the patient's family or relative of the abnormality of the patient, and alerts the family or relative. The transmission of this abnormal signal is based on the normal range that is unique to the patient, and is not in the normal range that is suitable for the general majority of people. Can be.

  As described above, the biological signal is not limited to the respiratory rate, but is not limited to the respiratory rate, heart rate, body temperature, blood pressure, SPO2 (arterial blood oxygen saturation), weight, sleep state, movement on the bed, falling from the bed (leaving from the bed). There are various patient condition detection signals. In any case, the respiratory rate, the heart rate, the sleep state, the body movement distance, and the like can be corrected by machine learning in a normal range unique to the patient, and a more accurate abnormality determination can be performed.

  In addition, the server 4 may be installed in a hospital that is the backbone of patient treatment, but is not necessarily a large hospital but has various installation modes such as private facilities. In addition to the hospitals and clinics, there are, of course, nursing homes as facilities where patients need medical care.

1, 1a, 1b: detecting device 2, 2a, 2b: input device 3, 3a, 3b: ID assigning device 4: server 5, control unit 6, execution unit 7, 7a, 7b: output device 10, 10a, 10b: Medical equipment

Claims (5)

Detecting the biological information of the user that is continuous data, a detecting unit that detects the illuminance and noise of the environment for detecting the biological information,
A first criterion used to determine an abnormality in biological information, and a storage unit capable of holding biological information for each user,
Based on the illuminance and noise, determine the accuracy of the biological information, when the biological information is accurate, perform an abnormality determination using the biological information, when the biological information is not accurate, An abnormality determination is performed using biological information obtained by correcting information based on the illuminance and the noise. When a desired number of biological information of the first user is stored in the storage unit, the first information is stored in the storage unit. Calculating a second criterion obtained by adjusting the first criterion; determining an abnormality of the first user based on the first user's biometric information or corrected biometric information and the second criterion; When the desired number of pieces of biological information are not stored in the storage unit, the abnormality of the second user is determined based on the biological information of the second user or the corrected biological information and the first reference. A control unit to
Watching system equipped with. The detection unit includes a sheet-like bag laid under the user, and a piezoelectric sensor installed in the bag, wherein the piezoelectric sensor detects a change in air pressure in the bag. The watching system according to claim 1. The detection unit includes a load sensor provided on a bed used by the first user or the second user, and detects whether the first user or the second user is in bed or leaves the bed, Calculating the movement of the center of gravity of the first user or the second user on the bed;
The control unit determines whether or not the range of movement of the center of gravity of the first user is abnormal based on the second criterion, and determines whether the range of movement of the first user is abnormal. 3. The watching system according to claim 1, wherein a range of movement is determined based on the first criterion whether the second user is abnormal. 4. The detection unit includes a first sensor and a second sensor,
The storage unit holds biological information of the first user acquired by the first sensor,
The control unit calculates the second criterion based on the biological information of the first user acquired by the first sensor,
4. The control unit according to claim 1, wherein the control unit determines an abnormality of the first user based on the biological information of the first user acquired by the second sensor and the second criterion. 5. Watching system. The watching system according to any one of claims 1 to 4, wherein the control unit notifies an external unit when the control unit determines that the abnormality is abnormal.

Images