Vital signs acquisition device, computer program, and vital signs acquisition system
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- NIHON KOHDEN CORP
- Filing Date
- 2022-11-14
- Publication Date
- 2026-06-30
Smart Images
Figure 0007882759000001 
Figure 0007882759000002
Abstract
Description
Technical Field
[0005]
[0001] The present disclosure relates to a device for acquiring vital signs from a subject, and a computer program executable by a processor mounted on the device. The present disclosure also relates to a vital sign acquisition system including the device and a management device for managing the attribute information of the subject.
Background Art
[0002] Various vital signs are acquired to grasp the condition of a subject. Medical staff regularly visit the subject and acquire predetermined vital signs through appropriate devices. This operation is called spot check. Patent Document 1 discloses a device for assisting the spot check operation of acquiring the respiratory rate, which is an example of vital signs.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] There is a demand to reduce the burden on medical staff related to the spot check operation.
Means for Solving the Problems
[0005] One example aspect that can be provided by the present disclosure is a vital sign acquisition device, an interface that receives a first detection signal corresponding to the vital sign from a first sensor that acquires the vital sign of a subject, an inference model that outputs the probability that the first detection signal is classified into each of a plurality of classes, A processor that outputs data corresponding to a vital sign associated with the first detection signal whose probability of being classified into one of the aforementioned multiple classes is above a threshold, It is equipped with.
[0006] One example of an embodiment that may be provided by this disclosure is a computer program that can be executed by a processor installed in a vital signs acquisition device, By being executed, the vital signs acquisition device will The sensor that acquires the subject's vital signs receives a detection signal corresponding to that vital sign. The detection signal is input to an inference model that outputs the probability that the detection signal is classified into each of several classes. The system outputs data corresponding to vital signs associated with the detection signal whose probability of being classified into one of the multiple classes output from the inference model is above a threshold.
[0007] One example of an embodiment that may be provided by this disclosure is a vital signs acquisition system, A first sensor that outputs a first detection signal corresponding to the subject's vital signs, A management device for managing the attribute information of the aforementioned subject, An inference model that outputs the probability that the first detection signal is classified into each of several classes, A processing device that outputs data corresponding to vital signs associated with the first detection signal whose probability of being classified into one of the aforementioned multiple classes is above a threshold to the management device, It is equipped with, The management device stores the data in association with the attribute information.
[0008] Spot checks typically involve healthcare professionals visiting a patient's bedside several times a day to obtain specific vital signs. These vital signs are recorded manually or via a device called a spot checker into an electronic medical record or other management database. While this work requires regular commitment, the patient is not always in a condition to provide good vital signs during visits, which contributes to the increased burden on healthcare professionals.
[0009] According to the configurations described in each of the above examples, data corresponding to the subject's vital signs acquired through the first sensor can be input into the vital sign acquisition device in real time. In addition, data corresponding to vital signs that have a high probability of belonging to one of several pre-set classes can be selectively acquired and stored in association with the subject's attribute information. In other words, the acquisition of vital signs to be used for recording in management databases such as electronic medical records can be automated without medical professionals having to visit the subject's bedside. Furthermore, because vital signs that meet the desired conditions are selected through the intervention of an inference model, data to be used for recording can be acquired in a state where the influence of the subject's condition and the acquisition environment is suppressed. Therefore, the burden on medical professionals involved in spot check work can be reduced. [Brief explanation of the drawing]
[0010] [Figure 1] This illustrates the functional configuration of a vital signs acquisition system according to one embodiment. [Figure 2] This shows the functional configuration of a vital signs acquisition system in a different example. [Modes for carrying out the invention]
[0011] An example of an embodiment will be described in detail with reference to the attached drawings.
[0012] Figure 1 illustrates the functional configuration of a vital signs acquisition system 10 according to one embodiment. The vital signs acquisition system 10 is a system for acquiring and managing the vital signs of a subject S.
[0013] As used herein, the term "vital signs" means at least one numerical value of pulse rate, respiration, body temperature, blood pressure, and level of consciousness, or the change in such value over time. The change in values over time may be visualized as waveforms or graphs.
[0014] The vital sign acquisition system 10 includes a first sensor 111. The first sensor 111 is configured to output a first detection signal DS1 corresponding to the vital signs of the subject S. The first detection signal DS1 may be an analog signal or a digital signal, depending on the specifications of the first sensor 111.
[0015] The first sensor 111 may have a well-known configuration corresponding to the vital signs to be acquired. For example, the first sensor 111 may be a non-restraint sensor installed on a patient's bed. Alternatively, the first sensor 111 may be mounted on a wearable device attached to the body of a subject S, or on a mobile device that can be carried by the subject S. Alternatively, the first sensor 111 may be mounted on a monitoring device, medical telemetry device, camera, thermal camera, smart speaker, etc., installed in a medical facility.
[0016] If the first sensor 111 is a non-restraint sensor installed on a patient's bed, vital signs such as respiratory rate and heart rate may be acquired. If the first sensor 111 is mounted on a camera or smart speaker, the level of consciousness may be acquired as a vital sign. If the first sensor 111 is mounted on a thermal camera, body temperature may be acquired as a vital sign. If the first sensor 111 is mounted on a wearable device or mobile device, heart rate, respiratory rate, blood pressure, and body temperature may be acquired as vital signs.
[0017] The vital sign acquisition system 10 includes a processing device 12. The processing device 12 includes an input interface 121, an inference model 122, a processor 123, and an output interface 124. The processing device 12 is an example of a vital sign acquisition device.
[0018] The input interface 121 is configured as a hardware interface that receives a first detection signal DS1. When the first detection signal DS1 is an analog signal, the input interface 121 includes an appropriate conversion circuit including an A / D converter. This description is similarly applicable to other signals and data that the input interface 121 described later can receive.
[0019] The inference model 122 is an algorithm configured to output, as an inference result, the probability that the first detection signal DS1 is classified into each of a plurality of classes. Examples of the plurality of classes include "with artifact" and "without artifact". The "with artifact" class corresponds to a state in which an artifact that cannot be ignored is superimposed on the first detection signal DS1. The "without artifact" class corresponds to a state in which there are no artifacts or the artifacts superimposed on the first detection signal DS1 are negligible. Another example of the plurality of classes includes whether the subject S is in a sleeping state, whether chain-stokes respiration, obstructive apnea, central apnea, etc. are occurring in the subject S, and the like. The probability takes a value between 0 and 1. The value 0 corresponds to 0%. The value 1 corresponds to 100%. The inference result may be a score (for example, any value from 1 to 5) corresponding to the inferred probability.
[0020] The inference model 122 can be generated by performing machine learning using teacher data. The teacher data can be appropriately configured according to the type of vital sign to be acquired and a plurality of classes to be set. Examples of algorithms used for machine learning include neural networks, decision trees, random forests, support vector machines, and the like.
[0021] The processor 123 is configured to input the first detection signal DS1 received by the input interface 121 to the inference model 122 and to acquire data corresponding to the probability output from the inference model 122. The processor 123 is configured to determine whether the probability of the first detection signal DS1 being classified into a specific class is above a threshold. For example, it is determined whether the probability of the first detection signal DS1 being classified into the "no artifact" class is above a threshold.
[0022] The processor 123 is configured to output vital sign data VD corresponding to the vital sign associated with the first detection signal DS1 whose probability is determined to be above a threshold, from the output interface 124. The vital sign data VD may be in the form of analog data or digital data, depending on the specifications of the device that receives the data.
[0023] The output interface 124 is configured as a hardware interface. When the vital sign data VD is in the form of analog data, the output interface 124 includes appropriate conversion circuits, including a D / A converter. This description also applies to other signals and data that the output interface 124 can output, as described later.
[0024] The vital signs acquisition system 10 includes a management device 13. The management device 13 is configured to manage the attribute information of the subject S. Examples of attribute information include the subject S's name, age, gender, and medical history. In other words, the management device 13 may be part of a management database system such as an electronic medical record.
[0025] The management device 13 may include storage such as semiconductor memory, hard disk drives, or magnetic tape drives. The management device 13 is configured to store the vital sign data VD of the subject S output from the processing device 12 in its storage, in association with the attribute information of the subject S.
[0026] Spot checks typically involve healthcare professionals visiting a patient's bedside several times a day to obtain specific vital signs. These vital signs are recorded manually or via a device called a spot checker into an electronic medical record or other management database. While this work requires regular commitment, the patient is not always in a condition to provide good vital signs during visits, which contributes to the increased burden on healthcare professionals.
[0027] With the above configuration, data corresponding to the vital signs of subject S acquired through the first sensor 111 can be input to the processing unit 12 in real time. In addition, data corresponding to vital signs that have a high probability of belonging to one of several pre-set classes can be selectively acquired and stored in association with the attribute information of subject S. In other words, the acquisition of vital signs to be used for recording in management databases such as electronic medical records can be automated without medical personnel having to visit subject S's bedside. Furthermore, because vital signs that meet the desired conditions are selected through the intervention of the inference model 122, vital sign data VD to be used for recording can be acquired in a state in which the influence of subject S's condition and the acquisition environment is suppressed. Therefore, the burden on medical personnel involved in spot check work can be reduced.
[0028] In particular, by setting the above-mentioned multiple classes to include artifacts superimposed on vital signs, it becomes possible to selectively acquire vital sign data VD corresponding to vital signs where the superposition of artifacts caused by, for example, the subject S's body movement can be ignored. This automatic selection process facilitates the acquisition of high-quality vital sign data VD, thereby reducing the burden on healthcare professionals involved in spot checks.
[0029] Alternatively, the above-mentioned multiple classes can be set up so that the elapsed time since the vital signs were acquired is relevant. For example, vital sign data VD corresponding to vital signs whose elapsed time since acquisition is less than a predetermined value (relatively recent) can be selectively acquired. By performing such selection automatically, it becomes easier to acquire vital sign data VD at the desired timing regardless of the timeliness of input from the first sensor 111, thereby reducing the burden on healthcare workers involved in spot check operations.
[0030] From the standpoint of timely input of vital signs, it is preferable that the operating period of the first sensor 111 is longer than the non-operating period. The term "operating period" used here does not necessarily refer only to the state in which the sensor is acquiring vital signs. For example, if the sensor is configured to return from a standby state and begin acquiring vital signs under predetermined conditions, the state in which background processing is being performed to detect those predetermined conditions falls under the "operating period" even if the vital sign acquisition operation is not taking place.
[0031] It is known that accurately obtaining respiratory status is relatively difficult when performing spot checks visually. Therefore, it is preferable that the vital signs of the subject S obtained by the first sensor 111 include respiratory status. In this case, the acquisition of respiratory status with improved accuracy can be easily automated. This can enhance the effect of reducing the burden on healthcare professionals involved in spot check work.
[0032] The processor 123 may be configured to acquire an index for evaluating the signs of injury or illness of subject S based on vital signs selected based on probabilities output by the inference model 122. An example of such an index is NEWS (National Early Warning Score). NEWS is acquired based on multiple types of vital signs. In this case, the vital sign data VD output from the output interface 124 is configured to include information corresponding to the index.
[0033] In the case of spot checks conducted through visits as described above, the infrequent nature of these checks meant that the possibility of failing to detect sudden changes in the subject's condition or signs thereof could not be ruled out. By including the above-mentioned indicators in the vital sign data VD, in addition to the real-time input of vital signs through the first sensor 111, it becomes easier to promptly detect sudden changes in the subject S's condition or signs thereof. This enhances the effect of reducing the burden on healthcare professionals involved in spot check operations.
[0034] As illustrated in Figure 1, the vital sign acquisition system 10 may include a second sensor 112. The second sensor 112 is configured to output a second detection signal DS2 corresponding to at least one of the subject S's level of consciousness and resting state. The second detection signal DS2 may be an analog signal or a digital signal, depending on the specifications of the second sensor 112. The input interface 121 of the processing unit 12 may also be configured to accept the second detection signal DS2. The processor 123 may be configured to determine whether to output vital sign data VD based on the second detection signal DS2.
[0035] For example, the second sensor 112 may be a camera that captures an image of the subject S. In this case, the second detection signal DS2 may be the signal corresponding to the image. The processor 123 uses appropriate image processing techniques to determine at least one of the subject S's level of consciousness and resting state. For example, if the subject S's level of consciousness is determined to be unsuitable for obtaining vital signs (e.g., sleeping when they should be awake), the processor 123 may be configured to prohibit output from the vital sign data VD output interface 124. In addition to or alternative to this, if the subject S is determined to be not in a resting state, the processor 123 may be configured to prohibit output from the vital sign data VD output interface 124.
[0036] The processor 123 may determine whether oxygen is being administered appropriately to the subject S based on the image acquired by the second sensor 112.
[0037] Alternatively, the second sensor 112 may be a motion sensor installed on the patient S's bedside. In this case, the second detection signal DS2 may be a signal corresponding to the patient S's body movement. The processor 123 determines the patient S's resting state based on this signal. If it is determined that the patient S is not at rest, the processor 123 may be configured to prohibit output from the vital sign data VD output interface 124.
[0038] This configuration allows for the restriction of outputting vital sign data (VD) corresponding to vital signs obtained under inappropriate circumstances. This reduces the occurrence of situations where data based on inappropriate vital signs is recorded in the management database, thereby easing the burden of data management related to spot check operations.
[0039] As illustrated by the dashed line in Figure 1, the second detection signal DS2 output from the second sensor 112 may be input to the management device 13 instead of the processing device 12. In this case, the management device 13 has an interface and processor similar to the input interface 121 and processor 123 of the processing device 12. The processor of the management device 13 may be configured to determine whether to save the vital sign data VD output from the processing device 12 based on the second detection signal DS2 received through the interface.
[0040] For example, if the control device 13 determines, based on the second detection signal DS2, that the subject S's level of consciousness is not suitable for obtaining vital signs (e.g., if the subject is asleep when they should be awake), the control device 13 may be configured to prohibit saving the vital sign data VD received from the processing device 12. In addition to or instead of this, if the control device 13 determines that the subject S is not in a resting state, the control device 13 may be configured to prohibit saving the vital sign data VD received from the processing device 12.
[0041] This configuration also allows for limiting the recording of vital sign data (VD) corresponding to vital signs obtained under inappropriate circumstances into the management database, thereby reducing the burden of data management related to spot check operations.
[0042] The processor 123 of the processing unit 12, which has the various functions described above, can be realized by a general-purpose microprocessor that works in cooperation with general-purpose memory. Examples of general-purpose microprocessors include CPUs, MPUs, and GPUs. Examples of general-purpose memory include ROMs and RAMs. In this case, the ROM may store computer programs that realize the various functions described above. ROM is an example of a non-temporary computer-readable medium that stores computer programs. The general-purpose microprocessor selects at least a portion of the program stored in the ROM and loads it onto the RAM, and then works in cooperation with the RAM to execute the above-described processes. The computer program may be pre-installed in the general-purpose memory, or it may be downloaded from an external server device via a communication network and then installed in the general-purpose memory. In this case, the external server device is an example of a non-temporary computer-readable medium that stores computer programs.
[0043] The processor 123 may be implemented by a dedicated integrated circuit such as a microcontroller, ASIC, or FPGA capable of executing the above-mentioned computer program. In this case, the above-mentioned computer program is pre-installed in a memory element included in the dedicated integrated circuit. This memory element is an example of a computer-readable medium that stores the computer program. The processor 123 can also be implemented by a combination of a general-purpose microprocessor and a dedicated integrated circuit.
[0044] The various configurations described herein are merely examples to facilitate understanding of this disclosure. Each example configuration may be modified or combined with others as appropriate within the scope of the intent of this disclosure.
[0045] The inference model 122 does not necessarily have to be mounted on the processing unit 12. As illustrated in Figure 2, the vital sign acquisition system 10 can have a configuration in which the first sensor 111, the second sensor 112, the processing unit 12, the management device 13, and the server device 14 are connected to each other via a communication network N. In this case, the inference model 122 can be mounted on the management device 13 or the server device 14, as long as data can be exchanged with the processor 123 of the processing unit 12.
[0046] The configurations listed below also constitute part of this disclosure. Item 1: An interface that receives a first detection signal corresponding to the vital signs from a first sensor that acquires the vital signs of the subject, An inference model that outputs the probability that the first detection signal is classified into each of several classes, A processor that outputs data corresponding to a vital sign associated with the first detection signal whose probability of being classified into one of the aforementioned multiple classes is above a threshold, It is equipped with Vital signs acquisition device. Item 2: One of the aforementioned classes relates to artifacts superimposed on the vital signs. The vital signs acquisition device described in item 1. Item 3: One of the aforementioned classes is related to the time elapsed since the vital signs were obtained. A vital signs acquisition device as described in item 1 or 2. Item 4: The aforementioned vital signs include respiratory information. A vital signs acquisition device as described in any one of items 1 to 3. Item 5: The aforementioned data includes indicators for evaluating disease signs in the subject. A vital signs acquisition device as described in any one of items 1 through 4. Item 6: The interface receives a second detection signal from the second sensor that corresponds to at least one of the subject's level of consciousness and resting state. The processor determines whether to output the data based on the second detection signal. A vital signs acquisition device as described in any one of items 1 through 5. Item 7: A computer program that can be executed by a processor installed in a vital signs acquisition device, By being executed, the vital signs acquisition device will The sensor that acquires the subject's vital signs receives a detection signal corresponding to that vital sign. The detection signal is input to an inference model that outputs the probability that the detection signal is classified into each of several classes. The inference model outputs data corresponding to vital signs associated with the detection signal whose probability of being classified into one of the multiple classes is above a threshold. Computer program. Item 8: A first sensor that outputs a first detection signal corresponding to the subject's vital signs, A management device for managing the attribute information of the aforementioned subject, An inference model that outputs the probability that the first detection signal is classified into each of several classes, A processing device that outputs data corresponding to a vital sign associated with the first detection signal whose probability of being classified into one of the aforementioned multiple classes is above a threshold to the management device. It is equipped with, The management device stores the data in association with the attribute information. Vital signs acquisition system. Item 9: The aforementioned first sensor has a longer operating period than its non-operating period. The vital signs acquisition system described in item 8. Item 10: The system includes a second sensor that outputs a second detection signal corresponding to at least one of the subject's level of consciousness and resting state. The processing device determines whether to output the data to the management device based on the second detection signal. A vital signs acquisition system as described in item 8 or 9. Item 11: The system includes a second sensor that outputs a second detection signal corresponding to at least one of the subject's level of consciousness and resting state. The management device determines whether to save the data based on the second detection signal. A vital signs acquisition system as described in item 8 or 9. Item 12: The management device assigns an index to the data to evaluate the disease signs of the subject. A vital signs acquisition system as described in any one of items 8 through 11. [Explanation of Symbols]
[0047] 111: First sensor, 112: Second sensor, 12: Processing unit, 121: Input interface, 122: Inference model, 13: Management device, DS1: First detection signal, DS2: Second detection signal, S: Subject, VD: Vital sign data
Claims
1. An interface that receives a first detection signal corresponding to the vital signs from a first sensor that acquires the vital signs of the subject, An inference model that outputs the probability that the first detection signal corresponds to a condition in which the influence of the subject's state and the environment in which the vital signs are acquired is suppressed, A processor that selectively outputs data corresponding to vital signs associated with the first detection signal whose probability is above a threshold, from among the vital signs acquired by the first sensor. It is equipped with Vital signs acquisition device.
2. The aforementioned probability corresponds to the state in which there are no artifacts superimposed on the vital signs, or the artifacts are negligible. A vital signs acquisition device according to claim 1.
3. The aforementioned probability corresponds to the state in which the elapsed time since the acquisition of the vital signs is less than a predetermined value. A vital signs acquisition device according to claim 1.
4. The aforementioned vital signs include respiratory information. A vital signs acquisition device according to claim 1.
5. The aforementioned data includes indicators for evaluating disease signs in the subject. A vital signs acquisition device according to claim 1.
6. The interface receives a second detection signal from the second sensor that corresponds to at least one of the subject's level of consciousness and resting state. The vital sign acquisition device according to claim 1, wherein the processor determines whether to output the data based on the second detection signal.
7. A computer program that can be executed by a processor installed in a vital signs acquisition device, By being executed, the vital signs acquisition device will The sensor that acquires the subject's vital signs receives a detection signal corresponding to that vital sign. The detection signal is input to an inference model that outputs the probability that the detection signal corresponds to a condition in which the influence of the subject's state and the environment in which the vital signs are acquired is suppressed. From among the vital signs acquired by the sensor, the data corresponding to the vital signs associated with the detection signal whose probability output from the inference model is above a threshold is selectively output. Computer program.
8. A first sensor that outputs a first detection signal corresponding to the subject's vital signs, A management device for managing the attribute information of the aforementioned subject, An inference model that outputs the probability that the first detection signal corresponds to a condition in which the influence of the subject's state and the environment in which the vital signs are acquired is suppressed, A processing device that selectively outputs to the management device data corresponding to the vital signs associated with the first detection signal whose probability is above a threshold, from among the vital signs acquired by the first sensor. It is equipped with, The management device stores the data in association with the attribute information. Vital signs acquisition system.
9. The aforementioned first sensor has a longer operating period than its non-operating period. A vital signs acquisition system according to claim 8.
10. The system includes a second sensor that outputs a second detection signal corresponding to at least one of the subject's level of consciousness and resting state. The processing device determines whether to output the data to the management device based on the second detection signal. A vital signs acquisition system according to claim 8.
11. The system includes a second sensor that outputs a second detection signal corresponding to at least one of the subject's level of consciousness and resting state. The management device determines whether to save the data based on the second detection signal. A vital signs acquisition system according to claim 8.
12. The vital sign acquisition system according to claim 8, wherein the management device assigns an index to the data for evaluating disease signs of the subject.