Vital sign measurement program, vital sign measurement method, and information processing device.

Abstract

To improve measurement accuracy. [Solution] The information processing device 10 generates point cloud information 4 indicating multiple positions on the body surface of a person 1 around the millimeter-wave sensor 2, based on a first signal 3 acquired from the millimeter-wave sensor 2. The information processing device 10 estimates the attributes of the person 1 based on the point cloud information 4. The information processing device 10 determines the values ​​of signal processing parameters based on the estimated attributes. The information processing device 10 generates a second signal 5 indicating the time change of the position on the body surface of the person 1 by performing signal processing on the first signal 3 according to the determined parameter values. The information processing device 10 then determines predetermined vital sign values ​​based on the second signal 5.

Description

Technical Field

[0001] The present invention relates to a vital measurement program, a vital measurement method, and an information processing apparatus.

Background Art

[0002] There is non-contact sensing technology using sensors such as millimeter wave sensors (sometimes called millimeter wave radars). Millimeter waves are radio wave signals with wavelengths ranging from 1 mm to 1 cm. In a millimeter wave sensor, the reflected wave of the transmitted millimeter wave is received, and information such as point cloud data indicating the position of the reflected object can be obtained. In non-contact sensing using a millimeter wave sensor, unlike a camera or the like, an image is not recorded. Therefore, if a state such as a vital sign of a person is detected by non-contact sensing using a millimeter wave sensor, it is possible to detect the state of the person while considering privacy. Such non-contact sensing using a millimeter wave sensor is useful as a technology for detecting the state of a person in a place where it is difficult to monitor with a camera, such as a toilet.

[0003] Note that a vital sign is an index indicating the life activities of the body, such as a pulse and a respiration rate. A vital sign is sometimes simply called a vital. As a non-contact sensing technology for the state of a person, for example, a highly accurate ecological information monitoring device that can be measured at an arbitrary position while wearing clothes has been proposed. A non-contact type sleep state measurement system that can accurately measure the sleep state regardless of the age or disease of the person to be measured has also been proposed. An electronic device that can detect the heartbeat of a human body or the like with good accuracy by transmitting and receiving radio waves has also been proposed. Furthermore, a millimeter wave mapping system that generates one or more point clouds and determines one or more vital signs for defining the mental state of a human has also been proposed.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 3

[0005] Millimeter-wave sensors can be used to acquire information about human vital signs such as respiration and heart rate. To accurately determine vital signs, it is crucial to properly remove noise from the signals received by the millimeter-wave sensor. Conventionally, a fixed parameter, unrelated to the attributes of the person being measured, is used to remove signals with frequencies outside the frequency band representing vital signs as noise. However, the effective frequency range for vital signs varies depending on the attributes of the person being measured (gender, age, etc.). Therefore, noise reduction that ignores the person's attributes may not be performed correctly, potentially reducing measurement accuracy.

[0006] In one respect, this project aims to improve measurement accuracy. [Means for solving the problem]

[0007] One proposal provides a vital signs measurement program that instructs a computer to perform the following processes: The computer generates point cloud information indicating multiple positions on the body surface of a person around the millimeter-wave sensor, based on a first signal acquired from the millimeter-wave sensor. The computer estimates the person's attributes based on the point cloud information. The computer determines the values ​​of signal processing parameters based on the estimated attributes. The computer generates a second signal indicating the time change of the person's position on the body surface by performing signal processing on the first signal according to the determined parameter values. The computer then determines predetermined vital sign values ​​based on the second signal. [Effects of the Invention]

[0008] According to one embodiment, measurement accuracy can be improved. [Brief explanation of the drawing]

[0009] [Figure 1] This figure shows an example of a vital sign measurement method according to the first embodiment. [Figure 2] This is a diagram showing an example of a monitoring system. [Figure 3] This figure shows an example of terminal device hardware. [Figure 4] This figure shows an example of the hardware for a vital signs measuring device. [Figure 5] This figure shows an example of noise reduction. [Figure 6] This block diagram shows an example of the data and functions of a vital signs measuring device. [Figure 7] This figure shows an example of the process for counting vital sign values. [Figure 8] This figure shows an example of attribute estimation processing. [Figure 9] This figure shows an example of a rule for determining age. [Figure 10] This figure shows an example of age determination using the infant age determination rules. [Figure 11] This figure shows an example of age determination using the rules for determining ages other than infants. [Figure 12] This flowchart shows an example of a procedure for measuring vital signs. [Figure 13] This figure shows an example of vital sign measurement. [Figure 14] This figure shows an example of noise processing according to the vital range. [Modes for carrying out the invention]

[0010] The following description of this embodiment will be made with reference to the drawings. Note that each embodiment can be implemented by combining multiple embodiments within a reasonable scope. 〔First Embodiment〕 The first embodiment is a vital measurement method for accurately measuring vital sign values by non-contact sensing using a millimeter-wave sensor.

[0011] FIG. 1 is a diagram showing an example of the vital measurement method according to the first embodiment. In FIG. 1, an information processing apparatus 10 for implementing the vital measurement method is shown. The information processing apparatus 10 can implement the vital measurement method, for example, by executing a vital measurement program.

[0012] The information processing apparatus 10 includes a storage unit 11 and a processing unit 12. The storage unit 11 is, for example, a memory or a storage device included in the information processing apparatus 10. The processing unit 12 is, for example, a processor included in the information processing apparatus 10. The information processing apparatus 10 may have a plurality of processors. Among the plurality of processes performed by the information processing apparatus 10, a certain process and another process may be executed by different processors, respectively.

[0013] The storage unit 11 stores, for example, parameter information 7. The parameter information 7 has, for example, values of parameters to be applied to signal processing associated with attributes related to a person. The values of the parameters are determined based on physiological knowledge about the person corresponding to the attribute. The physiological knowledge is, for example, the knowledge that infants and schoolchildren have a small tidal volume and a tendency to have a higher respiratory rate because their lungs are in the growth process.

[0014] Based on such physiological knowledge, in the parameter information 7, for example, a frequency band of a signal to be extracted as a heart rate or a respiratory rate of that age group is set in association with an attribute representing the age group. The age groups are, for example, infants (0 to 5 years old), schoolchildren (elementary school students), adults (junior high school students to 65 years old), elderly people (65 years old and above), etc.

[0015] The processing unit 12 determines predetermined vital sign values ​​for a person 1 located around the millimeter-wave sensor 2 based on the first signal 3 output from the millimeter-wave sensor 2. For example, the processing unit 12 generates point cloud information 4 indicating multiple locations on the body surface of the person 1 located around the millimeter-wave sensor 2 based on the first signal 3 acquired from the millimeter-wave sensor 2.

[0016] Next, the processing unit 12 estimates the attributes of person 1 based on the point cloud information 4. For example, the processing unit 12 estimates which age group person 1 belongs to as an attribute. When estimating the age, the processing unit 12 generates a set of points representing person 1 from the points included in the point cloud information 4, for example, by performing a process such as clustering. Then, the processing unit 12 estimates the age based on the generated set of points. For example, if the number of points included in the set of points is less than or equal to a predetermined number, it is estimated that person 1 is an infant. It is also possible to estimate that person 1 is an infant if the size of the region encompassing the set of points is less than or equal to a predetermined value.

[0017] Next, the processing unit 12 determines the values ​​of the signal processing parameters based on the estimated attributes. For example, the processing unit 12 obtains the parameter values ​​associated with the estimated attributes from the parameter information 7 stored in the memory unit 11. Then, the processing unit 12 determines the parameter values ​​to the obtained values. The parameters are, for example, the frequency bands to be extracted in noise processing.

[0018] The processing unit 12 performs signal processing on the first signal 3 according to the determined parameter values ​​to generate a second signal 5 that indicates the time change in the position of person 1 on the body surface. For example, the processing unit 12 generates the second signal 5 by removing frequency components other than those in the frequency band indicated by the determined parameter values.

[0019] The processing unit 12 determines predetermined vital sign values ​​based on the second signal 5. For example, the processing unit 12 counts the frequency of the periodically changing second signal 5. If the second signal 5 represents the heart rate of person 1, the heart rate per minute is determined. If the second signal 5 represents the respiration of person 1, the respiratory rate per minute is determined.

[0020] The processing unit 12 outputs vital data 6, which shows the values ​​of vital signs. For example, the processing unit 12 displays the values ​​of the vital signs shown in the vital data 6 on a monitor. The processing unit 12 may also transmit the vital data 6 to another device. Furthermore, the processing unit 12 may periodically store the generated vital data 6 as history information in the storage unit 11.

[0021] In this way, the information processing device 10 estimates the attributes of person 1 based on the first signal 3 acquired from the millimeter-wave sensor 2, and performs signal processing according to the parameter values ​​corresponding to those attributes. This ensures that appropriate signal processing is performed, and the values ​​of vital signs can be determined with high accuracy. For example, by determining the frequency band to be extracted in noise processing as a parameter value, the heart rate or respiratory rate per minute can be determined with high accuracy.

[0022] Furthermore, by determining attributes based on the point cloud information 4, the processing unit 12 can estimate the height, build, and other characteristics of person 1 from the height or width of the set of points representing person 1. This makes it easier to estimate the age.

[0023] The processing unit 12 may also estimate the gender of person 1 based on the point cloud information 4. In this case, the processing unit 12 estimates the age of person 1 based on the estimated gender and the point cloud information. For example, there are significant differences in physique between men and women once they reach adulthood. By estimating the gender of person 1 in advance and taking gender into account when estimating the age, the accuracy of the age estimation can be improved.

[0024] In estimating age, the processing unit 12 determines the values ​​of predetermined indicators (such as height and movement speed) that represent the physical characteristics or movements of person 1, for example, based on point cloud information 4. Then, the processing unit 12 estimates the age of person 1 based on the values ​​of the predetermined indicators for each gender and age group. By using statistical information in this way to estimate age, the accuracy of age estimation is improved.

[0025] [Second Embodiment] The second embodiment is a health monitoring system that uses a vital signs measuring device capable of accurately measuring the vital signs of a person being measured.

[0026] Figure 2 shows an example of a monitoring system. The vital sign monitor 200 is connected to the network 20, for example, wirelessly. The vital sign monitor 200 measures the vital signs (e.g., heart rate, respiratory rate) of the subjects 31-33 using a millimeter-wave sensor. The vital sign monitor 200 transmits data showing the measured values ​​of vital signs, for example, periodically.

[0027] A terminal device 100 is further connected to the network 20. The terminal device 100 is a computer used by a user to monitor the health status of, for example, persons 31-33. The terminal device 100 receives data showing vital sign measurements from the vital sign measuring device 200. The terminal device 100 stores the received data and, for example, detects abnormalities in vital signs based on the received data. If the terminal device 100 detects a person with abnormal vital signs, it displays an alarm message or outputs an alarm sound.

[0028] Figure 3 shows an example of terminal device hardware. The terminal device 100 is controlled as a whole by a processor 101. The processor 101 is connected to memory 102 and several peripheral devices via a bus 109.

[0029] The terminal device 100 may be a multiprocessor system having multiple processors. A collection of multiple processors in a multiprocessor system can be called a processor 101. A processor 101 may also be called a processor circuitry. Each of the multiple processors can execute some or all of the processes that are executed in the terminal device 100. When there are multiple related processes, two or more of those processes may be executed by different processors.

[0030] The processor 101 is, for example, a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or a DSP (Digital Signal Processor). At least some of the functions that the processor 101 implements by executing a program may be implemented by electronic circuits such as an ASIC (Application Specific Integrated Circuit) or a PLD (Programmable Logic Device).

[0031] Memory 102 is used as the main memory of the terminal device 100. Memory 102 temporarily stores at least a portion of the OS (Operating System) program and application programs to be executed by the processor 101. Memory 102 also stores various data used for processing by the processor 101. For memory 102, a volatile semiconductor memory device such as RAM (Random Access Memory) is used.

[0032] Peripheral devices connected to bus 109 include a storage device 103, a graphics controller 104, an input interface 105, an optical drive device 106, a device connection interface 107, and a network interface 108.

[0033] The storage device 103 electrically or magnetically writes and reads data from its built-in recording medium. The storage device 103 is used as an auxiliary storage device for the terminal device 100. The storage device 103 stores the OS program, application programs, and various data. For example, the storage device 103 can be an HDD (Hard Disk Drive) or an SSD (Solid State Drive).

[0034] The graphics controller 104 is an arithmetic unit that performs image processing. The graphics controller 104 is, for example, a GPU (Graphics Processing Unit). A monitor 21 is connected to the graphics controller 104. The graphics controller 104 displays images on the screen of the monitor 21 according to instructions from the processor 101. The monitor 21 can be an OLED (Electroluminescence) display device or a liquid crystal display device. If a GPU is used as the graphics controller 104, the graphics controller 104 can also perform complex numerical calculations such as matrix calculations.

[0035] The input interface 105 is connected to a keyboard 22 and a mouse 23. The input interface 105 transmits signals from the keyboard 22 and mouse 23 to the processor 101. Note that the mouse 23 is just one example of a pointing device; other pointing devices can also be used. Other pointing devices include touch panels, tablets, touchpads, and trackballs.

[0036] The optical drive device 106 uses laser light or the like to read data recorded on the optical disc 24 or write data to the optical disc 24. The optical disc 24 is a portable recording medium on which data is recorded in a way that makes it readable by the reflection of light. Examples of optical discs 24 include DVD (Digital Versatile Disc), DVD-RAM, CD-ROM (Compact Disc Read Only Memory), and CD-R (Recordable) / RW (ReWritable).

[0037] The device connection interface 107 is a communication interface for connecting peripheral devices to the terminal device 100. For example, a memory device 25 and a memory reader / writer 26 can be connected to the device connection interface 107. The memory device 25 is a recording medium equipped with a communication function with the device connection interface 107. The memory reader / writer 26 is a device that writes data to or reads data from the memory card 27. The memory card 27 is a card-type recording medium.

[0038] The network interface 108 is connected to the network 20. The network interface 108 transmits and receives data to and from other computers or communication devices via the network 20. The network interface 108 is a wired communication interface, for example, connected by cable to a wired communication device such as a switch or router. Alternatively, the network interface 108 may be a wireless communication interface, connected by radio waves to a wireless communication device such as a base station or access point.

[0039] The terminal device 100 implements the processing functions of the second embodiment by executing a program recorded on, for example, a computer-readable recording medium. The program describing the processing content to be executed by the terminal device 100 can be recorded on various recording media. For example, the program to be executed by the terminal device 100 can be stored in the storage device 103. The processor 101 loads at least a portion of the program in the storage device 103 into the memory 102 and executes the program. Alternatively, the program to be executed by the terminal device 100 can be recorded on a portable recording medium such as an optical disc 24, a memory device 25, or a memory card 27. The program stored on the portable recording medium becomes executable after being installed in the storage device 103, for example, under control from the processor 101. The processor 101 can also directly read and execute the program from the portable recording medium.

[0040] Figure 4 shows an example of the hardware of a vital signs measuring device. The vital signs measuring device 200 has a millimeter-wave sensor 210, memory 220, processor 230, and wireless communication interface 240. The millimeter-wave sensor 210 is a non-contact sensor that can measure the distance to objects around the vital signs measuring device 200 using millimeter waves.

[0041] The millimeter-wave sensor 210 includes a synthesizer 211, multiple transmitting antennas 212a, 212b, ..., multiple receiving antennas 213a, 213b, ..., a mixer 214, and an analog-to-digital converter (ADC) 215.

[0042] The synthesizer 211 generates a modulated wave (transmit wave). For example, the synthesizer 211 modulates using the FMCW (Frequency Continuous Modulation) method. The synthesizer 211 transmits the generated transmit wave to the transmitting antennas 212a, 212b, ... and the mixer 214.

[0043] The transmitting antennas 212a, 212b, ... transmit the signal wave towards the vital sign measuring device 200. The receiving antennas 213a, 213b, ... receive the reflected signal wave that is reflected off of the vital sign measuring device 200 by people or other objects in the surrounding area. The receiving antennas 213a, 213b, ... transmit the received reflected signal wave to the mixer 214.

[0044] Mixer 214 mixes the transmitted waves from transmitting antennas 212a, 212b, etc. with the reflected waves from receiving antennas 213a, 213b, etc. to generate multiple intermediate frequency (IF) signals. Mixer 214 transmits the generated IF signals to ADC 215.

[0045] The ADC215 converts the analog IF signal into a digital signal and transmits the digital signal to the processor 230. Memory 220 stores programs that describe the processes to be executed by the processor 230. Memory 220 also stores various data used to calculate vital sign values ​​based on signals output from the millimeter-wave sensor 210. Furthermore, memory 220 can also store vital sign values ​​measured periodically.

[0046] The processor 230 calculates the distance to a person around the vital sign measuring device 200, the direction (angle) of that person, and their movement speed, based on the signal output from the millimeter-wave sensor 210. For example, the processor 230 performs various processes, such as Fourier transform, on the signal acquired from the millimeter-wave sensor 210 to calculate the distance, the direction (angle) of the object, and their movement speed. Furthermore, the processor 230 calculates vital sign values ​​such as heart rate and respiratory rate based on information such as the distance, angle, and movement speed to the person being measured. The processor 230 is a CPU, MPU, or DSP.

[0047] The wireless communication interface 240 transmits data indicating the vital sign values ​​calculated by the processor 230 to the terminal device 100. This vital sign measuring device 200 can measure the vital signs of individuals in various spaces. For example, if the vital sign measuring device 200 is installed in an office, the vital signs of employees can be monitored. This allows for the estimation of, for instance, the employee's stress level and concentration level. If the vital sign measuring device 200 is installed in an office, it can also be used to detect employees who are experiencing harassment. Furthermore, if the vital sign measuring device 200 is installed around a bank ATM (Automatic Teller Machine), it can be used to identify customers who are being deceived by fraud.

[0048] If the Vital Signs Measuring Device 200 is installed in facilities such as daycare centers, it can be used to monitor the nap status of infants. Furthermore, if the Vital Signs Measuring Device 200 is installed in the shuttle buses of daycare centers, it can detect the presence of children who have been left behind on the bus. If the Vital Signs Measuring Device 200 is installed in elderly care facilities or detention facilities such as police stations, it can detect changes in the condition of the inmates in these facilities.

[0049] Thus, the vital signs measuring device 200 can be used in a variety of locations. In vital signs measuring devices 200 installed in various locations, the reflected waves received by the millimeter-wave sensor 210 contain various types of noise. Therefore, the vital signs measuring device 200 performs signal processing (noise processing) to remove this noise. One type of noise processing is the removal of signals with frequencies that are impossible to represent as vital signs.

[0050] Figure 5 shows an example of noise processing. For example, consider the case of measuring respiration. When the person being measured breathes, the position of the person's chest moves back and forth as shown in waveform 41. The vital sign measuring device 200 measures the position of the chest using the millimeter-wave sensor 210 and estimates the respiratory rate from the phase signal 42 that may be shown in the measurement result. In the phase signal 42, the horizontal axis is time and the vertical axis is phase. The fluctuation in phase represents the change in the position of the body surface of the person being investigated.

[0051] When the millimeter-wave sensor 210 of the vital signs measuring device 200 measures the position of a person being measured, the reflected waves from that person contain various noises. Therefore, the vital signs measuring device 200 performs noise processing on the phase signal 42, which indicates the time change in distance to the body surface of the person.

[0052] The vital signs measuring device 200 is designed to measure individuals of various age groups, so noise processing is performed assuming all age groups. Generally, a person's respiratory rate per minute is 0.1 Hz to 0.5 Hz (6 bpm to 30 bpm (Beats Per Minute)). Therefore, noise processing can be performed to, for example, retain only the frequency components between 6 bpm and 30 bpm and eliminate other frequency components. In the phase signal 43 after noise processing, fine jaggedness (components with frequencies higher than the vital range) has been removed.

[0053] It is known that the normal respiratory rate range depends on the age of the person being measured. For example, infants tend to have a higher respiratory rate (20 bpm or higher), while adults and the elderly tend to have a lower respiratory rate (10 bpm to 20 bpm, etc.). If signal processing is performed within a uniform range (6 bpm to 30 bpm) without considering these age-related changes in humans, a lot of noise will remain. A large amount of remaining noise degrades the accuracy of vital sign calculations.

[0054] Therefore, in the vital sign measuring device 200 according to the second embodiment, more appropriate values ​​for signal processing parameters are selected based on the attributes of the person being measured, and the values ​​of vital signs are estimated using those parameter values. This improves the accuracy of vital sign measurement.

[0055] Next, we will specifically explain the data used to measure vital signs using the vital sign measuring device 200, and the functions for measurement. Figure 6 is a block diagram illustrating an example of the data and functions of a vital signs measuring device. Memory 220 stores a posture estimation model 221, a sex determination model 222, age determination rules 223, and a vital signs range DB (Data Base) 224.

[0056] The posture estimation model 221 is a pre-trained machine learning model that estimates the posture of a person being measured based on a point cloud representing the position of the person's surface. The posture estimation model 221 is, for example, a neural network model. Using the posture estimation model 221, it is possible to estimate postural feature information such as whether the person being measured is hunched over like an elderly person, crawling like an infant, or what their body type is.

[0057] The gender determination model 222 is a pre-trained machine learning model that determines the gender of a person being measured based on point cloud feature information such as point cloud velocity and pose feature information estimated from the point cloud. The gender determination model 222 is, for example, a neural network model. The gender determination model 222 is, for example, a model trained on data showing the movements of people of various age groups. Data showing the movements of people of various age groups can be obtained by motion capture.

[0058] When training the gender determination model 222 based on data acquired by motion capture, the frame rate of the motion capture dataset is reduced to the number of point cloud data acquisitions per second by the millimeter-wave sensor 210. The number of point cloud data acquisitions per second by the millimeter-wave sensor 210 is, for example, 10 FPS (Frames Per Second). By matching the number of frames per second of the training data to the number of point cloud data acquisitions per second by the millimeter-wave sensor 210, the accuracy of gender determination based on posture features acquired from the millimeter-wave sensor 210 is improved.

[0059] Age determination rule 223 is information that shows the rules for determining the age of a person being measured based on point cloud feature information, posture feature information, gender information, etc. Using age determination rule 223, for example, it is possible to estimate the age of the person being measured, such as whether they are an infant, an adult, or an elderly person.

[0060] The Vitals Range DB224 is a database that shows appropriate value ranges for various vital signs for different age groups. For example, the Vitals Range DB224 has appropriate ranges set for heart rate per minute and respiratory rate per minute. In the example in Figure 7, the respiratory rate range for infants is 0.25Hz to 0.5Hz in signal frequency. This corresponds to a respiratory rate of 18bpm to 30bpm per minute. The respiratory rate range for adults is 0.15Hz to 0.4Hz in signal frequency. This corresponds to a respiratory rate of 12bpm to 25bpm per minute.

[0061] The processor 230 includes a point cloud acquisition unit 231, an attribute estimation unit 232, a vital range determination unit 233, a noise processing unit 234, and a vital counting unit 235. The point cloud acquisition unit 231 acquires a signal from the millimeter-wave sensor 210 based on reflected waves from the surroundings, and acquires a point cloud indicating the position of the person being measured based on that signal. The point cloud acquisition unit 231 calculates the distance to the point where the transmitted wave was reflected and the angle indicating the direction of that point by performing processing such as a Fourier transform on the signal acquired from the millimeter-wave sensor 210. The point cloud acquisition unit 231 generates a set of points (point cloud) in three-dimensional space from the distance and angle. The point cloud acquisition unit 231 can also calculate the movement speed of the person being measured based on the phase difference between the acquired signals.

[0062] The point cloud acquisition unit 231 transmits point cloud information, including the point cloud of the person being measured and their movement speed, to the attribute estimation unit 232. The point cloud acquisition unit 231 also transmits a phase signal indicating the change in position of points in the point cloud whose positions change periodically to the noise processing unit 234.

[0063] The attribute estimation unit 232 estimates the age of the person being measured based on the point cloud information. For example, the attribute estimation unit 232 inputs the point cloud information into the posture estimation model 221 to estimate the posture of the person being measured. The attribute estimation unit 232 also inputs the posture of the person being measured into the gender determination model 222 to estimate the gender of the person being measured.

[0064] The attribute estimation unit 232 then estimates the age of the person being measured based on point cloud features, posture features, gender, etc. For example, the attribute estimation unit 232 estimates the age using the age determination rule 223. The attribute estimation unit 232 then transmits information indicating the estimated age to the vital range determination unit 233.

[0065] The vital sign range determination unit 233 determines an appropriate range of vital sign values ​​according to the age of the person being measured. For example, the vital sign range determination unit 233 refers to the vital sign range DB 224 to determine an appropriate range for heart rate per minute and an appropriate range for respiratory rate per minute. The vital sign range determination unit 233 transmits vital sign range information, which indicates the determined range of vital sign values, to the noise processing unit 234.

[0066] The noise processing unit 234 generates a phase signal representing the skin movement of the person being investigated based on the point cloud information. The noise processing unit 234 then performs noise processing on the phase signal representing skin movement based on the vital range information. For example, the noise processing unit 234 performs a Fourier transform on the phase signal containing noise and extracts the phase signal in the frequency band indicated in the vital range information. The noise processing unit 234 transmits data showing the phase signal after noise processing to the vital counting unit 235.

[0067] The vital signs counting unit 235 counts vital signs based on data showing the phase signal after noise processing. For example, the vital signs counting unit 235 counts the heart rate per minute and the respiratory rate per minute. The vital signs counting unit 235 outputs vital data showing the counted vital signs to the wireless communication interface 240.

[0068] The vital data output from the vital counting unit 235 is transmitted to the terminal device 100 via the wireless communication interface 240. The functions of each element shown in Figure 6 can be realized, for example, by having the processor 101 execute the program module corresponding to that element.

[0069] Next, we will explain the process of counting vital sign values, which involves noise reduction tailored to the age of the individuals being studied. Figure 7 shows an example of the processing of counting vital sign values. The point cloud acquisition unit 231 generates a point cloud representing the body surface of the person being investigated based on the signal output from the millimeter-wave sensor 210. The point cloud acquisition unit 231 transmits point cloud information 51 indicating the generated point cloud to the attribute estimation unit 232. The point cloud acquisition unit 231 also generates a phase signal 52 indicating the time change of a predetermined position on the body surface of the person being investigated, and transmits the generated phase signal 52 to the noise processing unit 234. The predetermined position on the body surface for which the phase signal 52 is generated is, for example, a position near the chest of the person in question. The position near the chest is, for example, a position that vibrates periodically.

[0070] The attribute estimation unit 232 estimates the attributes of the person being investigated based on the point cloud information 51. The attribute estimation unit 232 transmits attribute information indicating the estimated attributes to the vital range determination unit 233. The vital range determination unit 233 determines the vital range of the person being investigated by referring to the vital range DB 224. The vital range DB 224 contains valid value ranges for various vital signs, associated with attributes of the person being investigated, for example. Attributes include age, such as infant or adult. Examples of vital sign value ranges include respiratory rate range and heart rate range. For example, if the attribute of the person being investigated is adult, the vital range determination unit 233 obtains the value range of each vital sign associated with the attribute "adult" in the vital range DB 224 and transmits it to the noise processing unit 234 as vital range information.

[0071] The noise processing unit 234 performs noise processing on the phase signal 52 based on vital range information. For example, the noise processing unit 234 extracts frequency components indicated by the respiratory rate range from the phase signal 52 and generates a phase signal 53 for respiratory rate counting. The noise processing unit 234 also extracts frequency components indicated by the heart rate range from the phase signal 52 and generates a phase signal 54 for heart rate counting. The noise processing unit 234 transmits the noise-removed phase signals 53 and 54 for each vital sign to the vital sign counting unit 235.

[0072] The vital signs counting unit 235 counts values ​​for each vital sign. For example, the vital signs counting unit 235 counts the respiratory rate and heart rate per minute based on the phase signals 53 and 54. In this way, by performing noise processing according to the attributes of the person being studied, it becomes possible to obtain highly accurate vital information. Specifically, the vital sign measuring device 200 utilizes the fact that the range of human vital signs (range of respiratory rate and heart rate) differs with age, and adjusts the values ​​of the noise processing parameters according to the attributes of the person being studied. This improves the accuracy of the vital information.

[0073] Next, we will explain the attribute estimation process in detail. Figure 8 shows an example of attribute estimation processing. The attribute estimation unit 232 takes multiple time-series point clouds 51a, 51b, ... acquired at regular intervals (e.g., 1 / 10 second intervals) as input to the posture estimation model 221 to estimate the posture of the person being investigated. The estimated posture information 61a, 61b, ... is, for example, a skeleton model representing the human skeleton. The time-series generated posture information 61a, 61b, ... represents the movements of the person being investigated. Since men and women have different postures, gender can be estimated based on the posture information 61a, 61b, ...

[0074] The attribute estimation unit 232 also generates point cloud feature information 62 that shows the characteristics of the point clouds 51a, 51b, ... The point cloud feature information 62 includes, for example, the number of points per point cloud, the movement speed of the person being investigated represented by the point clouds 51a, 51b, ... The point cloud feature information 62 may also include the extent of the point clouds.

[0075] The attribute estimation unit 232 takes posture information 61a, 61b, ... as input to the gender determination model 222 to estimate the gender of the person being investigated. As a result of gender estimation, the attribute estimation unit 232 obtains gender information 63 indicating whether the person is male or female. The attribute estimation unit 232 applies the point cloud feature information 62 and the gender information 63 to the age determination rule 223 to determine the age of the person being investigated. The attribute estimation unit 232 outputs age information 65 indicating the determined age.

[0076] Figure 9 shows an example of an age determination rule. For example, age determination rule 223 includes infant determination rule 223a and non-infant determination rule 223b. Infant determination rule 223a sets a numerical range for clusters that indicates the conditions for determining whether a person is an infant, for example, by associating it with the characteristics of the cluster corresponding to the person being investigated when the point cloud is clustered. In the example in Figure 9, the condition is set that the number of points within a cluster is 10 or more and less than 30. The numerical range for the number of points within a cluster is determined according to the resolution of the millimeter-wave sensor. Infant determination rule 223a also sets a condition that the cluster size is 1 m or less in height and 0.5 m or less in width. For example, if all of these conditions are met, the age of the person being investigated is determined to be an infant.

[0077] Rule 223b for determining whether a person is not an infant applies to individuals who have been determined not to be infants under Rule 223a. Rule 223b indicates that the statistically most likely age group is identified based on the individual's physical characteristics (height, weight, movement speed, etc.), and the individual's age is determined to be that identified age group. The individual's movement speed can be considered as their walking speed and is compared with statistical information on walking speed for each age group.

[0078] Figure 10 shows an example of age determination using infant identification rules. The millimeter-wave sensor 210 senses a point cloud generated from the movement and body movements of the person being investigated around the vital sign measuring device 200. Each point that makes up the point cloud is represented by coordinates in three-dimensional space. The point cloud also includes points generated by noise.

[0079] The attribute estimation unit 232 performs point cloud clustering to identify point clouds generated from people using clustering algorithms such as DBSCAN (Density-Based Spatial Clustering of Applications with Noise). In clustering, points with similar data densities are included in the same cluster 66. For example, when a point is selected, if the number of points within a certain distance from that point exceeds a threshold, a cluster containing the points within that distance is generated. Points included in cluster 66 are thought to represent people (the subjects of the survey). Points not included in cluster 66 are thought to be points generated by noise.

[0080] The attribute estimation unit 232 determines the height and width of the cluster 66 representing a person. The height is, for example, the distance between the furthest points on the surface of the curved shape (e.g., a spheroid) that encloses the cluster 66. If the curved shape is a spheroid rotated around the major axis of an ellipse, the major axis of the ellipse becomes the height of the cluster 66. The width is, for example, the length of the minor axis of the ellipse when the curved shape encloses the cluster 66 is a spheroid rotated around the major axis of an ellipse.

[0081] For example, if the number of points in cluster 66 is between 10 and 30, and the height of cluster 66 is 1m or less and the width is 0.5m or less, the person being investigated is determined to be an infant. Figure 11 shows an example of age determination using the non-infant determination rule. The non-infant determination rule 223b includes statistical information 67a, 67b, 67c, ... such as height, weight, and walking speed. For example, statistical information 67a shows the frequency (frequency distribution) of height for each height range for school children, adults, and the elderly, separated by gender.

[0082] The attribute estimation unit 232 refers to statistical information corresponding to the gender of the person being investigated and identifies the most likely age group based on the characteristics of that person. For example, consider the case where the person being investigated is female and 160 cm tall. Referring to the frequency distribution of heights for school-aged girls, the frequency of 160 cm is low. Referring to the frequency distribution of heights for adult women, the frequency of 160 cm is high. Referring to the frequency distribution of heights for elderly women, the frequency of 160 cm is low. Therefore, the age of the person being investigated is determined to be adult.

[0083] The attribute estimation unit 232 determines the age of the person being investigated based on each of several characteristics of the person being investigated (height, weight, speed of movement, etc.), and sets the age that is most frequently determined as the age of the person being investigated.

[0084] In this way, the age of the person being studied is determined. As a result, noise processing can be performed using parameters appropriate to the age group, and vital sign values ​​can be measured with high accuracy.

[0085] Figure 12 is a flowchart showing an example of the procedure for measuring vital signs. The process shown in Figure 12 will be explained below according to the step numbers. [Step S101] The point cloud acquisition unit 231 acquires the signal output from the millimeter-wave sensor 210.

[0086] [Step S102] The point cloud acquisition unit 231 acquires a point cloud indicating the surface positions of the person being investigated based on the acquired signals. The point cloud acquisition unit 231 generates point cloud information representing the point cloud.

[0087] [Step S103] The attribute estimation unit 232 estimates whether the person being investigated is an infant or young child based on the point cloud information. For example, if all the conditions shown in the infant determination rule 223a are met, the attribute estimation unit 232 determines that the person being investigated is an infant or young child. If the attribute estimation unit 232 determines that the person is an infant or young child, it proceeds to step S104. If the attribute estimation unit 232 determines that the person is not an infant or young child, it proceeds to step S105.

[0088] [Step S104] The attribute estimation unit 232 sets "infant" as the age information and proceeds to step S107. [Step S105] The attribute estimation unit 232 uses the gender determination model 222 to determine the gender of the person being investigated.

[0089] [Step S106] The attribute estimation unit 232 determines the age of the person being investigated according to the non-infant determination rule 223b. [Step S107] The vital range determination unit 233 obtains vital ranges corresponding to the age of the person being investigated. For example, the vital range determination unit 233 extracts records from the vital range DB 224 in which the age of the person being investigated is set as an attribute, and obtains one or more vital ranges such as respiratory rate range and heart rate range from those records.

[0090] [Step S108] The noise processing unit 234 performs noise processing on the signal acquired in step S101 based on the acquired vital range. For example, the noise processing unit 234 extracts only the frequency components within the vital range from the signal.

[0091] [Step S109] The vital signs counting unit 235 counts vital signs based on the signal after noise processing. For example, the vital signs counting unit 235 counts heart rate per minute, respiratory rate per minute, etc.

[0092] [Step S110] The vital sign counting unit 235 transmits vital data showing the values ​​of the counted vital signs. For example, the vital sign counting unit 235 outputs the vital data to the wireless communication interface. The wireless communication interface 240 transmits the vital data to the terminal device 100.

[0093] In this way, highly accurate vital data can be obtained that takes into account the age of the person being surveyed. Figure 13 shows an example of vital sign measurement. For example, consider the case where the respiratory rate of a person 33 per minute is measured using a vital sign measuring device 200. Based on the reflected waves of the transmitted waves sent to the surroundings by the millimeter-wave sensor 210 of the vital sign measuring device 200, a phase signal 71 representing the variation in distance to the skin surface of person 33 is generated. Point cloud information 72 indicating the position of the skin surface of person 33 in three-dimensional space is also generated based on the reflected waves. Furthermore, posture information 73 and point cloud feature information 74 are generated based on the point cloud information 72.

[0094] The point cloud feature information 74 includes information such as the number of points within a cluster ("24"), the cluster size ("height 0.5m, width 0.3m"), and the movement speed ("0m / s"). Comparing the point cloud feature information 74 with the infant identification rule 223a shown in Figure 9, the number of points within a cluster falls within the numerical range (10 or more and less than 30), and the cluster size also falls within the numerical range (height 1m or less, width 0.5m or less). Therefore, attribute estimation based on the infant identification rule 223a estimates that person 33 is an infant.

[0095] In the vital range DB224, the respiratory rate range for the attribute "infant" is "0.25Hz to 0.5Hz". Therefore, noise processing is performed to extract the frequency range of "0.25Hz to 0.5Hz" from the phase signal 71, and a respiratory waveform 75 indicating respiration is generated. Based on the respiratory waveform 75, the respiratory rate per minute (number of waves per minute) is counted. Then, vital data 76 indicating the respiratory rate is output.

[0096] In this way, the vital signs measuring device 200 estimates attributes (rough age) such as infant, school-aged child, adult, and elderly person from point cloud information 72 of 33 individuals. For example, the approximate body size is estimated from the scores within clusters, and it is determined whether the person is an infant or someone else.

[0097] If the subject is not an infant, their age (school-aged, adult, elderly, etc.) is estimated based on statistical information. Then, an appropriate range of vital signs corresponding to the estimated attribute is obtained and used as a parameter for noise reduction. As a result, highly accurate vital signs data is obtained.

[0098] Figure 14 shows an example of noise processing according to vital range. Person 32 is an elderly person, and person 33 is an infant. When noise processing is performed on the phase signals 81 and 82 obtained from persons 32 and 33 respectively using a uniform vital range, signals 83 and 84 with a lot of noise remain are output. Even if vital sign values ​​are measured based on these signals 83 and 84, only measurement results with low accuracy can be obtained.

[0099] On the other hand, in the vital sign measuring device 200 according to the second embodiment, the phase signal 81 of person 32, who is an elderly person, is subjected to noise processing within the vital sign range for the elderly, and the phase signal 82 of person 33, who is an infant, is subjected to noise processing within the vital sign range for infants. In this case, signals 85 and 86 with sufficient noise removal are output. By measuring the vital sign values ​​based on these signals 85 and 86, highly accurate measurement results can be obtained.

[0100] [Other embodiments] The attribute estimated by attribute estimation may be gender. In that case, signal processing is performed using gender-specific parameters (e.g., vital range).

[0101] The process for determining vital sign values ​​based on the output signal from the millimeter-wave sensor 210 can also be performed by a device outside the vital sign measuring instrument 200 (e.g., a terminal device 100). In this case, the signal output from the millimeter-wave sensor 210 is transmitted from the vital sign measuring instrument 200 to the external device.

[0102] Although embodiments have been illustrated above, the configurations of each part shown in the embodiments can be replaced with others having similar functions. Furthermore, other arbitrary components or processes may be added. Moreover, any two or more configurations (features) from the embodiments described above may be combined. [Explanation of symbols]

[0103] 1 person 2 mm wave sensor 3. First signal 4-point cloud information 5. Second signal 6. Vital Data 7 Parameter Information 10 Information Processing Devices 11 Storage section 12 Processing Units

Claims

1. Based on a first signal acquired from a millimeter-wave sensor, point cloud information is generated indicating multiple locations on the body surface of a person in the vicinity of the millimeter-wave sensor. Based on the aforementioned point cloud information, the attributes of the person are estimated. Based on the estimated attributes, the values ​​of the signal processing parameters are determined. By performing signal processing on the first signal according to the determined parameter values, a second signal is generated that indicates the time change of the person's position on the body surface. Based on the second signal, a predetermined value of a vital sign is determined. A vital signs measurement program that has a computer perform the processing.

2. In the process of determining the value of the parameter, the frequency band to be extracted in noise processing is determined as the value of the parameter based on the estimated attribute. In the process of generating the second signal, the second signal is generated by removing frequency components other than the frequency band indicated by the value of the parameter. The vital sign measurement program according to claim 1.

3. In the process of estimating the aforementioned attributes, the age of the person is estimated. The vital sign measurement program according to claim 1.

4. In the process of estimating the aforementioned attributes, a set of points representing the person is generated from the points included in the point cloud information, and the age is estimated based on the set of points. The vital sign measurement program according to claim 3.

5. In the process of estimating the aforementioned attributes, the gender of the person is estimated based on the point cloud information, and the age of the person is estimated based on the estimated gender and the point cloud information. The vital sign measurement program according to claim 3.

6. In the process of estimating the aforementioned attributes, based on the point cloud information, the values ​​of predetermined indicators representing the physical characteristics or actions of the person are determined, and based on the statistical information of the predetermined indicators for each gender and age group, the age of the person corresponding to the value of the indicator is estimated. The vital sign measurement program according to claim 5.

7. Based on a first signal acquired from a millimeter-wave sensor, point cloud information is generated indicating multiple locations on the body surface of a person in the vicinity of the millimeter-wave sensor. Based on the aforementioned point cloud information, the attributes of the person are estimated. Based on the estimated attributes, the values ​​of the signal processing parameters are determined. By performing signal processing on the first signal according to the determined parameter values, a second signal is generated that indicates the time change of the person's position on the body surface. Based on the second signal, a predetermined value of a vital sign is determined. A method of measuring vital signs in which the process is performed by a computer.

8. Based on a first signal acquired from a millimeter-wave sensor, point cloud information is generated indicating multiple locations on the body surface of a person in the vicinity of the millimeter-wave sensor. A processing unit that estimates the attributes of the person based on the point cloud information, determines the values ​​of signal processing parameters based on the estimated attributes, and generates a second signal indicating the time change of the person's position on the body surface by performing signal processing on the first signal according to the determined parameter values, and determines predetermined vital sign values ​​based on the second signal. An information processing device having

Images