Heart rate identification device, heart rate identification method, and program

Abstract

To identify a human heart rate in a shock state in a non-contact manner.SOLUTION: A heart rate identification device comprises: a reflected wave measurement unit which measures an intensity in a time domain of an electromagnetic wave reflected when microwaves are output toward a human; a PSD calculation unit which calculates a power spectral density (PSD) on the basis of the intensity of the electromagnetic wave in the time domain; and a heart rate identification unit which identifies the human heart rate on the basis of the calculated PSD. The PSD calculation unit performs an overlap process on the intensity of the electromagnetic wave in the time domain, calculates a PSD from each frame extracted by the overlap process, performs filtering process and addition average process on the PSD calculated from each frame by a comb filter having, as a fundamental frequency, a frequency of a component representing a human respiration rate to finally calculate the PSD.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a heart rate determining device, a heart rate determining method, and a program. [Background technology]

[0002] Among cardiovascular diseases, there is a condition called "shock." "Shock" is defined as "an acute syndrome that can lead to a life-threatening condition due to a sudden drop in blood pressure," and corresponds to a systolic blood pressure (SBP) of less than 90 mmHg. In particular, when the SBP is less than 80 mmHg, "impalpable compensatory tachycardia" is observed, and when the SBP reaches 60 mmHg, even the carotid arteries become impalpable.

[0003] Shock is a condition in which early detection is extremely important, even in home environments, as the mortality rate increases by 7.6% for every hour that appropriate treatment is delayed. Therefore, early detection in home environments is required. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Matsui T, Kobayashi T, Hirano M, Kanda M, Sun G, Otake Y, et.al. A Pneumonia Screening System based on Parasympathetic Activity Monitoring in Non-contact Way using Compact Radars Beneath the Bed Mattress. J Infect. 2020;81:e142-e144 [Non-patent document 2] Sun G, Trung NV, Hoi LH, Hiep PT, Ishibashi K, Matsui T. Visualization of epidemiological map using an Internet of Things infectious disease surveillance platform. Crit Care. 2020:24:400. Summary of the Invention [Problem to be solved by the invention]

[0005] A method has been developed to measure the respiratory rate and heart rate of a human subject without contact by irradiating the subject with electromagnetic waves such as microwaves and capturing the body surface movements from the reflected electromagnetic waves.However, it is difficult to determine the heart rate of a person in a state of shock using a non-contact measurement method. An object of the present invention is to provide a heart rate determination device that can determine the heart rate of a person in a shock state without contact. [Means for solving the problem]

[0006] One aspect of the present invention is a heart rate determination device that includes a reflected wave measurement unit that measures the intensity in the time domain of electromagnetic waves reflected when microwaves are output to a human, a PSD calculation unit that calculates a power spectral density (PSD) based on the intensity of the electromagnetic waves in the time domain, and a heart rate determination unit that determines the heart rate of the human based on the calculated PSD, wherein the PSD calculation unit performs overlap processing on the intensity of the electromagnetic waves in the time domain, calculates a PSD from each frame extracted by the overlap processing, and finally calculates the PSD by filtering the PSD calculated from each frame using a comb filter with a fundamental frequency that is the frequency of a component that indicates the human's respiratory rate and by averaging. [Effects of the Invention]

[0007] According to the present invention, the heart rate of a person in a state of shock can be determined without contact. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a diagram showing the configuration of a measurement system 1 according to the present embodiment. [Figure 2] 1 is a diagram showing the configuration of a heart rate specifying device 12 according to the present embodiment. [Figure 3] 10 is a diagram showing an example of the intensity of an electromagnetic wave in the time domain measured by the reflected wave measuring unit 122. FIG. [Figure 4] 10 is a diagram showing an example of a PSD calculated by the PSD calculation unit 123 and a corresponding frequency. FIG. [Figure 5] FIG. 10 is a diagram illustrating a method for calculating a PSD in the present embodiment. [Figure 6] 4 is a flowchart showing the operation of the heart rate specifying device 12 according to the present embodiment. [Figure 7] FIG. 10 is a diagram illustrating the correlation between the identified heart rate and the actual heart rate. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. 1 is a diagram showing the configuration of a measurement system 1 according to this embodiment. The measurement system 1 includes a Doppler radar 11 and a heart rate specifying device 12.

[0010] The Doppler radar 11 outputs microwaves to the body surface of the human P, who is the measurement target. The Doppler radar 11 is installed, for example, under a mattress or bed on which the human P lies. The frequency of the microwaves radiated from the Doppler radar 11 is, for example, 24 GHz. The human P, who is the measurement target, is, for example, a patient.

[0011] Electromagnetic waves reflected by the body surface of the person P are input to the Doppler radar 11. The heart rate identification device 12 measures the intensity of the electromagnetic waves input to the Doppler radar 11. The electromagnetic waves reflected by the body surface of the person P contain information about the pulse of the arteries on the body surface of the person P.

[0012] 2 is a diagram showing the configuration of the heart rate identifying device 12 according to this embodiment. The heart rate identifying device 12 includes a radar control unit 121, a reflected wave measurement unit 122, a PSD calculation unit 123, a heart rate identifying unit 124, a shock determination unit 125, and an output unit 126.

[0013] The radar control unit 121 controls the Doppler radar 11. The radar control unit 121 controls the Doppler radar 11 and causes the Doppler radar 11 to output microwaves.

[0014] The reflected wave measuring unit 122 measures, in the time domain, the intensity of the electromagnetic wave reflected from the person P and input to the Doppler radar 11. Fig. 3 is a diagram showing an example of the intensity of the electromagnetic wave in the time domain measured by the reflected wave measuring unit 122.

[0015] The PSD calculation unit 123 calculates the power spectrum density (PSD) of the electromagnetic wave based on the intensity of the electromagnetic wave in the time domain. The PSD calculation unit 123 calculates the PSD, which is the intensity of the electromagnetic wave in the frequency domain, by, for example, Fourier transforming the intensity of the electromagnetic wave in the time domain. The Fourier transform method is, for example, fast Fourier transform (FFT). Fig. 4 is a diagram showing an example of the PSD calculated by the PSD calculation unit 123 and the corresponding frequency. The PSD calculation unit 123 may calculate the PSD by calculating a spectrogram from the intensity of the electromagnetic wave in the time domain.

[0016] In the PSD calculated using conventional methods, the component indicating the respiratory rate and the higher harmonics and noise components derived from the component indicating the respiratory rate are strong, while the component indicating the heart rate is very small. In the example shown in Figure 4, a component indicating the respiratory rate is seen at approximately 0.5 Hz, and higher harmonics derived from the component indicating the respiratory rate are seen at approximately 1 Hz and 2 Hz, making it difficult to identify the component indicating the heart rate. Therefore, PSD calculation section 123 calculates the PSD so that the heart rate can be identified. The PSD calculation method will be described in detail below.

[0017] The PSD calculation unit 123 extracts frames, which represent the intensity of the electromagnetic wave over a predetermined period of time, from the intensity of the electromagnetic wave in the time domain. Here, the PSD calculation unit 123 performs overlap processing to extract frames so that consecutive frames overlap in time. The PSD calculation unit 123 calculates a PSD from each frame. As a result, the same number of PSDs as the number of extracted frames are calculated.

[0018] The PSD calculation unit 123 applies a comb filter to the calculated PSDs, with the frequency of the component representing the respiratory rate as its fundamental frequency, to remove the component representing the respiratory rate and higher harmonics derived from the component representing the respiratory rate. In the example shown in Fig. 4, the application of the comb filter removes the component representing the respiratory rate with a frequency of 0.5 Hz and higher harmonics derived from the component representing the respiratory rate with frequencies of 1 Hz, 1.5 Hz, and 2 Hz.

[0019] When applying a comb filter, for example, the frequency at which the intensity is greatest in the PSD calculated by the PSD calculation unit 123 may be determined to be the frequency of the component indicating the respiratory rate, and components with frequencies equal to the frequency of the determined component indicating the respiratory rate may be removed.

[0020] Furthermore, PSD calculation section 123 performs an averaging process on the PSD calculated from each frame and applied with the comb filter, thereby calculating an averaged PSD.

[0021] 5 is a diagram showing a method for calculating the PSD in this embodiment. PSD calculation unit 123 performs overlap processing on the intensity of the electromagnetic wave in the time domain, calculates the PSD from each frame, applies a comb filter to the PSD calculated from each frame, and performs averaging processing to calculate the final PSD.

[0022] The final PSD calculation is performed using overlap and averaging processes, which reduces noise components that occur during PSD calculation. Furthermore, a comb filter with a fundamental frequency that corresponds to the respiratory rate is applied, eliminating the respiratory rate component and higher harmonics derived from the respiratory rate component. This makes it easier to identify the heart rate component.

[0023] The PSD calculation unit 123 may calculate a final PSD based on PSDs calculated from a predetermined number of frames, or may perform averaging until a maximum peak is observed in the PSD calculated by averaging. "A maximum peak is observed" in the PSD means, for example, that the largest maximum value in the PSD is greater than or equal to a predetermined multiple of the other maximum values. When a maximum peak is observed in the PSD calculated by averaging, the PSD calculation unit 123 may terminate the averaging and use the PSD calculated at the time of termination as the final PSD. This allows for the calculation of a PSD that more reliably identifies a component indicating the heart rate.

[0024] Furthermore, if a maximum peak is found in a PSD calculated from one frame and to which a comb filter has been applied, PSD calculation unit 123 may use this PSD as the final PSD to be calculated without performing averaging processing.

[0025] Furthermore, the PSD calculation unit 123 applies a comb filter to the PSD calculated from each frame, and then calculates the arithmetic mean of the PSD to which the comb filter has been applied. However, the comb filter may be applied after calculating the arithmetic mean of the PSD calculated from each frame.

[0026] The heart rate identification unit 124 identifies the frequency of the component indicating the heart rate. The heart rate identification unit 124 identifies the frequency of the component with the maximum intensity in the PSD finally calculated by the PSD calculation unit 123 as the frequency of the component indicating the heart rate.

[0027] If the frequency of the component indicating the heart rate is a [Hz] and the heart rate is b [bpm], then there is a relationship between a and b, b = 60a. The heart rate determination unit 124 may calculate the heart rate from the frequency of the component indicating the heart rate using this formula.

[0028] The shock determination unit 125 determines whether the person P is in a state of shock based on the heart rate and respiratory rate. The respiratory rate is a value calculated based on the frequency of the component indicating the respiratory rate. If the frequency of the component indicating the respiratory rate is c [Hz] and the respiratory rate is d [bpm], then there is a relationship between c and d, d = 60c.

[0029] The shock determination unit 125 determines that the higher the respiratory rate and / or heart rate, the more the person is in a shock state. This is because when a person is in a shock state and blood pressure drops, the respiratory rate and heart rate increase compensatory. The shock determination unit 125 calculates a score, for example, using equation (1), and determines that the person is in a shock state if the calculated score is 0 or greater, and determines that the person is not in a shock state if the calculated score is less than 0.

number

[0030] In equation (1), HR is the heart rate, RR is the respiratory rate, and α, β, and γ are values ​​that are optimized through experiments, etc.

[0031] The output unit 126 outputs the determination result by the shock determination unit 125. The determination result is input to, for example, an external display device, and the determination result is displayed. The determination result may be transmitted to, for example, a device used by a medical professional. For example, when the output unit 126 determines that the person P is in a shock state, the output unit 126 may transmit a notification to the medical professional urging them to perform a shock test.

[0032] 6 is a flowchart showing the operation of the heart rate identification device 12 according to this embodiment. The radar control unit 121 controls the Doppler radar 11 to cause the Doppler radar 11 to output microwaves (step S11). The reflected wave measurement unit 122 measures the intensity of electromagnetic waves generated by the output of microwaves in the time domain (step S12). The PSD calculation unit 123 performs overlap processing on the intensity of the electromagnetic waves in the time domain and extracts frames of a certain period in which signals overlap (step S13). The PSD calculation unit 123 calculates the PSD from each frame (step S14).

[0033] The PSD calculation unit 123 applies a comb filter, whose fundamental frequency is the frequency of the component representing the respiratory rate, to the PSD calculated from each frame (step S15). This removes the component representing the respiratory rate and higher harmonics derived from the component representing the respiratory rate. The PSD calculation unit 123 calculates an average PSD by performing averaging on the PSD calculated from each frame and to which the comb filter has been applied (step S16). The PSD calculation unit 123 performs overlap processing and averaging to reduce noise components that occur during PSD calculation.

[0034] The heart rate identification unit 124 identifies the frequency of the component indicating the heart rate of the person P based on the averaged PSD (step S17). The shock determination unit 125 determines whether the person P is in a shock state based on the heart rate and respiratory rate of the person P (step S18). The output unit 126 outputs the determination result (step S19).

[0035] As described above, the heart rate identification system 1 can identify the heart rate during shock without coming into contact with the person P who is the measurement target. This makes it possible to automatically identify the heart rate of a person who may go into shock and automatically determine whether or not the person is in shock. Furthermore, by notifying a medical professional of the shock state determination result, it is possible to automatically prompt the medical professional to perform an examination, etc.

[0036] The PSD calculation unit 123 may calculate multiple PSDs for one frame by performing processing using multiple window functions on each frame. The PSD calculation unit 123 then calculates a final PSD by filtering the PSDs calculated for each frame and each window function using a comb filter whose fundamental frequency is the frequency of the component indicating the respiratory rate and averaging the PSDs. In this case, the PSD calculation unit 123 can calculate the same number of PSDs from a smaller number of frames. This allows the same number of PSDs to be calculated based on electromagnetic waves in a shorter time domain, enabling the heart rate identification device 12 to identify the heart rate and determine a shock state in a shorter time.

[0037] (experiment) The details of the experiment will be described below. Fig. 7 is a diagram showing the correlation between the identified heart rate and the actual heart rate. (a) shows the correlation between the heart rate identified by the comparative method and the actual heart rate, and (b) shows the correlation between the heart rate identified by the method of this embodiment and the actual heart rate. In the comparative method, overlap processing or averaging processing was not performed, and a comb filter was applied to the intensity of the electromagnetic wave in the time domain, with the frequency of the component indicating the respiratory rate as its fundamental frequency, and the frequency of the component with the maximum intensity was identified as the frequency of the component indicating the heart rate. In this experiment, heart rates were determined from 512 pieces of measurement data. The actual heart rates were measured by having the subjects wear an SpO2 monitor on their fingertips.

[0038] The correlation coefficient was −0.304 in the comparison method, and 0.704 in the method according to the present embodiment. This shows that the heart rate can be determined more accurately in the present embodiment based on the PSD calculated by the PSD calculation unit 123 than in the comparison method.

[0039] Furthermore, shock state was assessed for 12 hospitalized patients using the method of this embodiment. First, we created a dataset. We measured the blood pressure of 12 patients and determined whether they were in shock (whether their SBP was 90 mmHg or less). Of the 12 patients, 2 were diagnosed as being in shock and 10 were not. However, because SBP can fluctuate temporarily, we determined whether they were in shock based on SBP. A data set was created that linked the determined state of shock with the heart rate detected by the method of this embodiment during blood pressure measurement. The data set consisted of 277 pieces of data: 139 pieces of data measured from two people diagnosed with shock and 138 pieces of data measured from 10 people diagnosed with no shock. Using this data set, we verified whether a state of shock could be determined based on the detected heart rate. In the verification, the state of shock was determined by substituting α = 0.16, β = 0.15, and γ = 16.7 into equation (1).

[0040] The validity of the results was verified by five-fold cross-validation, and the sensitivity was 83%, specificity 88%, PPV 89%, and NPV 82%.

[0041] Other Embodiments One embodiment of the present invention has been described in detail above with reference to the drawings, but the specific configuration is not limited to that described above, and various design changes and the like are possible within the scope that does not deviate from the gist of the present invention.

[0042] The heart rate determination device 12 in the above-described embodiment may be partially or entirely implemented by a computer. In this case, a program for implementing the functions may be recorded on a computer-readable recording medium, and the program may be loaded into a computer system and executed. The term "computer system" as used herein includes an operating system (OS) and peripheral hardware. The term "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, as well as storage devices such as hard disks built into computer systems. The term "computer-readable recording medium" may also include media that dynamically store programs for a short period of time, such as communication lines used when transmitting programs via networks such as the Internet or telephone lines, or media that store programs for a fixed period of time, such as volatile memory within a computer system serving as a server or client. The program may be a program for implementing some of the functions described above, or may be a program that can be implemented in combination with a program already stored in the computer system, or may be implemented using a programmable logic device such as an FPGA (Field Programmable Gate Array). [Explanation of symbols]

[0043] 1 Measurement system, 11 Doppler radar, 12 Heart rate determination device, 121 Radar control unit, 122 Reflected wave measurement unit, 123 PSD calculation unit, 124 Heart rate determination unit, 125 Shock determination unit, 126 Output unit

Claims

1. a reflected wave measuring unit that measures the intensity in the time domain of the reflected electromagnetic wave when the microwave is output to a human; a power spectral density (PSD) calculation unit that calculates a power spectral density (PSD) based on the intensity of the electromagnetic wave in the time domain; a heart rate determination unit that determines a heart rate of the person based on the calculated PSD; Equipped with The PSD calculation unit Applying overlap processing to the electromagnetic wave intensity in the time domain, Calculating PSD from each frame extracted by the overlap processing; The PSD calculated from each frame is subjected to filtering using a comb filter with a fundamental frequency that is the frequency of the component indicating the human respiratory rate, and then subjected to averaging to finally calculate the PSD. Heart rate identification device.

2. The PSD calculation unit A comb filter with the fundamental frequency of the component representing the respiratory rate is applied to the PSD calculated for each frame. A final PSD is calculated by performing an averaging process on the PSD calculated for each frame and having the comb filter applied thereto.

2. The heart rate determining device of claim 1.

3. performing processing using a plurality of window functions on each of the frames to calculate a plurality of PSDs for each frame; The PSD calculated for each frame and each window function is filtered using a comb filter with the frequency of the component indicating the respiratory rate as the fundamental frequency, and then averaged to calculate the final PSD.

2. The heart rate determining device of claim 1.

4. a shock determination unit that determines whether the person is in a shock state based on the person's heart rate and respiratory rate; The heart rate determining device of claim 1 further comprising:

5. a reflected wave measuring step of measuring the intensity in the time domain of the electromagnetic wave reflected when the microwave is output to the human; a power spectral density (PSD) calculation step of calculating a power spectral density (PSD) based on the intensity of the electromagnetic wave in the time domain; determining a heart rate of the person based on the calculated PSD; and The PSD calculation step Applying overlap processing to the electromagnetic wave intensity in the time domain, Calculating PSD from each frame extracted by the overlap processing; The PSD calculated from each frame is subjected to filtering using a comb filter with a fundamental frequency that is the frequency of the component indicating the human respiratory rate, and then subjected to averaging to finally calculate the PSD. Heart rate determination method.

6. A method for determining a heart rate according to claim 5, program.

Images