JP2015217143A - Heartbeat measuring device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a heartbeat measuring device capable of calculating a correct heartbeat interval time with less amount of calculation and less influence of noises.SOLUTION: A heartbeat measuring device applies two types of simple filters that are a peak emphasis filter and a moving average filter to doppler signal data f(k) to generate smoothed differential intensity signal data f(k) (S2 to S5). The smoothed differential intensity signal data f(k) is obtained in such a manner that the doppler signal data f(k) is enhanced in its peaks but decreased in peaks caused from noises. On the basis of this smoothed differential intensity signal data f(k), the heartbeat measuring device determines plural heartbeat times (S6), and on the basis of these heartbeat times calculates a heartbeat interval time (S7), thereby, can obtain the correct heartbeat interval time with less amount of calculation and less influence of noises compared with conventional heartbeat measuring devices that use frequency analysis.

Description

  The present invention relates to a heartbeat measurement device that measures a heartbeat of a living body using the Doppler effect of radio waves.

  Conventionally, a radio wave is transmitted as a transmission wave to a living body such as a human body or an animal, a reflected wave with respect to the transmission wave is received, and a Doppler signal that is a signal having a frequency difference between the frequency of the transmission wave and the frequency of the reflected wave is obtained. 2. Description of the Related Art Heart rate measuring devices that include an output radio wave sensor (Doppler sensor) and measure a heart beat of a living body using the radio wave sensor are known (see Patent Document 1 and Patent Document 2).

  The detection device (body state detection tool) disclosed in Patent Literature 1 described above uses a radio wave sensor (microwave transmission / reception sensor) built in a bed to cope with the fine movements of the human body caused by heartbeat and respiration (reflection). A waveform of the displacement amount of the wave is generated, and by analyzing the frequency of the displacement amount waveform, the heartbeat and respiration components in the above waveform are separated, and then the heartbeat component waveform and the respiration component waveform after separation. Based on the above, heart rate and respiration rate are measured.

  Further, the detection device (heart rate estimation device) disclosed in Patent Document 2 described above includes a sensor for detecting a subject's movement (acceleration sensor or the like) in addition to a sensor for transmitting / receiving microwaves (microwave transmission / reception unit). Have. Then, from a plurality of peak frequencies (heartbeat candidates) obtained by frequency analysis of the reflected wave data received by the microwave transceiver, the peak frequency obtained by frequency analysis of the detection result by the motion detection sensor Is removed, and a heart rate is estimated based on the remaining peak frequency. Thereby, the estimation accuracy of the heartbeat is improved.

JP 2000-102515 A JP 2014-39666 A

  However, since the heartbeat measuring method in the conventional detection device disclosed in Patent Document 1 is based on the premise that frequency analysis is performed on an output signal (Doppler signal) from a radio wave sensor such as a microwave, the amount of calculation is large. It will increase. This heart rate measurement method uses frequency analysis to separate the heart rate and respiration components in the output signal from the radio wave sensor, and then uses the waveform of the separated heart rate component as it is to measure the heart rate. Therefore, it is difficult to detect fine fluctuations in the heartbeat interval time. Further, the heartbeat measuring method in the detection apparatus of Patent Document 2 is premised on frequency analysis of the output signal from the radio wave sensor, and thus the amount of calculation increases.

  The present invention solves the above-described problems, and an object of the present invention is to provide a heartbeat measuring device capable of calculating an accurate heartbeat interval time with a small amount of calculation and little influence of noise.

  In order to solve the above problems, the heartbeat measuring device of the present invention includes a radio wave transmission unit that transmits radio waves as transmission waves to a living body, a radio wave reception unit that receives reflected waves from the living body with respect to the transmission waves, A Doppler signal generator that generates a Doppler signal that is a signal having a frequency difference between the frequency of the transmission wave and the frequency of the reflected wave; and a digital signal obtained by converting the Doppler signal generated by the Doppler signal generator into a digital signal. In the heartbeat measuring device including a signal processing unit that performs signal processing, the signal processing unit samples the Doppler signal generated by the Doppler signal generation unit, and a signal value for a predetermined time obtained by the sampling Are generated by the Doppler signal data generation unit that generates Doppler signal data arranged in time series, and the Doppler signal data generation unit. By applying a peak enhancement filter to Doppler signal data, a peak enhancement signal data generation unit that generates peak enhancement signal data, and an absolute value of the peak enhancement signal data generated by the peak enhancement signal data generation unit are smoothed Based on the smoothed differential intensity signal data generated by the smoothed differential intensity signal data generating unit and the smoothed differential intensity signal data generating unit by applying the smoothing filter, A pulsation time determination unit that determines pulsation times of a plurality of heartbeats of the living body, and two pulsation times that are temporally adjacent to each other among the pulsation times of the plurality of heartbeats determined by the pulsation time determination unit And a heartbeat interval calculation unit that calculates a heartbeat interval time based on the difference time.

  In this heartbeat measuring device, the pulsation time determination unit determines the beat of the next heartbeat based on the determined pulsation time of the previous heartbeat and the previous heartbeat interval time calculated by the heartbeat interval calculation unit. It is preferable to predict a range of time that the motion time can take, and to determine the pulsation time of the next heartbeat based on the smoothed differential intensity signal data within the time range.

  The heartbeat measuring device further includes a respiration detecting unit that detects exhalation and inspiration of the living body, and the pulsation time determining unit detects the next time when the respiration detecting unit detects that the living body is exhaling. When the range of time that can be taken by the heartbeat of the heartbeat is shifted later in time, and the respiratory detector detects that the living body is inhaling, the time that can be taken by the heartbeat of the next heartbeat The range may be shifted forward in time.

  In this heartbeat measuring device, the pulsation time determination unit determines the beat of the next heartbeat based on the determined pulsation time of the previous heartbeat and the previous heartbeat interval time calculated by the heartbeat interval calculation unit. After estimating the probability distribution of motion time, the posterior distribution of the pulse time of the next heartbeat is obtained by Bayesian estimation using the probability distribution as a prior distribution and the smoothed differential intensity signal data as a likelihood function. It is preferable to determine the pulsation time of the next heartbeat based on the distribution.

  The heartbeat measuring device further includes a respiration detecting unit that detects exhalation and inspiration of the living body, and the pulsation time determining unit detects the next time when the respiration detecting unit detects that the living body is exhaling. When the respiratory detection unit detects that the living body is inhaling, the probability distribution of the next heartbeat beat time is temporally shifted. May be shifted forward.

  In this heartbeat measuring device, the pulsation time determination unit determines a time when the maximum value is maximum among the times when the smoothed differential intensity signal data is maximum within a predetermined time range. The beat interval calculation unit calculates a first beat interval time based on a result of frequency analysis of the peak enhancement signal data or the smoothed differential intensity signal data within the predetermined time range. It is preferable to calculate.

  According to the present invention, smoothed differential intensity signal data can be generated by applying two types of simple filters, a peak enhancement filter and a moving average filter, to Doppler signal data. The smoothed differential intensity signal data is signal data in which the peak due to noise is reduced while emphasizing the peak in the original Doppler signal data. Therefore, based on the smoothed differential intensity signal data, the beat time of a plurality of heartbeats is determined, and the heartbeat interval time is calculated based on the beat time of these heartbeats. Compared to the measuring apparatus, it is possible to obtain an accurate heartbeat interval time with a small amount of calculation and little influence of noise. Moreover, since the above-mentioned smoothed differential intensity signal data is signal data that emphasizes the peak in the original Doppler signal data, by determining the pulsation time of the heartbeat based on this smoothed differential intensity signal data, Unlike the heart rate measurement device that uses the heart rate component waveform separated by the conventional frequency analysis as it is, even when the heart rate interval time varies finely, the accurate heartbeat time and accurate heart rate interval time are obtained. be able to.

1 is a schematic electrical block configuration diagram of a heartbeat measuring device according to a first embodiment of the present invention. FIG. FIG. 2 is an electrical block configuration diagram of a signal processing unit in FIG. 1. The flowchart of the measurement process in the heart rate measuring device. The flowchart of the pulsation time candidate extraction process in FIG. The waveform figure of the Doppler signal data of the heart rate measuring device. The waveform figure of the peak emphasis signal data of the heartbeat measuring device. The wave form diagram of the smoothing differential intensity | strength signal data of the heart rate measuring device. The flowchart of the measurement process in the heart rate measuring device of the 2nd Embodiment of this invention. The electrical block block diagram of the heart rate measuring device. The flowchart of the measurement process in the heart rate measuring device of the 3rd Embodiment of this invention. The flowchart of the measurement process in the heart rate measuring device of the 4th Embodiment of this invention. 12 is a flowchart of a heartbeat beat time estimation process in FIG. 11. The flowchart of the measurement process in the heart rate measuring device of the 5th Embodiment of this invention. 14 is a flowchart of a heartbeat beat time estimation process in FIG. 13. The flowchart of the initial process performed in the modification of the 2nd thru | or 5th embodiment of this invention.

  Hereinafter, a heartbeat measuring device according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 shows an electrical block configuration of a heartbeat measuring device according to a first embodiment of the present invention. The heartbeat measuring device 1 includes a radio wave sensor 2 that is a kind of so-called Doppler sensor, and a signal processing unit 3. The radio wave sensor 2 transmits a radio wave such as a microwave as a transmission wave to the human body, receives a reflected wave with respect to the transmission wave, and performs Doppler according to a frequency difference between the transmission wave and the reflected wave. It is a device that outputs a signal.

  As shown in the figure, the radio wave sensor 2 includes a radio wave transmission unit 4, a radio wave reception unit 5, a Doppler signal generation unit 6, and a Doppler signal output unit 7. The radio wave transmission unit 4 converts a transmission signal output from the built-in transmitter into a radio wave, and transmits (radiates) the radio wave as a transmission wave to the human body. The radio wave receiver 5 receives a reflected wave from the human body with respect to the transmission wave transmitted from the radio wave transmitter 4 and converts it into a received signal. Based on the transmission signal output from the transmitter in the radio wave transmission unit 4 and the reception signal output from the radio wave reception unit 5, the Doppler signal generation unit 6 is the frequency of the difference between the frequency of the transmission wave and the frequency of the reflection wave. To generate a Doppler signal. The Doppler signal generation unit 6 extracts a Doppler signal from the mixer that mixes the transmission signal output from the transmitter in the radio wave transmission unit 4 and the reception signal output from the radio wave reception unit 5, and the mixed signal output from the mixer. It has a low-pass filter and the like. The Doppler signal output unit 7 outputs the Doppler signal generated by the Doppler signal generation unit 6 to the signal processing unit 3.

  As shown in FIG. 2, the signal processing unit 3 includes a Doppler signal data generation unit 31, a peak enhancement signal data generation unit 32, a smoothed differential intensity signal data generation unit 33, and a pulsation time determination unit 34. And a heartbeat interval calculation unit 35. The Doppler signal data generation unit 31 samples the (analog) Doppler signal output from the Doppler signal output unit 7 (generated by the Doppler signal generation unit 6), and a signal corresponding to a predetermined time obtained by this sampling. Doppler signal data in which values are arranged in chronological order is generated. The peak enhancement signal data generation unit 32 generates peak enhancement signal data by applying a peak enhancement filter to the Doppler signal data generated by the Doppler signal data generation unit 31. The smoothed differential intensity signal data generation unit 33 generates smoothed differential intensity signal data by applying a smoothing filter to the absolute value of the peak emphasized signal data generated by the peak emphasized signal data generation unit 32. . The pulsation time determination unit 34 determines pulsation times of a plurality of heartbeats (of the human body) based on the smoothed differential intensity signal data generated by the smoothed differential intensity signal data generation unit 33. The heart beat interval calculation unit 35 calculates the heart beat interval time based on the time difference between two beat times adjacent in time among the beat times of the plurality of heart beats determined by the beat time determination unit 34. To do.

  Next, the heartbeat interval time calculation process in the heartbeat measuring device 1 will be described. The radio wave sensor 2 is disposed so as to radiate radio waves toward the chest of the human body during heart rate measurement. The human body vibrates minutely due to periodic pulsation of the heart, and the radio wave sensor 2 outputs a Doppler signal having a signal value corresponding to a change in distance due to the vibration.

FIG. 3 is a flowchart of a heartbeat interval time calculation process performed in the signal processing unit 3 at the time of heartbeat measurement, and FIG. 4 is a flowchart of a beat time candidate extraction process in FIG. The signal processing unit 3 (Doppler signal data generating unit 31 of) the Doppler signal outputted from the Doppler signal output section 7, and sampled at certain time intervals (sampling period) T _S, the predetermined time obtained by the sampling Doppler signal data in which signal values of minutes are arranged in time series order is generated. Specifically, the signal processing unit 3, as shown in FIG. 3, at every sampling period T _S, the current time (to be precise, the elapsed time from the start of measurement) Get the t (S1), the By sampling (acquiring) Doppler signal data corresponding to the current time t (S2) until a predetermined time has elapsed (NO in S3), Doppler signal data for a predetermined time (for example, 10 seconds). Obtain f _d (k) (k = 1, 2,..., N). Here, N is a data number corresponding to the latest Doppler signal value at the current time, and the relationship between time t and k at the time of data acquisition is expressed as k = [t / T _S ]. Here, the symbol [x] represents the maximum integer not exceeding x.

FIG. 5 shows an example of a waveform in which each data included in the Doppler signal data f _d (k) is smoothly connected. In FIG. 5, a sharp peak is regularly observed, which corresponds to the timing of the heartbeat. In the following, the heartbeat timing is referred to as the heartbeat time, and the time interval between the heartbeat times that are adjacent in time (adjacent) among the heartbeat times is the heartbeat interval time. Call it.

When the acquisition of the Doppler signal data f _d (k) for a predetermined time is completed, the signal processing unit 3 (the peak enhancement signal data generation unit 32 thereof) applies a peak enhancement filter to the Doppler signal data f _d (k). Then, peak enhancement signal data f _p (k) is calculated (S4). The calculation used for the peak enhancement filter is an operation that emphasizes a steep signal change while removing the influence of gentle increase / decrease. For example, the peak enhancement filter can be used as a signal between adjacent (adjacent) signal data in terms of time. A differential filter for calculating the difference can be used. Differentiating filters, such as those that simply calculate the difference between adjacent signal data, those that calculate the difference between signal data separated by a certain distance, those that calculate the difference between multiple neighboring signal data, etc. It is possible to use a filter set according to the above. If the filter is g (s) (s = 1, 2,..., M), the application of the filter is a convolution operation of length M between the filter and the original signal data. Therefore, the peak enhancement signal data f _p (k) obtained by applying the peak enhancement filter to the Doppler signal data f _d (k) is expressed by the following equation.

Here, as the peak enhancement filter, a filter that satisfies the following conditions is usually used. However, in these equations, the filter length M of the peak enhancement filter is an odd number.

For example, g (s) = {− 1, 2, −1}, g (s) = {− 1, 0, 2, 0, −1}, g (s) = {− 1, −1, 4, −1, −1} all function as a peak enhancement filter. FIG. 6 shows peak enhancement signal data f _p (k) obtained by applying a peak enhancement filter to the Doppler signal data f _d (k) shown in FIG. 5 (a waveform obtained by smoothly connecting the data included therein).

Next, the signal processing unit 3 (the smoothed differential intensity signal data generation unit 33) calculates a moving average value of absolute values of the peak enhancement signal data f _p (k) for a predetermined number of data, and this moving average Smoothed differential intensity signal data f _m (k), which is signal data in which values are arranged in time series, is obtained. That is, the moving average filter h (s) having the filter length L is set, and the smoothed differential intensity signal data f _m (k) is calculated as in the following equation (S5).

Here, h (s) = {1 / L, 1 / L,..., 1 / L} is an example of the moving average filter. This example is an example of a filter that performs uniform averaging, but it is not necessarily required to be uniform averaging, and may be a Gaussian filter, for example. FIG. 7 shows the smoothed differential intensity signal data f _m (k) calculated from the peak enhancement signal data f _p (k) shown in FIG.

As shown in FIG. 7, the waveform of the smoothed differential intensity signal data f _m (k) is noise while enhancing a sharp peak with respect to the waveform of the original Doppler signal data f _d (k) shown in FIG. The peak resulting from is reduced. Therefore, by calculating the pulsation time of a plurality of heartbeats based on the smoothed differential intensity signal data f _m (k), the influence of noise compared to the case where the Doppler signal data f _d (k) is used as it is. It is possible to accurately estimate the heartbeat time with a small amount (stable).

Specifically, the signal processing unit 3 (the pulsation time determining unit 34) performs the heartbeat measurement according to the procedure shown in the flowchart of FIG. 4 based on the smoothed differential intensity signal data f _m (k) obtained in S5. A pulsation time is determined (a plurality of pulsation time candidates are extracted) (S6). First, for example, among the smoothed differential intensity signal data f _m (k) from 10 seconds before the current time t to the current time t, a specific time range (1 second in the example shown in the flowchart of FIG. 4, that is, , [1 / T _S ] data) from the smoothed differential intensity signal data f _m (k), after extracting all local maximum values, one of all local maximum values is selected. The time corresponding to the selected maximum value is set as the heartbeat pulsation time candidate. More specifically, as shown in the flowchart of FIG. 4, the signal processing unit 3 first sets (t−10) (10 seconds before the current time t) as an initial value in the index i. (S11) Calculate the maximum value among the local maximum values in the data of the smoothed differential intensity signal f _m (k) within the range after (i + 1) seconds after i seconds, and calculate the time corresponding to this maximum value. , in the range of from after the i seconds (i + 1) seconds, and beat time candidate _{T i} of the heartbeat (S12). If there is a heartbeat pulse time candidate T _i in S12 (when a local maximum value exists in the data of the smoothed differential intensity signal f _m (k) within a range of (i + 1) seconds after i seconds) ), the beat time candidate T _i of the heart, to add to beat time candidate set a heart rate (S13).

Next, the signal processing unit 3 partially overlaps the specific time range described above (for example, by shifting by 0.5 seconds, that is, by [0.5 / T _S ] data). Find beat time candidates. Specifically, after adding 0.5 to the index i, by performing the processing of S12 and S13, obtains the beat time candidate T _i for the next heartbeat. Then, the signal processing unit 3 repeats the processes of S12 to S14 until the index i becomes larger than (t-1) (NO in S15), thereby obtaining 10 seconds before the current time t (obtained in S1). To the current time t to extract a heartbeat pulse time candidate (determine a heartbeat pulse time candidate set A).

  The signal processing unit 3 (the heartbeat interval calculation unit 35) calculates a difference between two nearest beat times extracted from the beat time candidate extraction process (two beats adjacent in time). By calculating the difference in motion time), a numerical sequence of candidate heartbeat interval times is obtained. And the signal processing part 3 calculates | requires the average value or median value of the candidate of the heartbeat interval time contained in this numerical sequence as a heartbeat interval time representative value (S7). At this time, the signal processing unit 3 may calculate by removing the time that is clearly determined to be abnormal as the heartbeat interval time, for example, less than 0.5 seconds or 2 seconds or more. Furthermore, after removing a value that deviates from the representative value of the heartbeat interval time calculated as described above from a value included in the above-described numerical sequence of candidates for the heartbeat interval time, the heartbeat interval time representative is again obtained. By calculating the value, it can be made more robust against noise.

  The signal processing unit 3 (the heartbeat interval calculation unit 35) can output the heartbeat interval time representative value calculated as described above as a heartbeat interval in this time zone. For example, as shown in the flowchart of FIG. 4, by performing the above calculation using Doppler signal data for the past 10 seconds, one heartbeat interval value representing the past 10 seconds is obtained.

As described above, according to the heartbeat measuring device 1 of the present embodiment, the smoothed differential intensity is obtained by applying two types of simple filters, the peak enhancement filter and the moving average filter, to the Doppler signal data f _d (k). Signal data f _m (k) can be generated. The smoothed differential intensity signal data f _m (k) is signal data in which the peak due to noise is reduced while emphasizing the peak in the original Doppler signal data f _d (k). Therefore, by determining the beat times of a plurality of heartbeats based on the smoothed differential intensity signal data f _m (k) and calculating the heartbeat interval time based on the beat times of these heartbeats, the conventional frequency is obtained. Compared with a heartbeat measuring device using analysis, an accurate heartbeat interval time with less influence of noise can be obtained with a small amount of calculation. In addition, since the smoothed differential intensity signal data f _m (k) is signal data in which the peak in the original Doppler signal data f _d (k) is emphasized, the smoothed differential intensity signal data f _m (k). Unlike the heart rate measuring device that uses the waveform of the heart rate component separated by the conventional frequency analysis by determining the beat time of the heart rate based on the above, even when the heart rate interval time varies finely, Accurate heartbeat time and accurate heartbeat interval time can be obtained.

In the present embodiment, in the example shown in FIG. 3, the signal processing unit 3 collectively calculates Doppler signal data f _d (k) for a predetermined time (for example, 10 seconds), and then calculates the predetermined time. Calculation of peak enhancement signal data f _p (k) and smoothed differential intensity signal data f _m (k), extraction of pulsation time candidates, and representative value of heartbeat interval time based on Doppler signal data f _d (k) of Was calculated. However, every time the signal processing unit 3 samples (acquires) the Doppler signal data f _d (k) corresponding to a certain time t, the peak enhancement signal data f _p (k) and the smoothed differential intensity signal data f _m Calculation of (k), extraction of pulsation time candidates, and calculation of a representative value of heartbeat interval time may be performed.

  Next, the heartbeat measuring device 1 according to the second embodiment of the present invention will be described. The circuit configuration (hardware configuration) of the heartbeat measuring device 1 of the second embodiment is the same as that of the heartbeat measuring device 1 of the first embodiment. The difference between the heartbeat measuring device 1 and the heartbeat measuring device 1 of the first embodiment is mainly a method for determining the pulsation time of the heartbeat.

  With reference to the flowchart of FIG. 8, the heartbeat interval time calculation method in the heartbeat measuring apparatus 1 of the second embodiment will be described. The difference of the second embodiment from the first embodiment is mainly the method for determining the pulsation time of the heartbeat. Other points are basically the same as those in the first embodiment.

First, the signal processing unit 3 of the heartbeat measuring device 1 sets an initial value (“1”) to an index n of a heartbeat beat time (S21), and then S1, S2, S4 in FIG. A process similar to the process of S5 is performed. However, in the present embodiment, unlike the processing in the flowchart of FIG. 3, every time the signal processing unit 3 samples (acquires) Doppler signal data f _d (k) corresponding to a certain time t (S22 and S23). The peak enhancement signal data f _p (k) and the smoothed differential intensity signal data f _m (k) are calculated (S24 and S25).

Then, the signal processing unit 3 determines that the current time t is predetermined at the estimated time (t _R (n + 1) = t _R (n) + t _RRI (n)) of the next ((n + 1) th) heartbeat time. The smoothed differential intensity signal data f _m (k) is calculated until the time width _TP is added (NO in S26). Here, t _R (n) represents the pulsation time of the n-th heartbeat, and t _RRI (n) represents the n-th heartbeat interval time (the pulsation time t _R (n) of the n-th heartbeat). (N−1) represents the pulsation time t _R (n−1) of the second heartbeat.

Then, the signal processing unit 3, the current time t is equal to or time plus the predetermined time width _{T P} to the expected time of pulsation time of the next heartbeat _{(t R (n) + t} RRI (n) + T P) (YES in S26), the pulsation time of the next ((n + 1) th) heartbeat is estimated (S27).

Next, the estimation processing of the beat time of the heartbeat in S27 will be described. In general, heartbeats are not manipulatively manipulated, and if the person is in a healthy state, the previous heartbeat interval and the next heartbeat interval do not differ greatly. Therefore, it is possible to predict the pulsation time of the next heartbeat from the previous heartbeat interval time. Specifically, as described above, when the pulsation time of the n-th heartbeat is expressed as t _R (n), the n-th (n-th) heartbeat interval time is t _RRI (n) = t _R (N) −t _R (n−1). At this time, the pulsation time t _R (n + 1) of the next heartbeat can be expected to be close to t _R (n) + t _RRI (n). Therefore, the signal processing section 3, smoothing of setting a predetermined time width _{T P,} from time _{t R (n) + t RRI} (n) -T P, until time _{t R (n) + t RRI} (n) + T P of differential intensity data _f m with reference to (k), for example, data number argmax corresponding to the maximum value among the maximum value of the smoothed differential intensity signal data _f m (k) in that range _(f m (k)) From this, the beat time of the next heartbeat can be determined as t _R (n + 1) = argmax (f _m (k)) × T _S. Finally, the signal processing unit 3 calculates the heartbeat interval time by t _RRI (n + 1) = t _R (n + 1) −t _R (n) (S28), and the heartbeat beat time and the heartbeat interval time index are calculated. 1 is added to n (S29).

As described above, according to the heartbeat measuring device 1 of the second embodiment, the beat time t _R (n) of the determined previous heartbeat and the previous heartbeat interval time calculated by the heartbeat interval calculation unit 35. based on the t _RRI (n), the following heart beat time _t R (n + 1) takes may time ranging from (time _{t R (n) + t RRI} (n) -T P, time _t R (n _{) + t RRI (n) +} T to _P) to predict, based on the smoothed differential intensity signal data _f m (k) in the range of this time, to determine the beat time of the next heartbeat _t R (n + 1) I did it. This makes it possible to efficiently obtain (calculate) a stable heart beat time and a heart beat interval time with less noise influence for each beat. Further, as described above, the pulse time t _R (n + 1) of the next heartbeat is determined by referring only to the smoothed differential intensity data f _m (k) within a specific (predicted) time range. by the, beats time beat of the S27 estimation (determination) processing, smoothing data necessary for execution of the process corresponding to the aligned soon (time _{t R (n) + t RRI} (n) + T P The differential intensity signal data f _m (k) is acquired).

  Next, a heartbeat measuring device 1 according to a third embodiment of the present invention will be described. As shown in FIG. 9, the circuit configuration (hardware configuration) of the heartbeat measuring device 1 of the third embodiment is the same as that of the third embodiment except that it includes a respiration detection unit 41 that detects expiration and inspiration of the human body. Since it is the same as the heartbeat measuring device 1 of the first embodiment, the same reference numerals are given in FIG. 9 and the description thereof is omitted. As said respiration detection part 41, the well-known belt-shaped respiration sensor which can be attached to the chest and abdomen of a human body can be used, for example. This belt-shaped respiration sensor has, for example, a conductive fiber sensor, a piezoelectric film sensor, and the like, and detects a change in capacitance due to expansion and contraction of the belt according to the respiration of the subject, a change in pressure, etc. Detects exhalation and inspiration (respiration status).

  In addition, the heartbeat interval time calculation process in the heartbeat measuring device 1 of the present embodiment is based on the respiratory state detected by the respiratory detection unit 41 in the estimation process of the heartbeat beat time, and the beat time of the next heartbeat. This is different from the heartbeat measuring device 1 of the second embodiment in that the predicted time is adjusted. Since other processes are basically the same as those of the heart rate measuring apparatus 1 of the second embodiment, the same reference numerals are given in the flowchart of FIG. 10 and description thereof is omitted.

Next, the heartbeat interval time calculation process in the heartbeat measuring device 1 will be described in detail with reference to the flowchart of FIG. The signal processing unit 3 performs the same processing as the heartbeat measuring device 1 of the second embodiment from the acquisition of the current time t (S21) to the calculation of the smoothed differential intensity signal data f _m (k) (S25). After that, the adjustment processing of the predicted time of the pulsation time of the next heartbeat (the estimated pulsation time of the next heartbeat) is performed. Specifically, if the signal processing unit 3 determines that inspiration is being performed by the respiration detection unit 41 (YES in S31), the signal processing unit 3 is a time shorter by a predetermined time T _RS than the previous heartbeat interval time t _RRI (n). the _{(t RRI (n) -T RS} ), and sets the heartbeat interval for adjusting time work _{t RRIW,} from time _{t R (n) + t RRIW} -T P, time _t R _(n) of up to _{+ t} RRIW _{+ T P} With reference to the smoothed differential intensity data f _m (k), the pulsation time of the next heartbeat is estimated (S36). As a result, the range of time that can be taken by the pulse time t _R (n + 1) of the next heartbeat is shifted earlier by a predetermined time T _RS . On the other hand, when it is determined by the respiration detection unit 41 that the patient is exhaling (NO in S31, YES in S33), the signal processing unit 3 determines the predetermined time T from the previous heartbeat interval time t _RRI (n). _RS only long time _{(t RRI (n) + T} RS), and sets the heartbeat interval for adjusting time work _{t RRIW,} from time _{t R (n) + t RRIW} -T P, time _{t R (n) + t RRIW} + T _{With reference} to the smoothed differential intensity data f _m (k) up to _P , the pulsation time of the next heartbeat is estimated (S36). As a result, the range of time that can be taken by the pulse time t _R (n + 1) of the next heartbeat is shifted later by the predetermined time T _RS .

This is because even in healthy people, due to a phenomenon called sinus arrhythmia or respiratory arrhythmia, the heart rate increases during inhalation (beat interval time becomes shorter), and the heart rate decreases during exhalation (beat interval time). This adjustment is based on the tendency to be longer). Since there is an individual difference in the degree of the change, the value of the adjustment time _TRS can be automatically adjusted with reference to the past respiratory state and the heartbeat interval time. For example, the average value of the heartbeat interval time in the state after inspiration and before expiration and the average value of the heartbeat interval time in the state after expiration and before inspiration are calculated, and one half of the difference between these average values is adjusted time T may be set as the _RS.

The signal processing unit 3, smoothing the estimate (determine) the processing of the beat time of the heartbeat of the S36, the data required for execution of the process corresponding to the aligned soon (time _{t R (n) + t RRIW} + T P (At the timing when the converted differential intensity signal data f _m (k) is acquired) (YES in S35), the process is executed.

As described above, according to the heartbeat measuring device 1 of the third embodiment, when the signal processing unit 3 (the pulsation time determining unit 34) detects that the human body is exhaling by the respiration detecting unit 41, The range of possible time of the next heart beat time t _R (n + 1) is shifted later in time, and when the breath detecting unit 41 detects that the human body is inhaling, the next heart beat The range of time that the time t _R (n + 1) can take is shifted forward in time. Thereby, in addition to the effects in the second embodiment, the adjustment process of the time range that can be taken by the pulse time t _R (n + 1) of the next heart beat according to the fluctuation of the heart beat interval time due to breathing is performed. Can do. Therefore, since the estimation accuracy of the pulsation time of the next heartbeat can be improved, the heartbeat interval time can be calculated more accurately.

  Next, a heartbeat measuring device 1 according to a fourth embodiment of the present invention will be described. The circuit configuration (hardware configuration) of the heartbeat measuring device 1 of the fourth embodiment is the same as that of the heartbeat measuring device 1 of the first embodiment. The difference between the processing of the heartbeat measuring device 1 and the processing of the heartbeat measuring device 1 of the second embodiment is mainly the estimation method of the pulsation time of the heartbeat. Since the other processes are basically the same as those of the heart rate measuring apparatus 1 of the second embodiment shown in FIG. 8, in the flowchart of FIG. 11, the same processes as those in FIG. The description is omitted.

Next, a heartbeat interval time calculation process in the heartbeat measuring device 1 will be described with reference to the flowcharts of FIGS. 11 and 12. The heartbeat measuring device 1 of the present embodiment is the same as the heartbeat measuring device 1 of the second embodiment in that it predicts the pulsation time of the next heartbeat. However, in this embodiment, the beat time of the heart beat when the heart beat interval time closer to the same as the previous heart beat interval time is adopted instead of uniformly handling the range of the expected beat time of the next heart beat. This introduces the concept of probability distribution that the probability that the beat time of the next heartbeat is higher is higher. Specifically, the next heartbeat beat is obtained by Bayesian estimation using the probability distribution of the predicted beat time of the next heartbeat as a prior distribution and the smoothed differential intensity signal data f _m (k) as a likelihood function. Obtain the posterior distribution of time, and perform point estimation of the pulsation time of the next heartbeat from the posterior distribution.

More specifically, the signal processing unit 3 performs the same processing as S21 to S25 in FIG. 8 to calculate the smoothed differential intensity signal data f _m (k), and then beats the next heartbeat. When the data necessary for the estimation of time t _R (n + 1) is prepared (at the timing when the smoothed differential intensity signal data f _m (k) corresponding to time t _R (n) + t _RRI (n) + 3σ is acquired) ( (YES in S41), the pulsation time t _R (n + 1) of the next heartbeat is estimated (S42). However, (sigma) shows the standard deviation used as a parameter of normal distribution shown by S51 in FIG. 12 mentioned later.

As shown in the flowchart of FIG. 12, in the beat time estimation process of the heartbeat in S <b> 42, the signal processing unit 3 converts the probability distribution of the predicted beat time t _R (n + 1) of the next heartbeat into a prior distribution. Calculated as P _R (t) (S51). Here, as the prior distribution P _R (t) of t _R (n + 1), for example, as shown in the figure, t _R (n) + t _RRI (n) is an average value m, and a predetermined number of seconds (for example, A normal distribution having a standard deviation σ of about 0.1 to 0.2 seconds) can be used.

Next, the signal processing unit 3 uses the above prior distribution P _R (t) and the smoothed differential intensity signal data f _m (k) (= f _m ([t / T _S ])) as a likelihood function. By multiplying, the posterior distribution P _O (t) of the pulse time t _R (n + 1) of the next heartbeat is obtained (S52). This posterior distribution P _O (t) is not a probability distribution because it is not standardized, but it is only used for finally obtaining a point estimation value, and thus normalization is not necessary. The prior distribution P _R (t) is a continuous function, whereas the smoothed differential intensity signal data f _m (k) (= f _m ([t / T _S ])) is discrete in the time interval T _S. It is a function. However, the posterior distribution P _O (t) may be calculated by regarding f _m (t) as a function having an invariable value during the time interval T _S , as shown in S52 and the following equation. Then, the value of f _m (t) for an arbitrary t is obtained by interpolation or the like, and the posterior distribution P _O (t) is obtained by multiplying this f _m (t) and the prior distribution P _R (t). May be.

Next, the signal processing unit 3 obtains a point estimated value t _R (n + 1) of the heartbeat beat time as an expected value E [P _O (t)] of the posterior distribution P _O (t) (S53). Then, as shown in S28 of FIG. 11, the signal processing unit 3 calculates the pulse time of the next heartbeat (the estimated point value) t _R (n + 1) calculated in S53 and the pulse time of the previous heartbeat. Based on (point estimation value) t _R (n), the heartbeat interval time is calculated by t _RRI (n + 1) = t _R (n + 1) −t _R (n).

Also in the present embodiment, as in the second embodiment, the calculation of the pulsation time of the heartbeat is executed as soon as data necessary for this processing is available. The calculation of the posterior distribution P _O (t) can be defined (calculated) from t = 0 to t = + ∞. Here, for example, the value of t is m ± 3σ, that is, t _R (n) By determining to calculate the posterior distribution P _O (t) only within the range of + t _RRI (n) ± 3σ, the data range necessary for the calculation can be finite. In this case, the calculation of the pulsation time of the heartbeat is performed by using the smoothed differential intensity signal data f _m (k) corresponding to the time t _R (n) + t _RRI (n) + 3σ as shown in S41 in FIG. It is executed at the acquired timing.

As described above, according to the heartbeat measuring device 1 of the fourth embodiment, the signal processing unit 3 (the pulsation time determination unit 34) determines the pulsation time t _R (n) of the previous heartbeat that has been determined. Based on the previous heart beat interval time t _RRI (n) calculated by the heart beat interval calculating unit 35, the probability distribution of the pulse time t _R (n + 1) of the next heart beat is estimated. Then, the probability distribution is a prior distribution P _R (t), and the posterior distribution P _O (t of the next heartbeat is estimated by Bayesian estimation using the smoothed differential intensity signal data f _m (k) as a likelihood function. ) And the pulsation time (point estimate value) t _R (n + 1) of the next heartbeat is determined based on this posterior distribution. Thereby, it has the same effect as the heartbeat measuring device 1 of the second embodiment, and more accurately the beat time of the next heartbeat compared to the heartbeat measuring device 1 of the second embodiment. Can be predicted (estimated).

Next, a heartbeat measuring device 1 according to a fifth embodiment of the present invention will be described. The circuit configuration (hardware configuration) of the heartbeat measuring device 1 of the fifth embodiment is the same as that of the heartbeat measuring device 1 of the third embodiment shown in FIG. The method of calculating (generating) the smoothed differential intensity signal data f _m (k) and the method of estimating the pulsation time of the next heartbeat in the heartbeat measuring device 1 are the same as in the fourth embodiment. However, the heartbeat measuring device 1 of the present embodiment is different from the heartbeat measuring device 1 of the fourth embodiment in that the average value m of the prior distribution P _R (t) is adjusted according to the respiratory state.

Specifically, as shown in the flowchart of FIG. 13, when the signal processing unit 3 (the pulsation time determination unit 34) determines that the breathing detection unit 41 is inhaling (YES in S61), When the predetermined time T _RS for adjustment is used, m = t _R (n) + t _RRI (n) −T _RS is set (S62), and when it is determined by the breath detection unit 41 that exhalation is being performed (NO in S61, YES in S63), m = t _R (n) + t _RRI (n) + T _RS (S64). On the other hand, when it is determined by the breath detection unit 41 that neither inspiration nor inspiration is in progress (NO in S61, NO in S63), the signal processing unit 3 (the pulsation time determination unit 34) , _M = t _R (n) + t _RRI (n) (S65). Then, the signal processing unit 3 calculates the prior distribution P _R (t) using the adjusted average value m (S71 in FIG. 14), and based on the prior distribution P _R (t), the posterior distribution P _O. (T) is obtained (S52), and the pulsation time (point estimate value) t _R (n + 1) of the next heartbeat is calculated based on the posterior distribution P _O (t) (S53).

In the present embodiment, in the above example, the average value m of the prior distribution P _R (t) is adjusted according to the respiratory state. However, when exhaling and during inhaling, the prior value is compared with the state without breathing. it may be adjusted so as to widen the distribution width of the distribution P _{R (t).} Such an adjustment method can be realized by, for example, increasing the value of the standard deviation σ when using a normal distribution as the prior distribution P _R (t). Further, as described in the third embodiment, since the degree of change in heartbeat interval time during expiration and inspiration varies among individuals, the adjustment time _TRS may be changed according to the individual.

As described above, according to the heartbeat measuring device 1 of the fifth embodiment, when the signal processing unit 3 (the pulsation time determining unit 34) detects that the human body is exhaling by the respiration detecting unit 41, When the probability distribution of the pulsation time of the next heartbeat (the average value m of the prior distribution P _R (t)) is shifted later in time, when the respiration detecting unit 41 detects that the human body is inhaling, Shift the probability distribution of the heartbeat time of the previous time. Thereby, in addition to the effect in the fourth embodiment, the temporal distribution of the prior distribution P _R (t) of the pulsation time t _R (n + 1) of the next heart beat corresponding to the fluctuation of the heart beat interval time due to respiration. The position can be adjusted. Therefore, since the estimation accuracy of the pulsation time of the next heartbeat can be improved, the heartbeat interval time can be calculated more accurately.

Next, modified examples of the second to fifth embodiments will be described. In this modification, the initial value (t _R (1)) of the heartbeat time and the initial value (t _RRI (1)) of the heartbeat interval time in the second to fifth embodiments are appropriate. Perform initial processing to set a value.

In the heartbeat measuring devices 1 of the second to fifth embodiments, the initial values (t _R (1) and t _RRI (1)) of the heartbeat beat time and the heartbeat interval time are appropriate values (for example, heartbeats). The initial value t _R (1) of the pulsation time is the maximum time first appearing in the smoothed differential intensity signal data f _m (k), and the initial value t _RRI (1) of the heartbeat interval time is 1 second. Even if it is set to a value such as, the heartbeat time t _R (n) has a characteristic that it is gradually adjusted to the correct position. The tracking of the pulsation time can be performed. This is because even if the heartbeat prediction time at the beginning of measurement is greatly different from the actual time, the prediction has a width (time width T _P or 3σ), and the final (next) heartbeat beat This is because the motion time is determined by referring to the smoothed differential intensity signal data f _m (k) reflecting the actual peak position.

  However, in the heartbeat measuring devices 1 of the second to fifth embodiments, by performing the initial processing as shown in the flowchart of FIG. 15, the initial values of the heartbeat pulsation time and the heartbeat interval time are more appropriate values. Can be set to As a result, it is possible to shorten the time until stable measurement and realize a heart rate measuring apparatus 1 that is more convenient to use.

The flowchart of FIG. 15 shows the initial processing of the heartbeat measuring device 1 in this modification. In this figure, the same processes as those shown in FIG. 10 are denoted by the same reference numerals, and the description thereof is omitted. In this initial process, the signal processing unit 3 acquires the smoothed differential intensity signal data f _m (k) at a predetermined time interval T _i (for example, about 10 seconds) that includes a plurality of heartbeats. (S22 to S25). Then, (YES in S81) the signal processing unit 3, the acquisition of the smoothing differential intensity signal data _f m (k) of the time interval _{T i} is completed, the smoothed differential intensity signal data _f m (k time intervals _{T i} ) Is extracted (S82), and the time of the maximum value among these maximum values is set as the initial value t _R (1) of the heartbeat beat time (S83). That is, the signal processing section 3 (beats time determining unit 34), and out of the time a predetermined time interval T smoothing differential intensity signal data within _i f _{m (k)} is the maximum, the maximum value is the maximum Is the pulsation time t _R (1) of the first heartbeat. Thereby, it is possible to increase the probability that the initial value (beat time of the first heart beat) t _R (1) of the heart beat time corresponds to the true beat time.

The signal processing unit 3 (the heartbeat interval calculating unit 35 of) the relative peak enhancement signal of the time interval T _i min data f _{p (k)} or the smoothed differential intensity signal data f _{m (k),} high speed A power spectrum is calculated by performing Fourier transform (FFT) or the like (S84). In this power spectrum, the frequency at which the power is maximum is obtained from the frequencies (about 0.8 to 2 Hz) corresponding to the range in which the heartbeat interval time can be taken (about 0.5 to 1.2 seconds). The time calculated by the reciprocal is set to the initial value t _RRI (1) of the heartbeat interval time (S85). That is, the signal processing unit 3 (the heart rate interval calculation unit 35) performs frequency analysis on the peak enhancement signal data f _p (k) or the smoothed differential intensity signal data f _m (k) within a predetermined time interval T _i . Based on, the first heartbeat interval time t _RRI (1) is calculated. This makes it possible to estimate the initial value (first heartbeat interval time) t _RRI (1) of the heartbeat interval time that is close to the true heartbeat interval time.

At this time, using the initial value t _R (1) of the heartbeat beat time and the initial value t _RRI (1) of the heartbeat interval time, the smoothed differential intensity signal data f _m (k) obtained up to the current time t. If the estimation of the pulse time of the next heartbeat can be performed in accordance with the second to fifth embodiments within the range of), the estimation of the pulse time of the latest heartbeat is performed as far as possible within this range. It may be advanced (S86). At this time, the latest heartbeat interval time may be calculated as much as possible within the estimated heartbeat time range.

As described above, according to the heartbeat measuring device 1 of the present modification, the signal processing unit 3 (the pulsation time determining unit 34) performs the smoothed differential intensity signal data f _m (k) within the predetermined time interval T _i . ) Is a maximum time, and the time at which the maximum value is the maximum is the first heartbeat beat time (initial value of the heartbeat beat time) t _R (1), and the signal processing unit 3 ( The heartbeat interval calculation unit 35) of the first heartbeat frequency calculation unit 35) performs frequency analysis on the peak enhancement signal data f _p (k) or the smoothed differential intensity signal data f _m (k) within a predetermined time interval T _i . The heartbeat interval time (initial value of the heartbeat interval time) t _RRI (1) is calculated. As a result, the initial value of the heartbeat beat time (the beat time of the first heartbeat) t _R (1) and the initial value of the heartbeat interval time (the first heartbeat interval time) t _RRI (1) Since an appropriate value can be set, it is possible to improve the accuracy of predicting the pulsation time of the heartbeat in the initial stage of measurement, and to shorten the time until stable measurement.

  In addition, this invention is not restricted to the structure of the said embodiment, A various deformation | transformation is possible in the range which does not change the meaning of invention. For example, in each of the above-described embodiments, an example in which the measurement target by the heartbeat measuring device 1 is a human heartbeat (interval) has been shown, but the measurement target by the heartbeat measuring device is a heartbeat (interval) of an animal. May be.

  In the third and fifth embodiments, an example in which the respiration detection unit that detects the exhalation and inspiration of the human body is a so-called contact-type respiration detection device (belt-shaped respiration sensor) has been described. The respiration detection unit may be a so-called non-contact respiration detection device. Examples of the non-contact type respiratory detection device include those of the following respiratory detection system. That is, a reception signal having a frequency in the peripheral region of the frequency of the transmitted radio wave is extracted from the reception signal in the form of a digital signal based on the reception wave with respect to the transmission wave such as radio waves and ultrasonic waves. Multiply two signals (for example, sin ωt and cos ωt) orthogonal to each other (with a phase difference of 90 degrees). A signal obtained by removing the high frequency component from the signal obtained by multiplying the received signal by sin ωt is defined as an I (component) signal, and a signal obtained by removing the high frequency component from the signal obtained by multiplying the received signal by cos ωt is defined as a Q (component) signal. . Then, in a period of about 1 second, the amplitudes of these I signal and Q signal are constant. (1) If the phase of the I signal precedes that of the Q signal, it is determined as an intake state, and (2) If the phase of the Q signal precedes that of the I signal, it is determined as an expired state.

Furthermore, in the third embodiment, during exhalation and inhalation, the range of time that can be taken by the pulsation time t _R (n + 1) of the next heartbeat is set to a predetermined value as compared with a state where there is no breathing. You may adjust so that only time may spread.

  Further, in each of the above embodiments, an example in which the Doppler signal generation unit 6 and the signal processing unit 3 of the radio wave sensor 2 are separate circuits has been shown, but the Doppler signal generation unit and the signal processing unit are You may comprise with one microcomputer.

DESCRIPTION OF SYMBOLS 1 Heartbeat measuring device 3 Signal processing part 4 Radio wave transmission part 5 Radio wave reception part 6 Doppler signal generation part 31 Doppler signal data generation part 32 Peak emphasis signal data generation part 33 Smoothing differential strength signal data generation part 34 Beat time determination part 35 Heart rate interval calculation unit 41 Respiration detection unit

Claims (6)

A radio wave transmitter that transmits radio waves as transmission waves to a living body;
A radio wave receiving unit that receives a reflected wave from the living body with respect to the transmission wave;
A Doppler signal generator that generates a Doppler signal that is a signal having a frequency difference between the frequency of the transmitted wave and the frequency of the reflected wave;
In a heartbeat measuring device comprising a signal processing unit that converts the Doppler signal generated by the Doppler signal generation unit into a digital signal and performs digital signal processing,
The signal processing unit
Sampling a Doppler signal generated by the Doppler signal generator, and generating a Doppler signal data in which signal values for a predetermined time obtained by the sampling are arranged in chronological order; and
A peak enhancement signal data generation unit that generates peak enhancement signal data by applying a peak enhancement filter to the Doppler signal data generated by the Doppler signal data generation unit;
By applying a smoothing filter to the absolute value of the peak enhancement signal data generated by the peak enhancement signal data generation unit, a smoothed differential intensity signal data generation unit that generates smoothed differential intensity signal data;
Based on the smoothed differential intensity signal data generated by the smoothed differential intensity signal data generation unit, a pulsation time determination unit that determines pulsation times of a plurality of heartbeats of the living body,
A heartbeat interval calculating unit that calculates a heartbeat interval time based on a time difference between two beat times adjacent in time among a plurality of heartbeat beat times determined by the beat time determining unit; A heart rate measuring apparatus comprising:   The pulsation time determination unit is based on the determined previous pulsation time of the heartbeat and the previous pulsation time calculated by the HR calculation unit. The heart rate measuring apparatus according to claim 1, wherein a beat time of the next heart beat is determined based on the smoothed differential intensity signal data within the time range. Further comprising a respiration detector for detecting exhalation and inspiration of the living body,
When the respiration detecting unit detects that the living body is exhaling, the pulsation time determining unit shifts a time range that can be taken by the pulsation time of the next heartbeat later in time, and 3. The heartbeat according to claim 2, wherein when the detection unit detects that the living body is inhaling, the range of time that can be taken by the pulsation time of the next heartbeat is shifted forward in time. Measuring device.   The pulsation time determination unit calculates a probability distribution of the pulsation time of the next heartbeat based on the determined pulsation time of the previous heartbeat and the previous heartbeat interval time calculated by the heartbeat interval calculation unit. After the estimation, the probability distribution is set as a prior distribution, and a posterior distribution of the pulse time of the next heartbeat is obtained by Bayesian estimation using the smoothed differential intensity signal data as a likelihood function. The heartbeat measuring apparatus according to claim 1, wherein a beat time of the next heartbeat is determined. Further comprising a respiration detector for detecting exhalation and inspiration of the living body,
When the respiration detecting unit detects that the living body is exhaling, the pulsation time determining unit shifts the probability distribution of the pulsating time of the next heartbeat later in time, and the respiration detecting unit 5. The heartbeat measuring device according to claim 4, wherein when it is detected that the living body is inhaling, the probability distribution of the pulsation time of the next heartbeat is shifted forward in time. The pulsation time determination unit, as the pulsation time of the first heartbeat, the time when the maximum value of the smoothed differential intensity signal data is maximized within a predetermined time range,
The heartbeat interval calculation unit calculates a first heartbeat interval time based on a result of frequency analysis of the peak enhancement signal data or the smoothed differential intensity signal data within the predetermined time range. The heart rate measuring device according to any one of claims 2 to 5.

Images