JP2006055406A - Heartbeat detecting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a heartbeat detecting device capable of more accurately detecting heartbeat. <P>SOLUTION: This heartbeat detecting device has a doppler sensor 1, an input section 21 for inputting a signal output from the doppler sensor 1, a band pass filter 22 eliminating a predetermined frequency band of the signal inputted by the input section 21, a difference value calculation section 23 for calculating the difference value of the signal whose predetermined frequency band is eliminated by the band pass filter 22, a peak value extracting section 24 for extracting the peak value not less than a predetermined value of the difference value, and a heartbeat calculating section 25 for calculating the heartbeat from a time interval of the peak values adjacent each other. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a heartbeat detection device, and more particularly to a heartbeat detection device that performs heartbeat detection using a sensor that detects non-contact.

Conventionally, a heartbeat detection device detects a heartbeat cycle by FFT or the like based on the periodicity of a signal itself including respiration and heartbeat (see Patent Document 1 or 2).
JP-A-6-317676 JP 2003-275317 A

  However, for example, when trying to detect a heartbeat of a vehicle occupant or the like, an accurate heartbeat may not be detected due to noise such as vibration of the vehicle.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a heartbeat detecting device capable of detecting a more accurate heartbeat.

  The heartbeat detection device of the present invention includes a body surface motion detection sensor for detecting body surface motion, a sensor input means for inputting a signal output from the body surface motion detection sensor within a predetermined input time, and the sensor input means. A band-pass filter for removing a predetermined frequency band from the input signal, a difference value calculating means for calculating a difference value of the signal from which the predetermined frequency band has been removed by the band-pass filter, and a predetermined value among the difference values A peak value extracting unit that extracts a peak value that is equal to or greater than the value; and a peak value interval calculating unit that calculates a time interval between the adjacent peak values. The time interval calculated by the peak value interval calculating means corresponds to the heartbeat.

  According to the heartbeat detection device of the present invention, since the heartbeat is calculated based on the difference value of the signal from which the predetermined frequency band is removed by the bandpass filter, the heartbeat can be detected more accurately. By the way, in the electrocardiogram for detecting a heartbeat, an abruptly protruding R wave is extracted from the detected waveform. The extracted R wave interval (RRI) is extracted as a heartbeat. Here, the detected waveform (electrocardiographic waveform) appearing in the electrocardiogram is not a sine wave but a deformed waveform such as an R wave protruding.

  That is, according to the heartbeat detection device of the present invention, since the difference value is extracted, it is possible to reliably extract the portion of the heart motion derived from the rapidly protruding R wave. The ability to extract the part of the heart motion derived from the R wave means that the heartbeat can be detected more accurately. Therefore, according to the heartbeat detection device of the present invention, the heartbeat can be detected more accurately. Furthermore, the heartbeat time interval corresponding to the RRI of the electrocardiogram can be detected.

  As described above, the heartbeat detection device of the present invention includes a body surface motion detection sensor, a sensor input unit, a bandpass filter, a difference value calculation unit, a peak value extraction unit, and a peak value interval calculation unit. Prepare.

(1) Body surface motion detection sensor Here, as the body surface motion detection sensor, for example, a radio wave type Doppler sensor, an ultrasonic Doppler sensor, a piezoelectric sensor, a pneumatic sensor, or the like can be used. The body surface motion sensor may detect the motion of the surface of the human body, for example, with the back or buttocks in a state where a person is sitting on a chair. Note that the output signal of the body surface motion detection sensor generally includes breathing, body movement, etc. in addition to heartbeat. Further, the output signal includes vibrations caused by a vibration source around the person to be detected.

(2) Bandpass filter The bandpass filter generally leaves the upper limit of the heartbeat. Furthermore, the heartbeat is not a sine wave as described above. Furthermore, in the heartbeat detection device of the present invention, in order to be able to extract a part of the heart motion derived from the rapidly projecting R wave in the output signal of the sensor, a relatively large frequency band is left. Yes. For example, a frequency band other than the frequency band of 0.5 Hz to 15 Hz is removed by a band pass filter.

(3) Peak value extraction means The peak value extraction means may be as follows. The peak value extracting means is a maximum value extracting means for extracting a maximum value among the difference values, and an in-time maximum value extracting means for extracting a maximum value in time which is a maximum value within a predetermined extraction time from the maximum values. May be provided. In this case, the peak value interval calculation means calculates a time interval between adjacent maximum values in time.

  Thereby, a peak value corresponding to the heartbeat can be extracted. Here, the predetermined extraction time can be determined based on the maximum value of the heartbeat to be detected. For example, the period of the maximum value of the heartbeat to be detected is set as a predetermined extraction time. For example, when the maximum value of the heartbeat to be detected is 300 beats / minute, the predetermined extraction time is 0.2 seconds. Furthermore, the predetermined extraction time may be set to 0.2 seconds before and after each maximum value. That is, the in-time maximum value extracting means extracts only the maximum maximum value within 0.2 seconds before and after each maximum value.

  Further, the peak value extracting means extracts a local maximum value that extracts a local maximum value from the difference values, and an internal local maximum value that is the local maximum value belonging to a predetermined range based on the maximum local maximum value among the local maximum values. In-range maximum value extraction means for extraction may be provided. In this case, the peak value interval calculating means calculates a time interval between adjacent local maximum values.

  As a result, even when a local maximum value is generated due to noise or the like, only the local maximum value belonging to the predetermined range can be extracted, so that only the portion corresponding to the R wave can be extracted reliably.

  Further, the peak value extracting means extracts a maximum value for extracting a maximum value from the difference value, and a maximum value for time extraction means for extracting a maximum value in time that is a maximum value within a predetermined extraction time from the maximum value. And an in-range maximum value extracting means for extracting an in-range maximum value that is the in-time maximum value belonging to a predetermined range based on a maximum maximum value among the in-time maximum values. In this case, the peak value interval calculating means calculates a time interval between adjacent local maximum values. Thereby, the peak value corresponding to the heartbeat can be extracted more reliably. As a result, a more accurate heartbeat can be detected.

  The predetermined range may be as follows. For example, the predetermined range may be not less than a predetermined ratio of the maximum maximum value and not more than the maximum maximum value. Specifically, it is set to be not less than half of the maximum maximum value and not more than the maximum maximum value. The predetermined range may be not less than an average value of the maximum value or the maximum value in time and not more than the maximum maximum value. The predetermined range may be calculated based on a minimum maximum value and the maximum maximum value among the maximum value or the time maximum value. Specifically, it may be greater than or equal to an intermediate value between the maximum maximum value and the minimum maximum value and less than or equal to the maximum maximum value.

(4) Others Further, short-time peak value deletion means for deleting the peak value that becomes the time interval shorter than the maximum number of time intervals in which the frequency of the common time interval among the plurality of time intervals is the maximum. You may make it prepare. Thereby, RRI with higher accuracy can be calculated.

  The body surface motion detection sensor includes a plurality of sensors, and further includes a maximum number of time interval extraction means for extracting a maximum number of time intervals in which the frequency of the common time intervals is the maximum among the plurality of time intervals. May be. By performing various processes on the output signal from the sensor, a part corresponding to the R wave of the heartbeat may be removed or a part other than the R wave may remain. Even in such a case, it is possible to reliably detect the heartbeat by extracting the one having the maximum frequency of the time intervals of the common peak values.

  The peak value extracted by the peak value extracting means includes a maximum peak value where the maximum value of the difference value is equal to or greater than a first predetermined value and a minimum peak where the minimum value of the difference value is equal to or less than a second predetermined value. The peak value interval calculation means may calculate a time interval between the adjacent maximum peak values and a time interval between the adjacent minimum peak values. As described above, the waveform of the heartbeat has a sharply protruding R wave. The difference value calculated by the difference value calculating means of the heartbeat detecting device of the present invention is a very large value immediately before the portion corresponding to the R wave, and the falling portion immediately after the portion corresponding to the R wave. Is very small. That is, by using a difference value representing such a value, heartbeat detection can be performed using the maximum value and the minimum value. As a result, a large amount of data can be used, and as a result, a more accurate heartbeat can be detected.

  The heartbeat detecting device may detect the heartbeat of a vehicle occupant. By detecting the heartbeat of the occupant of the vehicle, the safety of the occupant and the comfort of the passenger compartment for the occupant can be improved.

(1) First Example Next, the present invention will be described more specifically with reference to a first example. The heartbeat detection device of the first embodiment is a device that detects the heartbeat of a vehicle occupant. That is, the heartbeat detecting device of the first embodiment is mounted on a vehicle.

(1.1) Description of Heart Rate First, before describing the heart rate detection device of the first embodiment, the heart rate will be described with reference to FIG. FIG. 1 shows an electrocardiogram waveform detected by an electrocardiogram. As shown in FIG. 1, a portion of the electrocardiogram that suddenly protrudes upward appears approximately every 1 second. The waveform of the portion that suddenly protrudes upward is called an R wave. The interval between adjacent R waves is called RRI. The heart rate corresponds to the average value of RRI. In addition, the protruding portion of the heartbeat suddenly appears on the lower side immediately before the portion of the heartbeat that protrudes upward.

(1.2) Configuration of Heartbeat Detection Device The configuration of the heartbeat detection device of the first embodiment will be described with reference to FIGS. FIG. 2 is a schematic diagram of a heartbeat detection device mounted on a vehicle. FIG. 3 shows a block diagram of the heart rate detection device. FIG. 4 shows signal waveforms generated by each unit.

  As shown in FIGS. 2 and 3, the heartbeat detection device includes a Doppler sensor 1 and an ECU 2. A Doppler sensor (body surface motion detection sensor) 1 is disposed at a backrest portion of a vehicle seat, and transmits a radio wave and receives the transmitted radio wave. And the Doppler sensor 1 can detect operation | movement of the body surface of the passenger | crew who seated on the vehicle seat. Specifically, the Doppler sensor 1 can detect the movement of the occupant's back surface. The movement of the back surface of the occupant includes vibration due to breathing, vibration due to body movement, and external vibrations such as a vehicle in addition to heartbeat.

  The ECU 2 inputs the detection signal detected by the Doppler sensor 1 and calculates the occupant's heart rate and RRI. As shown in FIG. 3, the ECU 2 includes an input unit 21, a bandpass filter 22, a difference value calculation unit 23, a peak value extraction unit 24, a heart rate calculation unit 25, and a short cycle peak value deletion unit 26. It is composed of

  The input unit (sensor input means) 21 inputs a detection signal output from the Doppler sensor 1. This input signal is shown in FIG. As shown in FIG. 4A, the input unit 21 inputs detection signals for 30 seconds (predetermined input time) among the detection signals output from the Doppler sensor 1. As described above, this signal includes heartbeat, respiration, body movement, and external vibration.

  The band pass filter 22 is a filter that removes frequency bands other than 0.5 Hz to 15 Hz. That is, the band pass filter 22 extracts only the frequency band of 0.5 Hz to 15 Hz from the signal input to the input unit 21. The signal waveform generated by the bandpass filter 22 is shown in FIG. In this way, low frequency components such as respiration and body movement of 0.5 Hz or less and high frequency components such as vibration by an engine of 15 Hz or more are removed. Here, the cycle of the heartbeat is 4.2 Hz when the measured maximum heartbeat is 250 beats / minute, but the bandpass filter 22 extracts up to 15 Hz. This is because the heartbeat waveform is set to 15 Hz so that the portion corresponding to the R wave can be reliably extracted instead of the sine wave as shown in FIG.

  The difference value calculation unit (difference value calculation means) 23 calculates a difference value signal based on the signal generated by the bandpass filter 22. The difference value is a difference between adjacent signal values among the signals generated by the band pass filter 22. The signal waveform generated by the difference value calculation unit 23 is shown in FIG. Here, in the difference value signal, a portion corresponding to the R wave portion of the heartbeat waveform appears as a large value.

  The peak value extraction unit (peak value extraction means) 24 extracts a peak value that is a difference value equal to or greater than a predetermined value based on the difference value signal generated by the difference value calculation unit 23. The peak value extraction unit 24 includes a first maximum value extraction unit 31, a second maximum value extraction unit 32, and a third maximum value extraction unit 33.

  The first maximum value extraction unit (maximum value extraction means) 31 extracts all the maximum values (first maximum values) from the difference value signal. This first maximum value is indicated by a circle in FIG. The second maximum value extraction unit (maximum value extraction means) 32 extracts the maximum value (maximum value within time) within the predetermined time T shown in Equation 1 among the first maximum values. Here, in the first embodiment, the maximum detected heart rate X is 300 [beats / minute]. That is, the predetermined time T is 0.2 seconds.

  Therefore, the second maximum value extraction unit 32 extracts the second maximum value that becomes the maximum value within 0.2 seconds from the extracted first maximum values. This second maximum value is indicated by a circle in FIG.

  The third maximum value extraction unit (in-range maximum value extraction means) 33 calculates the maximum value (third maximum value) included in the predetermined range among the second maximum values extracted by the second maximum value extraction unit 32. Extract. Here, the predetermined range is not less than one half of the maximum maximum value that is the maximum among the second maximum values, and is not more than the maximum maximum value. That is, in FIG. 4E, the maximum maximum value is a maximum value in the vicinity of about 10 seconds and is about 0.12V. Therefore, the predetermined range is a range of 0.06V to 0.12V. That is, a maximum value (third maximum value) included in 0.06 V to 0.12 V is extracted from the second maximum value. This third maximum value is indicated by a circle in FIG.

  First, the heart rate calculation unit (peak value interval calculation means) 25, based on the third maximum value extracted by the third maximum value extraction unit 33, the maximum value interval (peak value interval) that is an interval between adjacent third maximum values. Time interval). Further, the calculated maximum value interval is output to the short period peak value deletion unit 26 described later. Further, a corrected maximum value interval (peak value time interval) that is an interval between adjacent short cycle deletion signals is calculated based on a short cycle deletion signal generated by a short cycle peak value deletion unit 26 described later.

  The short cycle peak value deletion unit (short time peak value deletion unit) 26 performs processing for deleting the short cycle peak value based on the maximum value interval output from the heart rate calculation unit 25 to generate a short cycle deletion signal. Specifically, the short period peak value deletion unit 26 first generates a histogram for each common maximum value interval based on a plurality of maximum value intervals output from the heart rate calculation unit 25. Then, based on the generated hysteresis and gram, a third maximum value (short-period peak value) having a maximum value interval shorter than the maximum number of maximum value intervals is extracted. Then, the short period peak value is deleted from the third maximum value to generate a short period deletion signal. Then, the short cycle peak value deletion unit 26 outputs the generated short cycle deletion signal to the heart rate calculation unit 25 described above.

  Thus, the corrected maximum value interval generated by the heartbeat calculating unit 25 corresponds to RRI. The heart rate can be obtained based on the corrected maximum value interval. Thus, the heart rate and RRI of the vehicle occupant can be calculated.

(2) Second Example Next, a heartbeat detecting device of a second example will be described with reference to FIGS. FIG. 5 is a block diagram showing the heartbeat detecting device of the second embodiment. FIG. 6 shows a signal waveform input to the heart rate calculator 25. FIG. 7 is a histogram showing the frequency with respect to the maximum / minimum value interval. In the heartbeat detecting device of the second embodiment, the same reference numerals are given to the same components as those of the heartbeat detecting device of the first embodiment, and detailed description thereof is omitted.

  As shown in FIG. 5, the heartbeat detection device according to the second embodiment includes a plurality of Doppler sensors 1 a to 1 n and an ECU 3. Each of the Doppler sensors 1a to 1n is the same as the Doppler sensor 1 of the heart rate detecting device of the first embodiment, and a plurality of Doppler sensors 1a to 1n are arranged on the backrest portion and the seating surface portion of the vehicle seat.

  The ECU 3 inputs the detection signals detected by the Doppler sensors 1a to 1n, and calculates the occupant's heart rate. As shown in FIG. 3, the ECU 2 includes an input unit 21, a bandpass filter 22, a difference value calculation unit 23, a peak value extraction unit 27, and a heart rate calculation unit 28. Here, the input unit 21, the band pass filter 22, and the difference value calculation unit 23 are the same as those in the first embodiment.

  The peak value extraction unit (peak value extraction means) 27 extracts a peak value that is a difference value greater than or equal to a predetermined value or less based on the difference value signal generated by the difference value calculation unit 23. The peak value extraction unit 27 includes a first maximum / minimum value extraction unit 41, a second maximum / minimum value extraction unit 42, and a third maximum / minimum minimum value extraction unit 43.

  The first maximum / minimum value extraction unit 41 extracts all maximum values (first maximum value) and minimum values (first minimum value) from the difference value signal. The second maximum / minimum value extraction unit 42 extracts the maximum value within the predetermined time T shown in the above-described Expression 1 from the first maximum values. Further, the second maximum / minimum value extraction unit 42 extracts the minimum value within the predetermined time T shown in the above-described formula 1 from the first minimum values. Here, in the second embodiment, as in the first embodiment, the maximum detected heart rate X is 300 [beats / minute], so the predetermined time T is 0.2 seconds.

  The third maximum / minimum value extraction unit 43 extracts a maximum value (third maximum value) included in a predetermined range from the second maximum values extracted by the second maximum / minimum value extraction unit 42. Further, the third maximum / minimum value extraction unit 42 extracts a minimum value (third minimum value) included in the predetermined range from the second minimum values extracted by the second maximum / minimum value extraction unit 42. Here, the predetermined range is at least one half of the maximum maximum minimum value that is the maximum among the absolute values of the second maximum value and the second minimum value, and is equal to or less than the maximum minimum maximum value.

  Here, the waveform of the signal generated by the third maximum / minimum value extraction unit 43 is as shown in FIG. 6 for each of the Doppler sensors 1a to 1n. In FIG. 6, the third maximum value is indicated by a circle, and the third minimum value is indicated by a square mark.

  The heart rate calculation unit (peak value interval calculation means) 28 is based on the third maximum value for each Doppler sensor extracted by the third maximum value minimum extraction unit 43, and is a maximum value that is an interval between adjacent third maximum values. The interval is calculated for each Doppler sensor. Further, the heart rate calculation unit 28 sets the minimum value interval, which is the interval between the adjacent third minimum values, based on the third minimum value for each Doppler sensor extracted by the third maximum value minimum extraction unit 43. Calculate for each sensor.

  Furthermore, the heart rate calculation unit 28 generates a histogram for each common maximum value interval or minimum value interval (maximum minimum value interval) based on the calculated maximum value interval and minimum value interval. This histogram is shown in FIG.

  Then, the heart rate calculation unit 28 extracts a period in which the common maximum / minimum value interval is maximum. According to the histogram shown in FIG. 7, the maximum / minimum value interval is maximum at 1 second. Then, the heart rate is calculated based on the period of the maximum number of maximum / minimum value intervals extracted. In this case, since the heartbeat cycle is 1 second, the heart rate is 60 [beats / minute]. In this way, the heart rate of the vehicle occupant can be calculated.

(3) Others The predetermined range in the third maximum value extraction unit 33 and the third maximum value extraction unit 43 in the above embodiment is at least half of the maximum maximum value that is the maximum among the second maximum values. However, the maximum maximum value or less is set, but it is not limited to this. For example, the predetermined range may be not less than a predetermined ratio other than one half of the maximum maximum value and not more than the maximum maximum value. Furthermore, the predetermined range may be greater than or equal to the average value of the second maximum values and less than or equal to the maximum maximum value. Further, the predetermined range may be calculated based on the minimum maximum value and the maximum maximum value among the second maximum values. For example, it may be greater than or equal to an intermediate value between the maximum maximum value and the minimum maximum value but less than or equal to the maximum maximum value. Moreover, the numerical value in the said Example can be changed suitably.

The electrocardiogram waveform detected by the electrocardiogram is shown. 1 is a schematic diagram of a heartbeat detection device according to a first embodiment. The block diagram of a heart-rate detection apparatus is shown. The signal waveform produced | generated by each part is shown. It is a block diagram which shows the heart rate detection apparatus of 2nd Example. The signal waveform input into the heart rate calculation part 25 is shown. It is a histogram which shows the frequency with respect to the local maximum value interval.

Explanation of symbols

1: Doppler sensor 2, 3: ECU, 21: input unit (sensor input means), 22: band-pass filter, 23: difference value calculation unit (difference value calculation means), 24, 27: peak value extraction unit (peak) 25, 28: heart rate calculation unit (peak value interval calculation unit), 26: short period peak value deletion unit (short-time peak value deletion unit), 31: first maximum value extraction unit (maximum value extraction unit) 32: Second maximum value extraction unit (maximum value extraction means in time) 33: Third maximum value extraction unit (in-range maximum value extraction means) 41: First maximum minimum value extraction unit 42: Second maximum maximum minimum Value extraction unit, 43: third maximum / minimum value extraction unit

Claims (11)

A body surface motion detection sensor for detecting body surface motion;
Sensor input means for inputting a signal output from the body surface motion detection sensor within a predetermined input time;
A band-pass filter for removing a predetermined frequency band from the signal input by the sensor input means;
Difference value calculating means for calculating a difference value of the signal from which a predetermined frequency band has been removed by the bandpass filter;
Peak value extraction means for extracting a peak value equal to or greater than a predetermined value among the difference values;
A peak value interval calculating means for calculating a time interval between adjacent peak values;
A heartbeat detecting device comprising: The peak value extracting means includes
A maximum value extracting means for extracting a maximum value among the difference values;
In-time maximum value extraction means for extracting the maximum value in time that is the maximum value within a predetermined extraction time among the maximum values,
With
2. The heartbeat detecting device according to claim 1, wherein the peak value interval calculating means calculates a time interval between adjacent maximum values in time. The peak value extracting means includes
A maximum value extracting means for extracting a maximum value among the difference values;
In-range maximum value extracting means for extracting an in-range maximum value that is the maximum value belonging to a predetermined range based on a maximum maximum value among the maximum values;
With
2. The heartbeat detecting device according to claim 1, wherein the peak value interval calculating means calculates a time interval between adjacent maximum values in the range. The peak value extracting means includes
A maximum value for extracting a maximum value among the difference values; and
In-time maximum value extraction means for extracting the maximum value in time that is the maximum value within a predetermined extraction time among the maximum values,
A local maximum value extracting means for extracting a local maximum value in the range that is the local maximum value belonging to a predetermined range based on a maximum local maximum value among the local maximum values;
With
2. The heartbeat detecting device according to claim 1, wherein the peak value interval calculating means calculates a time interval between adjacent maximum values in the range.   The heart rate detection device according to claim 3 or 4, wherein the predetermined range is not less than a predetermined ratio of the maximum maximum value and not more than the maximum maximum value.   5. The heartbeat detection device according to claim 3, wherein the predetermined range is equal to or greater than an average value of the maximum value or the maximum value in time and equal to or less than the maximum maximum value.   The heart rate detection device according to claim 3 or 4, wherein the predetermined range is calculated based on a minimum maximum value and the maximum maximum value among the maximum value or the time maximum value.   Furthermore, short-time peak value deletion means for deleting the peak value that is the time interval shorter than the maximum number of time intervals in which the frequency of the common time intervals among the plurality of time intervals is maximum is provided. The heartbeat detection device according to any one of claims 1 to 7. The body surface motion detection sensors are plural,
Furthermore, it comprises the maximum number time interval extraction means which extracts the maximum number time interval whose frequency of the said time interval common among the said some time intervals is the maximum. The heartbeat detection device described in 1. The peak value extracted by the peak value extraction means is a maximum peak value where the maximum value of the difference value is equal to or greater than a first predetermined value, and a minimum peak value where the minimum value of the difference value is equal to or less than a second predetermined value. Yes,
The heart rate detection device according to claim 1 or 8, wherein the peak value interval calculation means calculates a time interval between the adjacent maximum peak values and a time interval between the adjacent minimum peak values.   The heartbeat detection device according to any one of claims 1 to 10, wherein the heartbeat detection device detects a heartbeat of a vehicle occupant.

Images