1. Technical Field
The present invention relates to a technology which measures the pulse wave of a living body.
2. Related Art
As a method of detecting a pulse wave of a living body, in particular, a human body, a method of measuring a pulse wave by performing photoelectric conversion (a photoelectric method) has been used. In this method, light of a wave length which is easily absorbed by blood is emitted from a light emitting device such as a light emitting diode. After the light passes through a living body or enters the living body, light which is dispersed due to tissue in the living body is received by a light receiving device, such as a photodiode or a phototransistor. Thereafter, a pulse wave is detected by converting the received light into an electrical signal. Although arteries are repeatedly dilated and constricted at the same period as that of a heartbeat, the degree of absorption of light which enters a living body is higher in a case in which the artery is dilated compared to a case in which the artery is constricted. Therefore, the intensity of light which is received by the light receiving device is changed according to pulsation. That is, the light absorption amount emitted from the light emitting device increases or decreases at a period which is the same as the period at which a blood vessel is dilated and constricted, and the intensity of reflected light is changed according to the increase and decrease. The pulse wave is measured based on the change.
However, when the pulse wave of a living body is measured, there are cases in which noise (hereinafter, referred to as “body motion noise”) is generated in the results of the measurement due to the movement of the living body (hereinafter, referred to as “body motion”). The body motion noise is generated because the pressurization or decompression state of the measurement position and the positional relationship between the light emitting device/the light receiving device and the living body are changed according to the body motion and affect the direction or amount of received light. As a technology which is used to reduce body motion noise, a pulse wave sensor which has a configuration in order to avoid a congestion state is disclosed in JP-A-2002-224064.
Blood vessels which are detection targets are present in the dermis of the skin of the living body. From a surface, capillary vessels and subpapillary vascular plexuses are formed, subcutaneous vascular plexuses are formed further below, and there is a tendency for blood vessels in deep portions to be thick. When a pulse wave is measured using the photoelectric method, the sensitivity of body motion noise differs between cases in which thick blood vessels are present under a sensor and cases in which thick blood vessels are not present under a sensor. When measurement is performed at a position at which thick blood vessels are present under a sensor, the influence of body motion noise increases. In a pressure pulse wave sensor disclosed in JP-A-2002-224064, a congestion state can be avoided but the influence of body motion noise is greatly received according to the position of a sensor.
An advantage of some aspects of the invention is to reduce the influence of noise which is generated due to the operation of a measurement position in a pulse wave measurement apparatus.
An aspect of the invention is directed to a pulse wave measurement apparatus including: a plurality of light receiving units that output measurement signals each indicative of a received amount of light which is irradiated to a measurement position where pulse waves are measured, and passes through or is reflected from the measurement position; an independent component analysis unit that performs an independent component analysis on the measurement signals which are output from the plurality of light receiving units, divides each of the measurement signals into a plurality of components, and calculates weighting factors of each of the plurality of components; a dispersion calculation unit that calculates a value indicative of a degree of dispersion of values of the weighting factors which are calculated by the independent component analysis unit for each of the plurality of components; a component specification unit that specifies a component of the plurality of components whose value indicative of the degree of dispersion calculated by the dispersion calculation unit satisfies a predetermined condition; and a pulse wave information generation unit that generates pulse wave information indicative of the pulse waves using the component specified by the component specification unit. According to this aspect, it is possible to reduce the influence of noise which is generated due to the operation of the measurement position in the pulse wave measurement apparatus.
The pulse wave measurement apparatus according to the aspect of the invention may be configured such that the pulse wave information generation unit extracts the component specified by the component specification unit by performing an operation using the weighting factors calculated by the independent component analysis unit on the measurement signals which are output from the plurality of light receiving units, and generates the pulse wave information indicative of the pulse waves based on the extracted component. According to this configuration, it is possible to reduce the influence of noise which is generated due to the operation of the measurement position in the pulse wave measurement apparatus.
The pulse wave measurement apparatus according to the aspect of the invention may be configured such that the component specification unit specifies a component whose value indicative of the degree of dispersion calculated by the dispersion calculation unit is smallest. According to this configuration, it is possible to reduce the influence of noise which is generated due to the operation of the measurement position in the pulse wave measurement apparatus.
The pulse wave measurement apparatus according to the aspect of the invention may be configured such that the component specification unit specifies a component whose value indicative of the degree of dispersion calculated by the dispersion calculation unit is within a predetermined range. According to this configuration, it is possible to reduce the influence of noise which is generated due to the operation of the measurement position in the pulse wave measurement apparatus.
The pulse wave measurement apparatus according to the aspect of the invention may be configured such that, when the plurality of components are specified by the component specification unit, the pulse wave information generation unit generates the plurality of specified components, and generates the pulse wave information based on the sum of the plurality of generated components. According to this configuration, it is possible to reduce the influence of noise which is generated due to the operation of the measurement position in the pulse wave measurement apparatus.
The pulse wave measurement apparatus according to the aspect of the invention may be configured such that the pulse wave measurement apparatus further includes a measurement signal specification unit that specifies a measurement signal whose value of the weighting factor of the component specified by the component specification unit satisfies the predetermined condition from among the measurement signals output from the plurality of light receiving units and the pulse wave information generation unit generates the pulse wave information indicative of the pulse waves based on the measurement signal specified by the measurement signal specification unit. According to this configuration, it is possible to reduce the influence of noise which is generated due to the operation of the measurement position in the pulse wave measurement apparatus.
The pulse wave measurement apparatus according to the aspect of the invention may be configured such that the component specification unit specifies a component whose value indicative of the degree of dispersion calculated by the dispersion calculation unit is greatest. According to this configuration, it is possible to reduce the influence of noise which is generated due to the operation of the measurement position in the pulse wave measurement apparatus.
The pulse wave measurement apparatus according to the aspect of the invention may be configured such that the measurement signal specification unit specifies a measurement signal whose value of the weighting factor of the component specified by the component specification unit is smallest. According to this aspect, it is possible to reduce the influence of noise which is generated due to the operation of the measurement position in the pulse wave measurement apparatus.
The pulse wave measurement apparatus according to the aspect of the invention may be configured such that the pulse wave information generation unit removes the component specified by the component specification unit from the measurement signal specified by the measurement signal specification unit, and generates the pulse wave information based on the measurement signal from which the component is removed. According to this configuration, it is possible to reduce the influence of noise which is generated due to the operation of the measurement position in the pulse wave measurement apparatus.
Another aspect of the invention is directed to a program causing a computer to perform: receiving measurement signals from a plurality of light receiving units that output the measurement signals each indicative of a received amount of light which is irradiated to a measurement position where pulse waves are measured, and passes through or is reflected from the measurement positions; performing an independent component analysis on the measurement signals which are output from the plurality of light receiving units, dividing each of the measurement signals into a plurality of components, and calculating weighting factors of each of the plurality of components; calculating a dispersion value indicative of a degree of dispersion of values of the weighting factors which are calculated in the independent component analysis for each of the plurality of components; specifying a component of the plurality of components whose value indicative of the degree of dispersion calculated in the calculating of the dispersion value satisfies a predetermined condition; and generating pulse wave information indicative of the pulse waves using the component specified in the specifying of the component. According to this aspect, it is possible to reduce the influence of noise which is generated due to the operation of the measurement position in the pulse wave measurement apparatus.
The invention will be described with reference to the accompanying drawings, wherein like numbers reference like elements.
As shown in
In
A signal processing unit 120 includes signal processing units 121, 122, 123, and 124. The signal processing unit 121 includes an amplifier (not shown) which acquires the measurement signal G1 output from the light receiving device 211 and amplifies the measurement signal G1, and an A/D conversion circuit (not shown) which quantizes the amplified measurement signal G1 using a predetermined sampling frequency. The signal processing unit 122 includes an amplifier (not shown) which acquires the measurement signal G2 output from the light receiving device 212 and amplifies the measurement signal G2, and an A/D conversion circuit (not shown) which quantizes the amplified measurement signal G2 using a predetermined sampling frequency. The signal processing unit 123 includes an amplifier (not shown) which acquires the measurement signal G3 output from the light receiving device 213 and amplifies the measurement signal G3, and an A/D conversion circuit (not shown) which quantizes the amplified measurement signal G3 using a predetermined sampling frequency. The signal processing unit 124 includes an amplifier (not shown) which acquires the measurement signal G4 output from the light receiving device 214 and amplifies the measurement signal G4, and an A/D conversion circuit (not shown) which quantizes the amplified measurement signal G4 using a predetermined sampling frequency. A timepiece unit 130 counts the time keeping clock signal of a clock supply unit 140 and times time. The clock supply unit 140 includes an oscillation circuit and a clock division circuit, supplies a reference clock signal to the control unit 110 by the oscillation circuit, and generates a time keeping clock signal for timing and supplies the generated time keeping clock signal to the control unit 110 by the clock division circuit. A display unit 150 includes a display 15, and displays various types of images, such as time information timed by the timepiece unit 130, a menu screen used to measure a pulse wave, and results of measurement, under the control of the control unit 110. An operation unit 160 includes an operational switch 16, and transmits an operation signal acquired through operations performed by the operational switch 16 to the control unit 110. A storage unit 180 includes a flag storage region 181 which stores a flag which is referred to when a pulse wave information generation process which will be described later is performed.
The independent component analysis unit 111 performs an independent component analysis (ICA) on a measurement signal Gi (1≦i≦n; n is an integral number which is equal to or greater than 2) which is output from each of a plurality of light receiving devices, and acquires the weighting factor wij of each component in a case in which the measurement signal Gi is divided into a plurality of components Sj (1≦j≦m; where n≧m, m is an integral number which is equal to or greater than 2). The plurality of components indicate a signal which is output from each signal source in a case in which it is assumed that a plurality of signal sources are present in a region in a body which is a measurement target. Here, each of the signal sources generates a signal based on the movement of a blood vessel which is in a measurement target region, and it is possible to specify a signal source which locates at a position where it is difficult to receive the body movement based on analysis using Equations which will be described later.
In this embodiment, the independent component analysis unit 111 performs the independent component analysis on the measurement signals G1, G2, G3, and G4 which are output from the respective light receiving devices 211, 212, 213, and 214, and acquires the weighting factors w11, w12, w44 of each of the components in a case in which each of the measurement signals G1, G2, G3, G4 is divided into a plurality of components S1, S2, S3, and S4. In the first embodiment, as described below, in a case in which it is assumed that an array G is Equation 1, an array S is Equation 2, and an array W is Equation 3, the independent component analysis unit 111 performs the independent component analysis and estimates four independent components S1, S2, S3, and S4 which cause the arrays G, S, and W to satisfy Equation 4. That is, the independent component analysis unit 111 calculates the weighting factors w11, w12, . . . w44 using any one of a method based on non-Gaussian maximization, a method based on maximum likelihood estimation, a method based on mutual information amount, a method based on uncorrelated nonlinear functions, and a method based on tensors such that the components S1, S2, S3, and S4 are statistically independent from each other. The independent component analysis performed by the independent component analysis unit 111 uses, for example, a method written by Noboru Murata, “introduction to independent component analysis”, first edition, Tokyo Denki University Press, July 2004”.
that is,
A dispersion calculation unit 112 calculates the dispersion σj (the degree of dispersion) of the values of the weighting factors wij, acquired by the independent component analysis unit 111, for each component Sj. In the first embodiment, the dispersion σj indicates the degree of the dispersion of values using a numerical value. For example, the dispersion when the weighting factors include 2, 3.5, 1, and 5 may be less than the dispersion when the weighting factors include 1, 13, −2, and 50. Here, the dispersion σj is defined as a standard deviation. Therefore, the dispersion σj shows that the greater the degree of dispersion is the greater the value of the dispersion is and that the lower the value is the lower the degree of dispersion is. The dispersion calculation unit 112 calculates the dispersion σj of the weighting factor wij corresponding to the component Sj (that is, the weighting factor wij included in the j-th column of the array W). That is, the dispersion calculation unit 112 calculates the dispersion of the weighting factor wij included in the array W for each column. More specifically, in the first embodiment, dispersion σ1 indicates the degree of dispersion of the weighting factors w11, w21, w31, and w41, and dispersion σ2 indicates the degree of dispersion of the weighting factors w12, w22, w32, and w42. Dispersion σ3 indicates the degree of dispersion of the weighting factors w13, w23, w33, and w43, and dispersion σ4 indicates the degree of dispersion of the weighting factors w14, w24, w34, and w44. The embodiment of the calculation of the degree of dispersion of the weighting factors performed by the dispersion calculation unit 112 is not limited to the calculation of the standard deviation for each component, and any calculation in which the degree of dispersion of weighting factors is calculated may be used.
A component specification unit 113 specifies one or more components which satisfy a condition in which the dispersion σj calculated by the dispersion calculation unit 112 is previously determined. In the first embodiment, the component specification unit 113 calculates the calculated dispersion σj for each of the components S1, S2, S3, and S4, and specifies a component in which the dispersion σj of the weighting factors is the smallest. The component specification unit 113 stores information (flag) indicative of the specified component in a flag storage region 181. The flag is referred to by the pulse wave information generation unit 114.
A pulse wave information generation unit 114 generates pulse wave information indicative of a pulse wave based on the measurement signals output from the signal processing unit 120. In the first embodiment, the pulse wave information generation unit 114 extracts a component (hereinafter, referred to as “specified component”) which is specified by the component specification unit 113 by performing an operation on the measurement signals G1, G2, G3, and G4 output from the signal processing unit 120 using the weighting factor wij. Since the measurement signals G1, G2, G3, and G4 and the components S1, S2, S3, and S4 satisfy the above-described Equation 4, the component Sj is calculated using the following Equation 5. The pulse wave information generation unit 114 extracts the specified component using the following Equation 5.
As described above, there are cases in which the component of body motion noise (hereinafter, referred to as “noise component”) is mixed with the measurement signal Gi according to the position of the light receiving device due to the influence of the thick blood vessels in the deep portions of the dermis 22 or the subcutaneous tissue 23. The magnitude of the influence of the noise component varies according to the position of the light receiving device, and the magnitude of the influence greatly varies even when the position of the light receiving device is slightly deviated. More specifically, in the example shown in
The pulse wave information generation unit 114 generates pulse wave information indicative of a pulse wave based on the calculated specified component. In the first embodiment, the pulse wave information generation unit 114 assumes that a time interval between the peaks of the wave form of the calculated specified component is a pulse period and that the appearance frequency of peaks of the wave form of the measurement signal during a predetermined time (1 minute or the like) is a pulse rate, and outputs information (pulse wave information) indicative of the pulse period and the pulse rate on the display unit 150. In addition, the pulse wave information generation unit 114 may, for example, chronologically store the information indicative of the pulse period and the pulse rate in the storage unit 180.
When the output of the measurement signals starts, the control unit 110 performs the independent component analysis on the measurement signals G1, G2, G3, and G4, which are output from the respective light receiving devices 211, 212, 213, and 214 and on which the A/D conversion is performed by the signal processing unit 120, in step S12, divides each of the measurement signals G1, G2, G3, and G4 into the plurality of components S1, S2, S3, and S4, and acquires weighting factors w11, w12, . . . , and w44 which are related to each of the components.
Subsequently, the control unit 110 calculates the dispersion σi of the acquired weighting factors for each of the components S1, S2, S3, and S4, specifies a component, in which the dispersion is the smallest, in step S13, and stores information indicative of the specified component (specific character component) in the flag storage region 181.
Once the processes up to step S13 have been completed, the control unit 110 generates pulse wave information based on the component specified in step S13, and displays an image indicative of the pulse wave information on the display unit 150. More specifically, the control unit 110 calculates the specified component using the measurement signals G1, G2, G3, and G4 which are output from the signal processing unit 120 in step S14, detects the time interval between peaks of the waveform of the calculated specified component as a pulse period and the appearance frequency of the peaks of the waveform of the measurement signal during a predetermined time as a pulse rate, and displays an image indicative of the detected pulse period and pulse rate (pulse wave information) on the display unit 150. Meanwhile, if the control unit 110 does not receive the operation of measuring a pulse wave through the operation unit 160 (step S10; NO), the control unit 110 stands by until the operation is performed.
The control unit 110 repeatedly performs the process in step S14 until an operation of terminating the measurement of the pulse wave is performed through the operation unit 160 (step S15; NO). The control unit 110 terminates the process when the operation of terminating the measurement of the pulse wave is performed through the operation unit 160 (step S15; YES).
Due to the difference in density of the blood vessels of a living body and the influence of the distribution of blood vessels having different thicknesses, the level of each of the collector electrical potentials varies according to the measurement position of the living body as shown in
In the first embodiment, among the components S1, S2, S3, and S4 which are estimated by performing the independent component analysis, a component, in which the dispersion of the weighting factors is the smallest, is used as the component of a signal used to measure a pulse wave. Since the component in which dispersion of the weighting factors is the smallest is a component in which the influence of noise is the smallest, the measurement of a pulse wave, in which the influence of noise is suppressed, is performed by using such a signal.
Subsequently, a second embodiment according to the invention will be described. The second embodiment and the above-described first embodiment differ in that the processes performed by the control unit 110 are different, and the other processes are the same as those in the above-described first embodiment. Here, description will be made below while focusing on the differences between the second embodiment and the first embodiment. In addition, for the components of the second embodiment which are the same as the components of the first embodiment, the same reference numerals are used as in the first embodiment.
The component specification unit 116 specifies one or more components Sj in which dispersion σj calculated by the dispersion calculation unit 112 satisfies a predetermined condition. In the second embodiment, the component specification unit 116 specifies the component Sj in which the calculated dispersion σj is the greatest (that is, j is specified).
As described above, according to the position of the light receiving device, there are cases in which a noise component is mixed due to the influence of the deep portions of the dermis 22 or a thick blood vessel in the subcutaneous tissue 23. The magnitude of the influence of the noise component differs according to the position of the light receiving device, and the magnitude of the influence is greatly changed even when the position of the light receiving device is slightly deviated. More specifically, for example, in the example shown in
The measurement signal specification unit 117 specifies a measurement signal Gi, whose value of the weighting factor wij (that is, a weighting factor wij included in a j-th column of the array W) of the component Sj specified by the component specification unit 116 is the smallest, from among the measurement signals G1, G2, G3, and G4. For example, when the component S1 is specified by the component specification unit 116, the measurement signal specification unit 117 specifies i, the value of which is the smallest, of the weighting factor wi1 (that is, w11, w21, w31, and w41). When the subscript i is specified, the measurement signal Gi is specified. In addition, for example, when the component S2 is specified by the component specification unit 116, the measurement signal specification unit 117 specifies i, the value of which is the smallest, of the weighting factor wi2 (that is, w12, w22, w32, and w42).
Since the measurement signal Gi specified by the measurement signal specification unit 117 is a measurement signal in which the weighting factor wij of the component Sj specified as a component which includes a large number of noise components is the smallest, the measurement signal Gi is a measurement signal in which the influence of noise components is the smallest. As described above, the second embodiment is configured such that a measurement signal in which the influence of noise is the smallest is selected from among the measurement signals output from the plurality of light receiving devices, and a pulse wave is measured using the selected measurement signal. The measurement signal specification unit 117 stores information (a flag) indicative of the specified measurement signal Gi in the flag storage region 181. The flag is referred to by the pulse wave information generation unit 118.
The pulse wave information generation unit 118 generates pulse wave information indicative of a pulse wave based on the measurement signal specified by the measurement signal specification unit 117. In the second embodiment, the pulse wave information generation unit 118 regards a time interval between the peaks of the wave form of the measurement signal specified by the measurement signal specification unit 117 as a pulse period, regards the appearance frequency of peaks of the wave form of the measurement signal during a predetermined time (1 minute) as a pulse rate, and outputs information (pulse wave information) indicative of the pulse period and the pulse rate on the display unit 150. In addition, the pulse wave information generation unit 118 may, for example, chronologically store the information indicative of the pulse period and the pulse rate in the storage unit 180.
When the output of the measurement signals starts, the control unit 110 performs the independent component analysis on the measurement signals G1, G2, G3, and G4, which are output from the respective light receiving devices 211, 212, 213, and 214 and on which the A/D conversion is performed by the signal processing unit 120, in step S112, divides each of the measurement signals G1, G2, G3, and G4 into the plurality of components S1, S2, S3, and S4, and acquires weighting factors w11, w12, . . . , and w44 which are related to each of the components.
Subsequently, the control unit 110 calculates the dispersion σj of the acquired weighting factors for each of the components Si, S2, S3, and S4, and specifies a component Sj, in which the dispersion is the greatest, in step S113. Among the measurement signals G1, G2, G3, and G4 in step S114, the control unit 110 specifies a measurement signal in which the weighting factor wij of the component Sj which is specified in step S113 is the smallest, and stores information indicative of the specified measurement signal in the flag storage region 181.
Once the processes up to step S114 have been completed, the control unit 110 generates pulse wave information indicative of a pulse wave based on the measurement signal specified in step S114 in step S115. More specifically, the control unit 110 detects, from among the measurement signals G1, G2, G3, and G4 which are output from the signal processing unit 120, a time interval between peaks of the wave form of a measurement signal indicated by the flag stored in the flag storage region 181 as a pulse period, an appearance frequency of peaks of the waveform of the measurement signal during a predetermined time as a pulse rate, and displays an image indicative of the detected pulse period and the pulse rate on the display unit 150. Meanwhile, if the control unit 110 does not receive the operation of measuring a pulse wave through the operation unit 160 (step S110; NO), the control unit 110 stands by until the operation is performed.
The control unit 110 repeatedly performs the process in step S115 until an operation of terminating the measurement of the pulse wave is performed through the operation unit 160 (step S116; NO). The control unit 110 terminates the process when the operation of terminating the measurement of the pulse wave is performed through the operation unit 160 (step S116; YES).
Due to the difference in density of the blood vessels of a living body and the influence of the distribution of blood vessels having different thicknesses, the level of each of the results of measurement varies according to the measurement position of the living body as shown in
In the second embodiment, among the components S1, S2, S3, and S4 which are estimated by performing the independent component analysis, a component, in which the dispersion of the weighting factors is the greatest, is specified as a component which includes a large number of noise components, and a measurement signal in which the influence of the specified component is the smallest is used as a signal used to measure the pulse wave. The measurement of a pulse wave, in which the influence of body motion noise is suppressed, is performed using such a measurement signal.
The invention is not limited to the above-described embodiments, and may be performed in such a way as to be modified as below. In addition, modification examples below may be combined.
(1) In each of the above-described embodiments, the pulse wave measurement apparatus 1 has the configuration shown in
(2) In each of the above-described embodiments, the light emitting and receiving unit 210 has the configuration which includes the light emitting device 215 that emits light of a wave length of green light, and the light receiving devices 211, 212, 213, and 214 which receive light of a wave length of green light. However, light which is emitted and received by the light emitting and receiving units is not limited to green light, and light of other wave lengths, such as blue light or infrared light, may be used. In addition, in the above-described embodiments, the light emitting and receiving unit 210 has the configuration which includes the single light emitting device 215 and the plurality of light receiving devices 211, 212, 213, and 214. However, the number of light emitting devices may be plural. For example, a configuration in which the light receiving device corresponds to the light emitting device one to one may be used. In this case, like the above-described first embodiment, the control unit 110 specifies a component in which the dispersion of the weighting factors satisfies a predetermined condition by performing the above-described independent component analysis, the dispersion calculation process, and the component specification process on the measurement signals which are indicative of the results of detection performed by the respective light receiving devices. In addition, as well as in the above-described second embodiment, even when the light receiving device corresponds to the light emitting device one to one, the control unit 110 specifies a measurement signal in which the component in which the dispersion of the weighting factors satisfies the predetermined condition is the smallest by performing the above-described independent component analysis, the dispersion calculation process, and the component specification process on the measurement signals which are indicative of the results of detection performed by the respective light receiving devices.
(3) In each of the above-described embodiments, the pulse wave measurement apparatus 1 has the configuration which includes four light receiving devices 211, 212, 213, and 214. However, the number of light receiving devices is not limited to four and a larger or smaller number of light receiving devices may be used. In addition, in the above-described embodiments, the control unit 110 estimates four components S1, S2, S3, and S4 by performing the independent component analysis on each of four measurement signals G1, G2, G3, and G4 which are output from the respective four light receiving devices 211, 212, 213, and 214. However, the number of measurement signals and the number of estimated components are not limited to four and a larger or smaller number of measurement signals and a larger or smaller number of estimated components may be used. In brief, the control unit 110 may estimate a plurality of components (independent components) by performing the independent component analysis on a plurality of measurement signals. However, it is necessary that the number of measurement signals n and the number of estimated components m satisfy a relationship of n m. The condition n≧m is a condition which is necessary to estimate components using the independent component analysis.
(4) In the above-described first embodiment, the control unit 110 estimates a component in which the dispersion of the values of the weighting factors is the smallest from among the plurality of components which are estimated by performing the independent component analysis, and generates pulse wave information using the specified component. The component specification embodiment is not limited thereto. For example, the control unit 110 may specify two components, that is, a component in which the dispersion of the weighting factors is the smallest and a component in which the dispersion of the weighting factors is the second smallest. When the plurality of components are specified, the control unit 110 may calculate the sum of the plurality of specified components (or the sum of weightings), and may generate the pulse wave information based on the results of calculation.
In addition, as another example of the component specification embodiment according to the above-described first embodiment, for example, the control unit 110 may specify one or more components in which the dispersion σi of the values of the weighting factors is smaller than a predetermined threshold (the degree of dispersion is within a predetermined range). In brief, the control unit 110 may specify a component in which the dispersion of the values of the weighting factors satisfies a predetermined condition (dispersion is small). In addition, when a plurality of components are specified, the control unit 110 may calculate the sum of the plurality of specified components (or the sum of weightings), and may generate the pulse wave information based on the results of calculation.
(5) In the above-described first embodiment, the control unit 110 performs the component specification process (the processes in steps S12 to S13 in
(6) The above-described first embodiment has the configuration in which the control unit 110 specifies a component, in which the dispersion of the weighting factors satisfies a predetermined condition, from among the components estimated by performing the independent component analysis on the plurality of measurement signals, and measures the pulse wave using the specified component, thereby reducing errors generated due to body motion noise which varies according to a measurement position. In addition to this, a configuration in which the control unit 110 performs a process of removing body motion noise (for example, body motion noise generated due to blood congestion) which does not vary due to a measurement position may be used. More specifically, for example, the control unit 110 may determine whether or not blood congestion exists by measuring the change in the volume of blood vessels at the measurement position, and, when it is determined to be the blood congestion, may perform a filtering process of removing a noise component generated due to the blood congestion on the specified component.
(7) In addition, as another example, for example, a configuration in which an acceleration sensor is provided in the pulse wave measurement apparatus 1 is used, and an operation of the measurement target person may be detected by the acceleration sensor, and the control unit 110 may determine whether or not there is blood congestion based on the results of detection. In this case, for example, when the control unit 110 may detect the direction of the measurement position by the acceleration sensor, and may determine whether or not the posture of the measurement position is easy to be congested due to the influence of weight. When it is determined that the posture of the measurement position is easy to be congested, the control unit 110 may perform the filtering process of removing a noise component generated due to the blood congestion on the specified component based on a predetermined algorithm.
(8) In the above-described second embodiment, the control unit 110 specifies a component in which the dispersion of the values of the weighting factors is the greatest as a component which includes a large amount of noise from among the plurality of components estimated by performing the independent component analysis. The component specification embodiment is not limited thereto. For example, the control unit 110 may specify two components, that is, a component in which the dispersion of the weighting factors is the greatest and a component in which the dispersion of the weighting factors is the second greatest. In addition, as another example of the component specification embodiment, for example, the control unit 110 may specify one or more components in which the dispersion σi of the values of the weighting factors is greater than a predetermined threshold. In brief, the control unit 110 may specify a component in which the dispersion of the values of the weighting factors satisfies a predetermined condition (dispersion is great).
When a plurality of components is specified as a component which includes a large amount of noise, the control unit 110 may calculate, for example, the total sum of the weighting factors of the specified component for each measurement signal, and may specify a measurement signal in which the result of calculation is the smallest as a measurement signal which is used to generate the pulse wave information. More specifically, for example, when two components, that is, the component S1 and the component S2 are specified as components which include a large amount of noise, the control unit 110 may calculate the sum of the weighting factor wi1 of the component S1 and the weighting factor wi2 of the component S2 (w11+w12+w21+w22+w31+w32, and w41+w42) for each measurement signal Gi, and may specify the measurement signal Gi in which the result of calculation is the smallest. In addition, as another example, for example, when a plurality of components is specified as a component which includes a large amount of noise, the control unit 110 may calculate the product of the weighting factors of the specified component for each measurement signal, and may specify a measurement signal in which the result of calculation is the smallest. In brief, any configuration in which the control unit 110 specifies a measurement signal which satisfies a predetermined condition (it is considered that the influence of noise is small) from among the measurement signals Gi may be used.
(9) In the above-described second embodiment, the control unit 110 specifies a measurement signal at a timing in which the operation of measuring a pulse wave is performed by a measurement target person. However the timing in which the measurement signal specification process is performed is not limited to the above timing. For example, the measurement signal specification process may be performed at a timing in which the power of the pulse wave measurement apparatus 1 is turned on. In addition, for example, the measurement signal specification process may be performed at a timing in which an operation of performing initial setting is performed by the measurement target person. In brief, the control unit 110 may perform the measurement signal specification process at any of the timings, and may measure the pulse wave using the specified measurement signal when the pulse wave measurement process is performed.
(10) The above-described second embodiment has the configuration in which the control unit 110 specifies one or more components as a component which includes a large amount of noise from among components estimated by performing the independent component analysis on the plurality of measurement signals, and measures the pulse wave using a measurement signal in which the weighting factors of the specified component is the smallest, thereby reducing errors generated due to body motion noise which varies according to a measurement position. In addition to this, a configuration in which the control unit 110 performs a process of removing body motion noise (for example, body motion noise generated due to blood congestion) which does not vary due to a measurement position may be used. More specifically, for example, the control unit 110 may determine whether or not blood congestion exists by measuring the change in the volume of blood vessels at the measurement position, and, when it is determined to be the blood congestion, may perform a filtering process of removing a noise component generated due to the blood congestion on the specified measurement signal.
In addition, as another example, for example, a configuration in which an acceleration sensor is provided in the pulse wave measurement apparatus 1 is used, and an operation of the measurement target person may be detected by the acceleration sensor, and the control unit 110 may determine whether or not there is blood congestion based on the results of detection. In this case, for example, when the control unit 110 may detect the direction of the measurement position by the acceleration sensor, and may determine whether or not the posture of the measurement position is easy to be congested due to the influence of weight. When it is determined that the posture of the measurement position is easy to be congested, the control unit 110 may perform the filtering process of removing a noise component generated due to the blood congestion on the specified measurement signal based on a predetermined algorithm.
(11) In the above-described second embodiment, a configuration in which the control unit 110 removes a component Sj which includes a large amount of noise and which is specified in step S113 from the measurement signal specified in step S114, and generates a pulse wave information based on the measurement signal from which the component Sj which includes a large amount of noise is removed may be used. Since the measurement signals G1, G2, G3, and G4 and the components S1, S2, S3, and S4 satisfy the above-described Equation 4, the component Sj is calculated using the above-described Equation 5. The pulse wave information generation unit 118 specifies the component Sj which includes a large amount of noise using the above-described Equation 5, and performs the filtering process of removing the specified component Sj from the measurement signal Gi specified in step S14.
(12)
(13) The above-described first embodiment may be combined with the second embodiment. That is, a configuration may be used in which the control unit 110 of the pulse wave measurement apparatus 1 generates pulse wave information based on a component which is specified by the component specification unit 113 (refer to
(14) A program which is executed by the control unit 110 of the pulse wave measurement apparatus 1 may be provided in a state in which the program is stored in a magnetic recording medium such as a magnetic tape or a magnetic disc, an optical recording medium such as an optical disc, a magneto-optical medium, and a computer-readable recording medium such as a semiconductor memory. In addition, it is possible to download the program via a communication network such as the Internet. Meanwhile, as a control device which performs such control, various types of devices other than a CPU can be applied, for example, a dedicated processor may be used.
The entire disclosure of Japanese Patent Application No. 2012-020766, filed Feb. 2, 2012, No. 2012-020767, filed Feb. 2, 2012 and No. 2012-219093 filed Oct. 1, 2012 are expressly incorporated reference herein.
Number | Date | Country | Kind |
---|---|---|---|
2012-020766 | Feb 2012 | JP | national |
2012-020767 | Feb 2012 | JP | national |
2012-219093 | Oct 2012 | JP | national |