JP5742146B2 - INFORMATION PROCESSING METHOD, INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING DEVICE, DISPLAY DEVICE, INFORMATION PROCESSING PROGRAM, AND COMPUTER-READABLE RECORDING MEDIUM CONTAINING THE PROGRAM - Google Patents

Description

  The present invention relates to an information processing method, an information processing system, an information processing device, a display device, an information processing program, and a computer-readable recording medium on which the program is recorded.

  It is extremely important from the viewpoint of health management to continuously grasp the state of breathing in daily life. It is also important to know how you are breathing during exercises that you can easily do to improve your health, such as walking and jogging. For example, it is known that a person can breathe at a constant rhythm during running and can easily run by synchronizing the breathing rhythm with the foot rhythm. Therefore, if the exercise and breathing states can be monitored simultaneously in real time, exercise management suitable for one's body is possible, which is very convenient.

In the biological signal monitoring in such a situation, it is desired that the sensor does not give the subject a sense of incongruity or restraint.
In recent years, inertia sensors such as accelerometers and gyro sensors that have been remarkably reduced in size and weight due to the advancement of MEMS technology have attracted attention as such sensors. An inertial sensor is a sensor that exerts a force to detect the movement of a moving body, but recently it has also been proposed that a respiratory rhythm can be detected using an acceleration sensor or a gyro sensor.

JP 2006-247374 A JP 2000-350716 A

However, the reports so far, including these proposals, have mainly focused on detecting respiratory rhythm in a resting state. That is, there is a problem that the respiratory rhythm cannot be accurately detected from the signal measured during the exercise of moving the body greatly.
One of the purposes of the present case has been created in view of such problems, and aims to accurately detect a respiratory rhythm from signals measured during regularly repeated exercise.

  In addition, the present invention is not limited to the above-described object, and is a function and effect derived from each configuration shown in the embodiment for carrying out the present invention, which is another object of the present invention. Can be positioned as one.

As a result of diligent research by the present inventors, it has been found that by using an inertial sensor, a respiratory rhythm can be accurately detected without being obstructed by the movement of the body during regularly repeated exercise.
In other words, the gist of this case is
An extraction process for extracting a respiratory rhythm of the living body from the body movement signal by filtering a signal based on a repetitive rhythmic movement accompanying the voluntary movement of the living body (hereinafter referred to as a body movement signal) is provided. It is an information processing method.

Further, the gist of the present case is that a body motion signal detection unit that detects a signal based on a repetitive rhythmic motion accompanying a voluntary motion of a living body as a body motion signal, and filtering the body motion signal, the body motion signal An information processing system comprising: a rhythm extraction unit that extracts the respiratory rhythm.
Further, the gist of the present invention is that a rhythm that extracts a respiratory rhythm of the living body from the body motion signal by performing a filtering process on a signal (hereinafter referred to as a body motion signal) based on a repetitive rhythm movement accompanying the voluntary movement of the living body. An information processing apparatus including an extraction unit.

The gist of the present invention is a display device characterized by having an output unit for outputting the results extracted by the information processing device.
Furthermore, the gist of the present case is to extract a respiratory rhythm of the living body from the body motion signal by performing a filtering process on a signal (hereinafter referred to as a body motion signal) based on a repetitive rhythm movement accompanying the voluntary movement of the living body. An information processing program that causes a computer to execute a process.

  In addition, the gist of the present case is an extraction that extracts a respiratory rhythm of the living body from the body motion signal by performing a filtering process on a signal (hereinafter referred to as a body motion signal) based on a repetitive rhythm movement accompanying the voluntary movement of the living body. A computer-readable recording medium having an information processing program recorded thereon, wherein the process is executed by a computer.

  It becomes possible to accurately detect a respiratory rhythm from a signal based on voluntary movement.

It is a figure which shows the structure of the system as an example of embodiment. It is a flowchart for demonstrating the respiration rhythm extraction process of the system as an example of embodiment. (A) and (B) are signals obtained by the system as an example of the embodiment. It is a flowchart for demonstrating operation | movement of the time constant determination part as an example of embodiment. It is a figure which shows the change of the dispersion | variation with respect to the change of the time constant as an example of embodiment. It is a flowchart for demonstrating operation | movement of the respiration rhythm extraction part as an example of embodiment. It is a flowchart for demonstrating operation | movement of the exercise rhythm extraction part as an example of embodiment. It is a flowchart for demonstrating the cycle ratio determination process of the system as an example of embodiment. It is a flowchart for demonstrating operation | movement of the ratio determination part as an example of embodiment. (A)-(C) are the figures for demonstrating the pattern matching method as an example of embodiment. It is the cycle ratio obtained by the system as an example of the embodiment. It is a flowchart for demonstrating operation | movement of the determination part as an example of embodiment. It is the cycle ratio obtained by the system as an example of the embodiment. It is the cycle ratio obtained by the system as an example of the embodiment. It is a figure which shows the change of the dispersion | variation with respect to the change of the time constant as an example of embodiment. It is the cycle ratio obtained by the system as an example of the embodiment. It is the cycle ratio obtained by the system as an example of the embodiment. It is the cycle ratio obtained by the system as an example of the embodiment.

Hereinafter, embodiments of the information processing method will be described with reference to the drawings.
[A] Description of Embodiment FIG. 1 is a diagram illustrating a configuration of a system according to an example of an embodiment. The system 1 includes a body motion signal detection device 10 and an information processing device 20. The body motion signal detection device 10 is communicably connected to the information processing device 20 via, for example, wired such as the Internet or wireless such as a wireless LAN (Local Area Network) or Bluetooth (registered trademark).

  The body motion signal detection device 10 detects (measures) a non-invasive and continuous rhythmic motion accompanying a voluntary motion of a subject (living body), for example. Hereinafter, the repeated rhythmic movement accompanying the voluntary movement may be simply referred to as rhythmic movement. In other words, the voluntary movement is an exercise that can be consciously controlled by the subject. In addition, when the voluntary movement is a movement that is repeated regularly, the rhythm and the respiratory rhythm of the voluntary movement itself can be obtained more accurately. In addition, the voluntary exercise | movement repeated regularly includes not only that the completely same exercise | movement is repeated, but repeating substantially the same exercise | movement.

  Here, the voluntary exercise is, for example, one having a shorter cycle (preferably less than half) than the breathing rhythm of the subject during the voluntary exercise, such as walking, jogging, running, cycling, swimming, gymnastics, weight training. , Physical strength measurement (lifting up / down steps, repeated side jumps), juggling (lifting beanbags, soccer balls), etc. In addition, the repetitive rhythmic movement accompanying the voluntary movement includes, for example, a rhythmic movement of breathing and walking itself accompanying the walking which is the voluntary movement.

In the present application, breathing exercise or the like does not correspond to voluntary exercise. Non-invasive means, for example, that the body of the subject is not damaged and that the subject is not burdened.
The body motion signal detection device 10 is configured to be portable, for example. The attachment position of the body motion signal detection device 10 to the subject is not particularly limited as long as it is a part that can detect the movement of the body, but it is desirable to attach it to the trunk of the subject in order to accurately capture the respiratory rhythm, More preferably, it is desirable to attach to the subject's abdomen or chest.

The body motion signal detection device 10 includes, for example, a body motion signal detection unit 11, a storage unit 12, and an interface unit 13. The body motion signal detection unit 11, the storage unit 12, and the interface unit 13 are connected to be communicable with each other.
The body motion signal detection unit 11 detects (measures), for example, a repetitive rhythm motion associated with voluntary movement as a body motion signal (a signal based on a repetitive rhythm motion associated with a voluntary movement of a living body). That is, the body motion signal detection unit 11 detects a signal based on a repetitive rhythm movement accompanying a voluntary movement of a living body as a body motion signal. From a different viewpoint, the body motion signal detection unit 11 detects, for example, a repetitive rhythmic motion accompanying a voluntary motion of a living body as one body motion signal. Specifically, the subject's rhythmic movement, for example, changes in force, spatial changes in body position, sounds emitted from the body, waves such as electromagnetic waves, changes in fine energy, or changes in the field around the body, etc. Detect (measure) as a body motion signal. That is, the body motion signal detection unit 11 measures a signal based on repetitive rhythm movement accompanying voluntary movement.

  Here, the body motion signal detection unit 11 is realized by an inertial sensor such as an acceleration sensor, a speed sensor, or a gyro sensor, for example. About the inertial sensor which detects the said body motion signal, it selects suitably according to the kind of signal to detect, for example. Usually, when detecting a walking rhythm, an acceleration sensor that measures acceleration of body movement is preferably used, but the present invention is not limited to this.

In addition, as the acceleration sensor, one of one axis to three axes can be arbitrarily used, but three axes for detecting acceleration acting in three directions of the vertical direction, the horizontal front-rear direction, and the horizontal left-right direction at the time of walking. Although an acceleration sensor is preferably used, the present invention is not limited to this.
Note that the body motion signal detection unit 11 measures the body motion signal at a predetermined sampling frequency (for example, 100 Hz), for example.

  The storage unit 12 is a storage device capable of storing various types of information, such as RAM (Random Access Memory), HDD (Hard Disk Drive), SSD (Solid State Drive), flash memory, and the like. Specifically, the storage unit 12 stores the body motion signal obtained by the body motion signal detection unit 11, for example. Moreover, the memory | storage part 12 may be provided detachably with respect to the body motion signal detection apparatus 10, for example, and may be fixed to the body motion signal detection apparatus 10. FIG.

  The interface unit 13 is an interface connected to the information processing apparatus 20 so as to be communicable, for example. When the body motion signal detection device 10 is connected to the information processing device 20 via wireless, for example, the interface unit 13 includes an antenna. Further, when the body motion signal detection device 10 is connected to the information processing device 20 via a wire, for example, the interface unit 13 is a wire connection terminal. For example, the storage unit 12 is provided detachably with respect to the body motion signal detection device 10, and when the storage unit 12 is detached from the body motion signal detection device 10 and connected to the information processing device 20, The interface unit 13 may not be provided in the body motion signal detection device 10. The information processing device 20 is, for example, a PC (Personal Computer), and extracts a signal (hereinafter sometimes simply referred to as a respiratory rhythm) representing the subject's respiratory rhythm from the body motion signal obtained by the body motion signal detection device 10. To do.

The information processing apparatus 20 includes, for example, a central processing unit 21, a storage unit 22, an output unit 23, and an interface unit 24. Here, the central processing unit 21, the storage unit 22, the output unit 23, and the interface unit 24 are connected to be communicable with each other.
The storage unit 22 is a storage device that can store data such as RAM, HDD, SSD, and the like, and stores various types of information.

The central processing unit 21 is a processing device that performs various calculations and controls by executing various application programs stored in the storage unit 22, thereby realizing various functions.
For example, the central processing unit 21 executes an information processing program stored in the storage unit 22 to thereby execute a time constant determination unit 211, a respiratory rhythm extraction unit 212, an exercise rhythm extraction unit 213, a ratio determination unit 214, and a determination unit. 215 and the output control unit 216 function. That is, the information processing program is a program that causes the central processing unit 21 to function as the time constant determination unit 211, the respiratory rhythm extraction unit 212, the exercise rhythm extraction unit 213, the ratio determination unit 214, the determination unit 215, and the output control unit 216. is there.

The time constant determination unit 211 acquires, for example, the body motion signal obtained by the body motion signal detection unit 11 from the storage unit 12, and based on the acquired body motion signal, a signal representing the respiratory rhythm and the rhythm of the voluntary motion itself (Hereinafter, simply referred to as “motor rhythm”) is calculated.
Specifically, the time constant determination unit 211 determines the time constant by performing the following processing.

  (1) The time constant determination unit 211 applies the body motion signal X to a high-pass filter or a band-pass filter characterized by a certain time constant T. Here, for example, as a high-pass filter, when F (X, T) is used to describe the process of smoothing the body motion signal X with a zero phase moving average filter with a time width (time constant) T, the output waveform (signal) The process expressed by Y = XF (X, T) is performed. Further, for example, as a bandpass filter, processing represented by Y = F (X−F (X, T), T / 2.5) is performed. Here, the zero phase moving average filter refers to a moving average filter whose phase shift is zero. The zero phase moving average filter can be realized by using various known methods, and detailed description thereof is omitted. An example of the time constant of the bandpass filter is T / 2.5, but the present invention is not limited to this.

  (2) Next, the time constant determination unit 211 digitizes the regularity of the output waveform Y. The time constant determination unit 211 obtains, for example, CV (= standard deviation / average value: Coefficient of Variation) of absolute values of values at the maximum and minimum peaks of the output waveform Y. More specifically, the time constant determination unit 211 extracts the maximum and minimum peaks from the output waveform Y, obtains the absolute value of the amplitude of the waveform at the peak position, and calculates the standard deviation and the average from the obtained waveform amplitude. Find the value. Then, the time constant determination unit 211 calculates the variance, that is, CV by dividing the standard deviation by the average value.

  (3) Furthermore, the time constant determination unit 211 obtains regularity of the output waveform Y when the time constant T is changed. For example, the time constant determination unit 211 graphs the regularity of Y when the time constant T is changed. Specifically, the time constant determination unit 211 obtains a change in the regularity of the output waveform Y with respect to the time constant T, for example, by calculating the CV at each time constant T when the time constant T is changed. .

  (4) The time constant determination unit 211 obtains a minimum point from the change in regularity of the output waveform Y (change in CV) with respect to the time constant T obtained in the process of (3) above. Then, the time constant determination unit 211 separates the motion rhythm from the body motion signal, for example, the time constant T corresponding to the minimum point with the smaller time constant T among the obtained minimum points (for example, two minimum points). Thus, it is determined as a time constant for extraction (hereinafter sometimes referred to as a second time constant). Furthermore, the time constant determination unit 211 extracts, for example, the time constant T corresponding to the minimum point having the larger time constant T among the obtained minimum points by separating the respiratory rhythm from the body motion signal. It is determined as a constant (hereinafter sometimes referred to as a first time constant). That is, the time constant determination unit 211 selects a time constant that enables separation and extraction of a waveform having regularity as much as possible. In other words, the time constant determination unit 211 selects a plurality of minimum points (for example, two points), and sets the time constant T corresponding to the minimum point with the larger time constant T as the first time constant and the smaller of the time constant T. The time constant T corresponding to the local minimum point is determined as the second time constant. That is, the time constant determination unit 211 determines the first time constant and the second time constant based on the output when the bandpass filter is applied to the body motion signal.

In addition, it is preferable from the viewpoint of noise removal that the process of (1) to (4) includes the process of performing N-order integration (N = 1, 2, 3,...) Of the body motion signal. When performing N-order integration, it is preferable to perform the process (1) (filter process) N times or more. The order of the integration operation and the filtering process may be any order as long as the following conditions are satisfied, for example.
・ Do not perform the filter process more than once in succession.
・ At least once in the filter process, the integration process is completed.
-When the filter process is performed twice or more, the filter is a high-pass filter except for the last one.

For example, the respiratory rhythm extraction unit 212 acquires the body motion signal obtained by the body motion signal detection unit 11 from the storage unit 12, and uses the first time constant (T1) determined by the time constant determination unit 211, A respiratory rhythm is extracted from the body motion signal by performing a filtering process.
Specifically, the respiratory rhythm extraction unit 212 extracts the respiratory rhythm from the body motion signal by performing the following processing (filtering processing), for example. In addition, the following process is an illustration and is not limited to this.

  (5) For example, the respiratory rhythm extraction unit 212 obtains a low frequency component of the body motion signal by applying a zero phase moving average filter having a first time constant to the body motion signal. That is, the respiratory rhythm extraction unit 212 performs processing represented by F (X, T1) on the body motion signal. Next, the respiratory rhythm extraction unit 212 subtracts the low frequency component obtained by performing the process represented by F (X, T1) from the body motion signal. That is, the respiratory rhythm extraction unit 212 calculates Y = X−F (X, T1). In other words, the respiratory rhythm extraction unit 212 performs high-pass filtering on the body motion signal using the first time constant.

Note that this high-pass filtering makes it possible to prevent divergence, for example, in an integration process described later.
(6) Next, the respiratory rhythm extraction unit 212 performs integration on the signal after the high-pass filtering. The order of integration is arbitrary. For example, the respiratory rhythm extraction unit 212 performs second-order integration on the signal after high-pass filtering.

(7) The respiratory rhythm extraction unit 212 performs the same process as (5) on the integrated signal X1 obtained by the process (6). That is, the respiratory rhythm extraction unit 212 performs processing represented by F (X1, T1) on the integrated body motion signal. In other words, the respiratory rhythm extraction unit 212 performs high-pass filtering on the signal after integration using the first time constant.
(8) Next, the respiratory rhythm extraction unit 212 removes the motion rhythm by performing low-pass filtering on the signal X2 obtained by the process (7). That is, the process represented by F (X2, T3) is performed on the signal obtained by the process (7), with the time constant of the low-pass filtering being T3. Here, the time constant T3 of the low-pass filtering can be set to T3 = T1 / 2.5, but is not limited to this, and may be a value larger than the second time constant T2.

That is, the respiratory rhythm extraction unit 212 performs bandpass filtering on the body motion signal using the first time constant by the processes (5) to (8).
(9) Furthermore, the respiratory rhythm extraction unit 212 creates an envelope connecting the maximum values of the signal obtained by the process of (8) and an envelope connecting the minimum values, and these two envelopes A process of subtracting the signal having the average of the above from the signal obtained by the process (8) is performed. Since the amplitude of the signal obtained by this processing changes with the amplitude value 0 interposed therebetween, the phase can be accurately obtained when the Hilbert transform method described later is used. Note that this processing may be omitted when a pattern matching method described later is used.

Through the processes (5) to (9), the respiratory rhythm extraction unit 212 extracts the respiratory rhythm from the body motion signal. That is, the respiratory rhythm extraction unit 212 functions as a rhythm extraction unit that extracts a respiratory rhythm from a body motion signal by performing a filtering process on the body motion signal.
In the process (6), the body motion signal is second-order integrated, but the present invention is not limited to this, and the integration process may not be performed. Further, the present invention is not limited to the second order integration, and Nth order integration may be performed (N = 1, 2, 3,...). In addition, when performing N-order integration, it is preferable to perform the process (filter process) of (5) [or (7)] N times or more. Furthermore, although the high-pass filtering is performed twice in the above (5) and (7), it is not limited to this. The order of the integration operation and the filtering process may be any order as long as the following conditions are satisfied, for example.
・ Do not perform the filter process more than once in succession.
・ At least once in the filter process, the integration process is completed.
-When the filter process is performed twice or more, the filter is a high-pass filter except for the last one.

The exercise rhythm extraction unit 213 acquires, for example, the body motion signal obtained by the body motion signal detection unit 11 from the storage unit 12, and uses the second time constant (T2) determined by the time constant determination unit 211, The motion rhythm is extracted from the body motion signal by performing a filtering process.
Specifically, the exercise rhythm extraction unit 213 extracts the exercise rhythm from the body motion signal, for example, by performing the following process (filtering process). In addition, the following process is an illustration and is not limited to this.

  (10) For example, the motion rhythm extraction unit 213 obtains a low frequency component of the body motion signal by applying a zero-phase moving average filter having a second time constant to the body motion signal. That is, the movement rhythm extraction unit 213 performs processing represented by F (X, T2) on the body movement signal. Next, the exercise rhythm extraction unit 213 subtracts the low frequency component obtained by performing the process represented by F (X, T2) from the body motion signal. That is, the movement rhythm extraction unit 213 obtains Y = X−F (X, T2). In other words, the exercise rhythm extraction unit 213 performs high-pass filtering on the body motion signal using the second time constant.

Note that this high-pass filtering makes it possible to prevent divergence, for example, in an integration process described later.
(11) Next, the motion rhythm extraction unit 213 performs integration on the signal after the high-pass filtering. The order of integration is arbitrary. For example, the motion rhythm extraction unit 213 performs second-order integration on the signal after high-pass filtering.

(12) The movement rhythm extraction unit 213 performs the same processing as (10) on the signal X2 after integration in the processing (11). That is, the motion rhythm extraction unit 213 performs processing represented by F (X2, T2) on the integrated body motion signal. In other words, the motion rhythm extraction unit 213 performs high-pass filtering on the signal after integration using the second time constant.
Through the processes (10) to (12), the exercise rhythm extraction unit 213 extracts the exercise rhythm from the body motion signal. That is, the exercise rhythm extraction unit 213 functions as a rhythm extraction unit that extracts a rhythm (exercise rhythm) of voluntary exercise from the body motion signal by performing filtering processing on the body motion signal.

In the process of (11), the body motion signal is second-order integrated. However, the present invention is not limited to this, and the integration process may not be performed. Further, the present invention is not limited to the second order integration, and Nth order integration may be performed (N = 1, 2, 3,...). In addition, when performing N-order integration, it is preferable to perform the process (filter process) of (10) [or (12)] N times or more. The order of the integration operation and the filtering process may be any order as long as the following conditions are satisfied, for example.
・ Do not perform the filter process more than once in succession.
・ At least once in the filter process, the integration process is completed.
-When the filter process is performed twice or more, the filter is a high-pass filter except for the last one. Note that the processing similar to (9) may be performed after the processing of (12).

  The ratio determination unit 214 determines, for example, the ratio (for example, cycle ratio) of the phase change amount in a predetermined time between the respiratory rhythm and the exercise rhythm. Here, the phase is an index representing how many rotation angles each point on the rhythm waveform corresponds to, assuming that a rhythm that repeats in time is a movement rotating on the circumference. For example, there is a difference of 360 degrees between the phases of two adjacent peak points in a sine wave. The cycle ratio refers to, for example, the number of repetitions of exercise rhythm during one breath. More specifically, the cycle ratio is a value indicating, for example, how many movement rhythms or how many movement rhythm peaks exist within one respiratory rhythm. The fixed time is, for example, a time when the phase of any one rhythm (for example, the phase of the respiratory rhythm) changes by an integral multiple of 360 degrees, but is not limited to this, and the fixed time is arbitrary. Value.

The ratio determination unit 214 includes a phase calculation unit 224 and a ratio calculation unit 234.
For example, the phase calculation unit 224 calculates the phase change amount of the respiratory rhythm and the motion rhythm for a certain period of time by calculating the phase of the respiratory rhythm and the motion rhythm. The phase calculation unit 224 calculates the phases of the respiratory rhythm and the exercise rhythm using, for example, the Hilbert transform method or the pattern matching method. Note that the method for calculating the phase is not limited to the Hilbert transform method or the pattern matching method.

  Here, the Hilbert transform method is a method for specifically calculating the phase at a predetermined position of the waveform, and the pattern matching method is not a method for directly specifying the phase, but is a method for finding points where the phases are mutually shifted by 360 degrees. is there. That is, the pattern matching method is a method of finding a point whose phase is shifted by an integral multiple of 360 degrees from a predetermined position of the waveform. In other words, the pattern matching method calculates a relative phase (for example, 360 degrees) with respect to a predetermined position of the waveform.

The ratio calculation unit 234 calculates, for example, the ratio between the phase change amount of the respiratory rhythm and the phase change amount of the exercise rhythm calculated by the phase calculation unit 224. For example, the ratio calculation unit 234 calculates a cycle ratio. First, the case where the phase calculation unit 224 and the ratio calculation unit 234 calculate the ratio of the phase change amount using the Hilbert transform method will be described.
(A) When using the Hilbert transform method The Hilbert transform is a mathematical method for deriving a corresponding imaginary part Y (t) from an arbitrary real time series signal X (t). Thereby, the phase calculation part 224 can obtain | require directly the phase (theta) (t) of X (t) from the following (1) Formula. That is, the phase calculation unit 224 can obtain the phase of the respiratory rhythm and the movement rhythm at a predetermined time from the following equation (1).
(Equation 1)
Z (t) = X (t) + iY (t) = A (t) exp (iθ (t)) (1)
Further, assuming that the phase of the respiratory rhythm and the motion rhythm at an arbitrary time t is θB (t) degree and θG (t) degree, respectively, by calculating the phase of the respiratory rhythm from the above equation (1), for example, θB ( A time t1 such that t1) = θB (t) +360 can be obtained. That is, the phase calculation unit 224 obtains the phase change amount θB (t1) −θB (t) for a certain time t−t1 with respect to the respiratory rhythm. Further, the phase calculation 224 calculates the phase of the movement rhythm at time t and time t1 from the equation (1), so that the phase change amount θG (t1) −θG (t) with respect to the movement rhythm for a certain time. Ask for. That is, the phase calculation unit 224 functions as a phase calculation unit that calculates the amount of change in phase over a certain time for each of the respiratory rhythm and the exercise rhythm.

Then, the ratio calculation unit 234 calculates, for example, the respiratory rhythm phase change θB (t1) −θB (t) and the exercise rhythm phase change θB (t1) −θB (t) calculated by the phase calculation unit 224. The ratio [θG (t1) −θG (t)] / [θB (t1) −θB (t)] is calculated. For example, the ratio calculation unit 234 calculates the cycle ratio by calculating [θG (t1) −θG (t)] / 360.
Next, a case where the phase calculation unit 224 and the ratio calculation unit 234 calculate the phase change amount ratio using the pattern matching method will be described.
(B) When using the pattern matching method The pattern matching method is a method for quantifying the similarity between two signals. Although there are many definitions and calculation methods of similarity, for example, a method described in “Image processing engineering (Ryoichi Suematsu, Hirohisa Yamada, Corona)” or the like is used. The most representative is the autocorrelation coefficient. For example, the following calculation is performed for a three-dimensional body motion signal.

  First, an appropriate reference wave is selected from the body motion data. The coordinates of the reference wave are expressed by the following equation (2).

  Here, it is assumed that the average value of p x, y, z is zero. The autocorrelation coefficient is calculated by the following equation (3).

This is a so-called scalar quantity and does not depend on the coordinate system. That is, the same value is obtained even when the body motion signal detection device 10 causes a rotational shift at the wearing site during body motion measurement. The same calculation is performed for a one-dimensional signal.
When the pattern matching method is used, a time t1 whose phase is shifted by an integer multiple of 360 degrees can be easily obtained from the phase of the rhythm waveform at an arbitrary time t. In other words, the phase calculation unit 224 can obtain a point whose phase is shifted by an integer multiple of 360 degrees from the phase of the respiratory rhythm and the motion rhythm at time t by using pattern matching. That is, the phase calculation unit 224 calculates a relative phase (for example, 360 degrees) with respect to the respiratory rhythm and motion rhythm points at time t. That is, the phase calculation unit 224 calculates the amount of change in phase (for example, 360 degrees) for a certain time with respect to the respiratory rhythm.

  In addition, the phase calculation unit 224 calculates, for example, how many cycles of the exercise rhythm or how many peaks of the exercise rhythm there are in a certain time from time t to time t1. Here, calculating the number of cycles of the motion rhythm or the number of peaks of the motion rhythm in a fixed time is equivalent to calculating the phase change of the motion rhythm in a fixed time. That is, the phase calculation unit 224 calculates the amount of change in phase for a certain time with respect to the movement rhythm. That is, the phase calculation unit 224 functions as a phase calculation unit that calculates the amount of change in phase over a certain time for each of the respiratory rhythm and the exercise rhythm. Note that the number of cycles of exercise rhythm and the number of peaks of exercise rhythm calculated by the phase calculation unit 224 are not limited to integers.

Then, the ratio calculation unit 234 calculates the ratio between the phase change amount of the respiratory rhythm calculated by the phase calculation unit 224 and the phase change amount of the exercise rhythm.
For example, the determination unit 215 determines whether the cycle ratio obtained by the ratio determination unit 214 is a value within a predetermined range. Furthermore, the determination unit 215 can also determine whether or not the cycle ratio is within a predetermined range for a predetermined time. The predetermined range and the predetermined time can be arbitrarily set by an administrator of the system 1 or the like, for example. Here, the predetermined range can be set, for example, as a range of ± 1 of the target cycle ratio, but is not limited thereto. The predetermined time can be set to 30 minutes, for example, but is not limited thereto. That is, the predetermined range and the predetermined time can be set to arbitrary values according to the purpose. For example, when jogging at a constant speed of 30 minutes, after the movement reaches a steady state, the predetermined range is set to 4 ± 1 and the predetermined time is set to 30 minutes.

  The output control unit 216 controls the output unit 23. For example, when the output unit 23 includes a display unit of the display, the output control unit 216 controls the display state of the output unit 23 to display various information on the output unit 23. For example, the output control unit 216 causes the display unit 23 to display the cycle ratio obtained by the respiratory rhythm and exercise rhythm and the ratio determination unit 214 extracted by the respiratory rhythm extraction unit 212 and the exercise rhythm extraction unit 213, respectively. The cycle ratio may be displayed on the output unit 23 as a graph, or may be displayed as a message such as “Cycle ratio is **”. In addition, for example, when the determination unit 215 determines that the cycle ratio is not a value within a predetermined range, a warning to that effect is given in addition to the display of the cycle ratio or instead of the display of the cycle ratio. Display. For example, if the cycle ratio is determined not to be within the specified range, for example, “Respiratory rhythm is disturbed → Stop exercise, reduce exercise intensity” or “Respiration and exercise are It doesn't fit. "

Further, if the output unit 23 warns, for example, by an alarm or vibration, the output control unit when the determination unit 215 determines that the cycle ratio is not a value within a predetermined range, for example. 216 controls the output unit 23 to cause the output unit 23 to issue a warning.
When the output unit 23 is provided outside the information processing apparatus 20, the output control unit 216, for example, the respiratory rhythm and the exercise rhythm extracted by the respiratory rhythm extraction unit 212 and the exercise rhythm extraction unit 213, respectively. In addition, the display state of the output unit 23 is controlled by transmitting various information such as the cycle ratio obtained by the ratio determining unit 214 to the output unit 23 via wireless or wired communication.

The output unit 23 outputs various types of information under the control of the output control unit 216. For example, the output unit 23 is a display, and if the cycle ratio obtained by the ratio determination unit 214 or the determination unit 215 determines that the cycle ratio is not a value within a predetermined range, for example, a warning is given. Is displayed.
Further, the output unit 23 may be a unit that informs about a change in state or a sudden abnormality by an alarm or vibration, for example, and the determination unit 215 has a cycle ratio that is not a value within a predetermined range. If it is determined, a warning is given by an alarm or vibration.

  Note that the output unit 23 may not be provided in the information processing apparatus 20 and may be provided outside the information processing apparatus 20. Here, the case where the output unit 23 is provided outside the information processing device 20 is, for example, the case where the output unit 23 is included in the body motion signal detection device 10, or the body motion signal detection device 10 and the information processing device. 20 is not included in either of them. When provided outside the information processing apparatus 20, for example, the information processing apparatus 20 is connected to the output unit 23 via wireless or wired connection. That is, the output unit 23 functions as an output unit that outputs a result extracted by the information processing apparatus. The output unit 23 functions as a display device including an output unit that outputs a result extracted by the information processing device.

The interface unit 24 is, for example, an interface that is communicably connected to the body motion signal detection device 10. When the information processing apparatus 20 is connected to the body motion signal detection apparatus 10 via wireless, for example, the interface unit 24 includes an antenna. Further, when the information processing apparatus 20 is connected to the body motion signal detection apparatus 10 via a wired line, for example, the interface unit 24 is a wired connection terminal. For example, the storage unit 12 is provided detachably with respect to the body motion signal detection device 10, and when the storage unit 12 is detached from the body motion signal detection device 10 and connected to the information processing device 20, The interface unit 24 also functions as a slot to which the storage unit 12 is connected, for example.
The respiratory rhythm extraction process of the system 1 as an example of the embodiment configured as described above will be described with reference to the flowchart (steps A1 to A3) shown in FIG. In the following, an example of a specific method for extracting a rhythm based on an integral signal will be described in detail for the case of using an acceleration signal during walking, but in the case of other inertial sensors or other signals during exercise But the same process can be applied.

  First, the body motion signal detector 11 detects a repetitive rhythmic motion accompanying a voluntary motion as a body motion signal (step A1). FIG. 3A shows a state in which the body motion signal detection device 10 (3-axis acceleration sensor) is attached to the center of the abdomen of the subject, and the body motion signal detection device 10 (body motion signal detection unit 11) performs 100 Hz sampling. It is a part of the measured acceleration signal of continuous walking for 200 seconds. In FIG. 3A, only the acceleration change in the traveling direction among the three axes is shown. Next, based on the body motion signal detected by the body motion signal detection unit 11, the information processing apparatus 20 determines a time constant for extracting a respiratory rhythm and a time constant for extracting an exercise rhythm ( Step A2: Time constant determination process). Then, using the time constant determined in step A2, the respiratory rhythm and the exercise rhythm are extracted from the body motion signal (step A3: extraction process).

Next, details of the time constant determination unit 211, that is, detailed operations of step A2 in FIG. 2 will be described with reference to the flowchart (steps A21 to A26) shown in FIG.
First, the time constant determination unit 211 performs filtering on the body motion signal detected by the body signal detection unit 11 using a predetermined time constant T (step A21). For example, the time constant determining unit 211 performs high-pass filtering represented by Y = XF (X, T) on the body motion signal, performs second-order integration, and again returns Y = XF ( Perform high-pass filtering represented by X, T). Furthermore, the time constant determination unit 211 performs low-pass filtering represented by Y = F (X, T / 2.5) after high-pass filtering, for example. That is, the time constant determination unit 211 performs bandbus filtering on the body motion signal.

  Next, the time constant determination unit 211 obtains the absolute value CV of the values at the maximum and minimum peaks of the output waveform Y, for example (step A22). Then, the time constant determining unit 211 obtains a change in CV with respect to a change in the time constant T, for example, by changing the time constant T and performing the processes in steps A21 and A22 (step A23). FIG. 5 is a diagram illustrating an example of a change in CV (dispersion) with respect to a change in time constant (filter time) T obtained in step A23. Next, the time constant determination unit 211 obtains a minimum point from the change in CV with respect to the change in the time constant T obtained in step A23. For example, as shown in FIG. 5, the time constant determination unit 211 obtains two local minimum points (marks ● and ♦ in FIG. 5) having a small CV value. Then, the time constant determination unit 211 sets the time constant T (for example, 0.3) corresponding to the local minimum point (marked with ● in FIG. 5) having the smaller time constant T to the second time constant and the large time constant T. A time constant T (for example, 1.3) corresponding to the local minimum point (marked by ◆ in FIG. 5) is determined as the first time constant (step A24). If the first time constant and the second time constant have been determined, the process of step A2 ends (see Yes route of step A25). On the other hand, for example, when a plurality of local minimum points cannot be obtained and the time constant determination unit 211 cannot determine the first time constant (see No route in step A25), for example, the output control unit 216 The output unit 23 is warned that the respiratory rhythm cannot be extracted (step A26). At this time, for example, the output control unit 216 sends a message to the output unit 23 such as “breathing cannot be measured correctly → adjust the sensor position” or “breathing cannot be measured correctly → synchrony could not be evaluated”. You may warn by displaying. That is, step A2 (steps A21 to A24) is an example of a time constant determination process for determining a first time constant based on an output when a bandpass filter is applied to a body motion signal.

Next, details of the breathing rhythm extraction unit 212, that is, detailed operations of step A3 in FIG. 2 will be described with reference to a flowchart (steps A31 to A36) shown in FIG.
First, the respiratory rhythm extraction unit 212 performs high-pass filtering on the body motion signal detected by the body motion signal detection unit 11 using the first time constant (step A31). Next, the respiratory rhythm extraction unit 212 performs N (for example, 2) floor integration (step A32). Then, high-pass filtering is again performed on the signal after integration using the first time constant (step A33). Next, the respiratory rhythm extraction unit 212 performs low-pass filtering on the signal obtained in step A33 using a time constant (for example, 0.6) larger than the second time constant (step A34). Furthermore, the respiratory rhythm extraction unit 212 creates an envelope connecting the maximum values of the signal after low-pass filtering and an envelope connecting the minimum values (step A35). Then, the rhythm extraction unit 212 subtracts the average of these envelopes from the low-pass filtered signal (step A36).

  The broken line in FIG. 3B shows an example of the signal obtained in step A36. The waveform indicated by the broken line corresponds to an abdominal motion trajectory due to breathing of the subject, that is, a respiratory rhythm. This is because it is generally difficult to perform regularly repeated exercises other than abdominal exercises due to breathing during voluntary exercises (particularly during voluntary exercises that are regularly repeated). That is, step A3 (steps A31 to A36) is an example of an extraction process for extracting a respiratory rhythm of the living body from the body motion signal by performing a filtering process on a signal based on a repetitive rhythmic movement accompanying the voluntary movement of the living body. It is.

Next, the details of the exercise rhythm extraction unit 213, that is, the detailed operation of step A3 in FIG. 2 will be described with reference to the flowchart (steps A37 to A39) shown in FIG.
First, the exercise rhythm extraction unit 213 performs high-pass filtering on the body motion signal detected by the body motion signal detection unit 11 using the second time constant (step A37). The movement rhythm extraction unit 213 performs N (for example, 2) order integration (step A38). Then, high-pass filtering is performed again on the signal after integration using the second time constant (step A39). The solid line in FIG. 3B shows an example of the signal obtained in step A39. The waveform indicated by the solid line corresponds to, for example, voluntary movement of the subject who is walking. That is, the waveform indicated by the solid line in FIG. 3B corresponds to the movement rhythm.

Next, the determination process of the system 1 as an example of the embodiment configured as described above will be described with reference to the flowchart (steps A4 to A6) shown in FIG.
First, the ratio determination unit 214 determines, for example, a cycle ratio from the respiratory rhythm and the exercise rhythm extracted by the respiratory rhythm extraction unit 212 and the exercise rhythm extraction unit 213, respectively (step A4). Then, the determination unit 215 determines whether or not the cycle ratio calculated in step A4 is a value within a predetermined range, for example (step A5). Based on the determination result of the determination unit 215, the output control unit 216 causes the output unit 23 to output the determination result (step A6). The determination result is, for example, whether or not the cycle ratio is within a predetermined range.

Next, the details of the ratio determining unit 214, that is, the detailed operation of step A4 in FIG. 8 will be described with reference to the flowchart (steps A41 to A44) shown in FIG.
First, the phase calculation unit 224 selects an arbitrary point at an arbitrary time t of the respiratory rhythm (step A41). The phase calculation unit 224 calculates the phase t using, for example, Hilbert transform or pattern matching, thereby calculating a time t1 at a point where the phase is shifted 360 degrees from an arbitrary point (step A42). That is, the phase calculation unit 224 calculates the phase change amount at time t−t1 for the respiratory rhythm. Next, the phase calculation unit 224 calculates the amount of phase change at time t-t1 of the exercise rhythm (step A43: phase calculation process). For example, the ratio calculation unit 234 calculates the ratio of the respiratory rhythm phase change amount (for example, 360 degrees) and the exercise rhythm phase change amount as the cycle ratio at time t-t1 (step A44: ratio calculation process).

  Here, FIG. 10 is a diagram for explaining an example of a signal obtained when the pattern matching method is used. An example of detailed operation of the ratio determining unit 214 when pattern matching is used will be described with reference to FIG. Here, steps A42 and A43 are an example of a phase calculation process for obtaining a phase change amount for a predetermined time for each of the respiratory rhythm and voluntary movement rhythm extracted by the extraction process. Step A44 is an example of a ratio calculation process for calculating the ratio between the phase change amount of the respiratory rhythm and the phase change amount of the rhythm based on the rhythm movement.

  FIG. 10A shows a part of the waveform extracted from the data of FIG. 3B. The solid line shows the movement rhythm and the broken line shows the respiratory rhythm. Here, the phase calculation unit 224 calculates the autocorrelation coefficient by selecting the reference wave having a width of 2 seconds for the respiratory rhythm and the reference wave having a width of 0.4 seconds for the motion rhythm, centering on the time indicated by *. The result is FIG. 10 (B). With respect to the respiratory rhythm, the positions of two points whose phases are shifted by 360 degrees are indicated by dotted lines. The interval between the two points (corresponding to time t-t1) corresponds to one cycle of the respiratory rhythm. Similarly, FIG. 10C shows an autocorrelation coefficient obtained from a reference wave centered around a circle in FIG.

  The phase calculation unit 224 determines, from the waveform shown in FIG. 10B or FIG. 10C, the number of motion rhythm cycles or the motion rhythm at time t-t1 when the phase change amount of the respiratory rhythm is 360 degrees. The number of peaks is calculated, that is, the phase change amount is calculated. Then, the ratio calculation unit 234 calculates the cycle ratio. That is, the phase calculation unit 224 calculates the movement rhythm phase change amount at time t-t1. Then, the ratio calculation unit 234 calculates the cycle ratio by dividing the movement rhythm phase change amount by the respiratory rhythm phase change amount (360 degrees) at time t-t1 (in the example of FIG. 10, the cycle ratio is 5).

  Next, with respect to the data illustrated in FIG. 3A, the result of an example in which the cycle ratio is calculated every time the respiratory rhythm phase increases by 3.6 degrees using the Hilbert transform method is shown by the solid line in FIG. When this is converted into a time interval, the cycle ratio is obtained every 0.02 to 0.04 seconds. In the case of FIG. 11, the cycle ratio is almost constant at 5, and it can be said that the synchrony between respiration and movement is high.

Further, with respect to the data illustrated in FIG. 3A, the result of obtaining the cycle ratio every time interval 0.07 seconds using the pattern matching method is shown by a broken line in FIG. From FIG. 11, it can be seen that results having the same tendency as the results obtained by the Hilbert transform method are obtained.
Next, details of the determination unit 215, that is, detailed operations of step A5 in FIG. 8 will be described with reference to a flowchart (steps A51 to A54: determination process) shown in FIG.

  First, the determination unit 215 determines whether or not the cycle ratio determined by the ratio determination unit 214 is within a predetermined range (step A51). If the cycle ratio is within a predetermined range (see Yes route in step A51), it is determined whether or not a predetermined time has elapsed (step A52). If the predetermined time has elapsed (see the Yes route in step A52), the determination unit 215 determines that the synchrony is high (step A53). That is, the determination unit 215 determines that the synchrony is high, for example, when the cycle ratio is a value within a predetermined range over a predetermined time. On the other hand, if the cycle ratio is not within the predetermined range (see No route in step A51), it is determined that the synchrony is low (step A54). That is, steps A51 to A54 are an example of a determination process for determining whether the ratio of the phase change amounts is within a predetermined range over a predetermined time.

[B] Description of real-time processing In order to be able to apply the above-described method, it is necessary to use data corresponding to a time (in many cases, for example, 3 seconds or more) in which the phase of the respiratory rhythm changes at least by one cycle. Therefore, it is suitable for performing signal processing on a long-time body motion signal measured in advance.
On the other hand, in order to perform this processing in real time, it is necessary to obtain the cycle ratio from shorter time data. Therefore, when performing real-time processing (analysis), for example, a small time width Δt of 3 seconds or less is set, and the phase change ΔθB of the respiratory rhythm and the movement rhythm between time t and time t + Δt are set. The phase change ΔθG is obtained, and the cycle ratio is approximately calculated from ΔθG / ΔθB. Specifically, for example, the phase calculation unit 224 calculates the phase change amount of the respiratory rhythm and the exercise rhythm at time t−t + Δt using Hilbert transform, pattern matching, etc., and the ratio calculation unit 234 A ratio between the calculated phase change amount of the respiratory rhythm and the phase change amount of the motion rhythm is obtained.

  For real-time processing, it is necessary to perform analysis using data measured and accumulated up to the current time. The problem is the effect of data boundaries. In other words, when signal processing such as spectrum analysis or Hilbert transform is performed, the error increases near the data end points (measurement start point and end point). That is, the cycle ratio at the current time may include an error based on such signal processing. In order to estimate this, the time t1 (16 seconds) is assumed to be the current time for the acceleration data illustrated in FIG. 3A, and only the data up to this time is used, and the time width Δt = 1.11 seconds is used. An approximate cycle ratio was calculated from the phase change. A result is shown with the broken line of FIG. Note that the solid line in FIG. 13 is the same as the solid line in FIG. 10 and is a result of an accurate calculation method using the entire data. When these are compared, the difference between the two is large at time t1. However, a value with almost no error is obtained at time t2 which goes back about 1.5 seconds from time t1. Therefore, an accurate cycle ratio is required with a time delay of less than 2 seconds. This is sufficient performance for practical real-time processing.

  When performing real-time processing, for example, a time constant T may be set in advance for the information processing apparatus 20. Even if this time constant T is not determined from the body motion signal, although there are some differences in respiratory rhythm and exercise rhythm depending on the subject, it is not very different, so it can be determined empirically as a general value. Is possible. Further, the time constant T may be calculated in real time from the body motion signal by the method described above.

  As described above, according to the system 1 in the example of the present embodiment, the respiratory rhythm can be accurately extracted from the body motion signal obtained from the repetitive rhythm movement accompanying the voluntary movement of the subject. That is, according to the system 1 of the present application, the respiratory rhythm can be accurately extracted from the body motion signal obtained during the voluntary exercise that is not in a resting state. In other words, according to the system 1 of the present application, it is possible to accurately extract the respiratory rhythm even when not in a resting state.

  Moreover, according to the system 1 in the example of the present embodiment, the ratio (for example, cycle ratio) between the phase change of the respiratory rhythm and the phase change of the exercise rhythm in a predetermined time is calculated by paying attention to the phase. The cycle ratio and the like can be accurately calculated from data having no peak. That is, according to the system 1 in the example of the present embodiment, since the focus is on the phase, the cycle ratio can be accurately obtained without paying attention to the peak of the respiratory rhythm and the motion rhythm extracted from the body motion signal. Can do.

Furthermore, according to the system 1 in the example of the present embodiment, the cycle ratio can be accurately calculated, so that the synchrony can be accurately evaluated.
In addition, according to the system 1 in the example of the present embodiment, by using pattern matching, the waveform after autocorrelation has less noise than the original body motion signal and thus has a clear peak. Since the position can be specified accurately, the cycle ratio and the like can be obtained with high accuracy.

Furthermore, according to the system 1 in the example of the present embodiment, real-time processing can be performed accurately.
[C] Others The disclosed technique is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present embodiment.

  For example, in the example of the present embodiment, in FIG. 3 (A), only the acceleration change in the traveling direction (front-rear direction) of the three axes is shown, and based on this signal, the respiratory rhythm and the exercise rhythm are extracted, Although the cycle ratio is calculated, it is not limited to this. For example, when measuring a multi-dimensional signal using a three-axis sensor or the like, the waveform for extracting the respiratory rhythm and the exercise rhythm may be the same signal or different waveforms. Specifically, for example, the respiratory rhythm may be extracted from an acceleration signal in the front-rear direction (front-rear direction), and the exercise rhythm may be extracted from an acceleration signal in the vertical direction. However, in the case of exercise such as walking and jogging, the rhythm cycle (stride cycle) in the left-right direction is about twice the rhythm cycle (step cycle) in the vertical direction and the front-back direction. When calculating the cycle ratio, a correct value can be obtained by doubling the calculated cycle ratio. The solid line in FIG. 14 shows the waveform when the cycle ratio is calculated as described above using the Hilbert transform from the acceleration signal in the left-right direction. In FIG. 14, the dotted line indicates the cycle ratio calculated using the Hilbert transform from the acceleration in the traveling direction (front-rear direction). It can be seen from FIG. 14 that both agree well.

  Further, in the example of the present embodiment, in the example of the present embodiment, in FIG. 3A, only the acceleration change in the traveling direction of the three axes is shown, and the respiratory rhythm and the motion rhythm are calculated based on this signal. Although extracted, it is not limited to this. For example, the time constant determination unit 211 performs the processes (1) to (4) on signals in two or more axes (for example, all three axes), and the minimum point corresponding to the respiratory rhythm in the process (4). It is also possible to select a signal that minimizes the variance value. For example, when the time constant determination unit 211 obtains the characteristics shown in FIG. 15 by performing the processes (1) to (4) on the signals in all three axes, the time constant determination unit 211 The time constant is determined by using the signal. In FIG. 15, the signal in the vertical and horizontal directions has a clear minimum point in the vicinity of the filter time (time constant) of 1.5 seconds, but no minimum point is observed in the signal in the front-rear direction. In such a case, the time constant determination unit 211 does not use a signal in the left-right direction to determine the time constant. That is, the time constant determination unit 211 extracts an optimal signal for extracting a respiratory rhythm from signals in a plurality of axial directions.

For example, in the example of the present embodiment, only the acceleration change in the traveling direction (front-rear direction) of the three axes is shown in FIG. 3A, and the cycle ratio is changed based on this signal as shown in FIG. Although it is calculated, it is not limited to this. For example, the cycle ratio may be calculated using acceleration in the vertical direction or the horizontal direction.
Furthermore, in the example of the present embodiment, the body motion signal detection device 10 does not include the central processing unit 21, but is not limited thereto, and the body motion signal detection device 10 includes the central processing unit 21. Also good. In this case, the central processing unit 21 executes an information processing program stored in a storage device (for example, the storage unit 12) in the body motion signal detection device 10 or a storage unit (not shown) outside the body motion signal detection device 10. Thus, the above-described function is exhibited. That is, in this case, the body motion signal detection device 10 also functions as an information processing device.

  In the example of the present embodiment, the body motion signal detection device 10 includes the body motion signal detection unit 11 and the storage unit 12, but is not limited thereto. For example, the body motion signal detection device 10 may include the body motion signal detection unit 11 but may not include the storage unit 12. In this case, for example, the body motion signal detection device 10 and the storage unit 12 are connected by wire or wirelessly, and the body motion signal detected by the body motion signal detection unit 11 is transferred to the storage unit 12 via wire or wirelessly. Stored. In this case, the storage unit 12 and the information processing device 20 are connected by wire or wirelessly, and the information processing device 20 acquires a body motion signal from the storage unit 12.

  Further, when the body motion signal detection device 10 and the information processing device 20 (or the storage unit 12) are connected via wireless, the body motion signal detection device 10 detects the body detected by the body motion signal detection unit 11. It has a function of transmitting a moving signal to the information processing apparatus 20 (or the storage unit 12) via the interface unit 13 that is an antenna. This transmission function is realized by executing a program stored in a storage unit (not shown) by a processing unit (not shown) provided by the body motion signal detection device 10 such as a central processing unit.

  In one example of the present embodiment, the information processing apparatus 20 determines the first time constant and the second time constant from the body motion signal, and calculates the respiratory rhythm and the exercise rhythm based on the first time constant and the second time constant. Although extracted, it is not limited to this. For example, the frequency characteristics of the body motion signal are obtained by spectrum analysis such as FFT (Fast Fourier Transform) or wavelet analysis, and the frequency range of respiratory rhythm and motion rhythm is specified from the result of the spectrum analysis. Then, for example, a respiratory rhythm and an exercise rhythm may be extracted by applying a bandpass filter corresponding to each frequency band to the body motion signal.

In addition, the body motion signal is decomposed into each mode waveform (Intrinsic Mode Function) by EMD (Empirical Mode Decomposition) and EEMD (Ensemble Empirical Mode Decomposition), and from this result, mode waveforms corresponding to respiratory rhythm and motion rhythm (for example, A strong waveform) may be selected.
Further, in an example of the present embodiment, after the time constant determination unit 211 determines the time constant, the respiratory rhythm extraction unit 212 and the exercise rhythm extraction unit 213 respectively filter the respiratory rhythm and the body motion rhythm from the body motion signal by filtering processing. Although extracted, it is not limited to this. For example, in the process (3), all output waveforms Y when the time constant T is changed are stored in the storage unit 22. Then, the respiratory rhythm extraction unit 212 and the exercise rhythm extraction unit 213 respectively select waveforms corresponding to the first time constant and the second time constant determined in the process (4) from all the output waveforms Y. It is good to do.

In the example of the present embodiment, the respiratory rhythm extraction unit 212 and the exercise rhythm extraction unit 213 each extract the respiratory rhythm and the exercise rhythm from the body motion signal. However, the present invention is not limited to this. Only the respiratory rhythm may be extracted from the body motion signal by the extraction unit 212.
Furthermore, in an example of the present embodiment, two local minimum points having a small CV value are obtained from the change in CV with respect to the change in the time constant T, and the time constant T corresponding to the local minimum point having the smaller time constant T is determined as the first constant. The time constant T corresponding to the minimum point with the larger time constant T or time constant T is determined as the first time constant, but is not limited to this. For example, the time constant corresponding to the local minimum point itself may not be the first time constant or the second time constant, but the time constant near the local minimum point may be the first time constant or the second time constant. For example, in the example of FIG. 5, for example, the first time constant may be 1.5 and the second time constant may be 0.4.

An example of this embodiment is applicable not only to people but also to animals such as pets, livestock, and horses. For example, since it is known that a horse is synchronized with one complete step and one breath at the time of gallop, using an example of this embodiment makes it possible to manage the health of the horse, for example.
Furthermore, in the example of the present embodiment, the respiratory rhythm and the exercise rhythm are extracted from the body motion signal using the two time constants determined by the time constant determination unit 211, but the present invention is not limited to this. For example, the time constant determination unit 211 determines a time constant of 3 or more, and the exercise rhythm extraction unit 212 and the respiratory rhythm extraction unit 213 extract 3 or more rhythms from the body motion signal using the time constant of 3 or more. Also good. For example, the subject may extract another rhythm in addition to the respiratory rhythm and the exercise rhythm from the body motion signal. For example, three rhythms of a respiratory rhythm, a walking rhythm, and a juggling rhythm during walking may be extracted from a body motion signal detected when the subject juggles while walking. In this case, the time constant determining unit 211 has three local minimum CV values from the change in CV with respect to the time constant T obtained from the output when the high-pass filter or the band-pass filter is applied to the body motion signal. Select a point. For example, when it is assumed that the cycle of the respiratory rhythm is the longest, the cycle of the juggling rhythm during walking is long, and the cycle of the walking rhythm is shortest, the time constant determination unit 211 obtains The time constant T corresponding to the minimum point having the largest time constant T among the three minimum points obtained is set as the time constant for extracting the respiratory rhythm from the body motion signal, and the next of the obtained minimum points is A time constant T corresponding to a minimum point having a large time constant T is determined as a time constant for separating and extracting the rhythm of juggling during walking from the body motion signal. Further, the time constant determination unit 211 determines the time constant T corresponding to the minimum point having the smallest time constant T among the obtained minimum points as a time constant for extracting the walking rhythm separately from the body motion signal. To do. Then, using the three time constants determined by the time constant determination unit 211, the respiratory rhythm extraction unit 212 and the exercise rhythm extraction unit 213 determine the respiratory rhythm, the exercise rhythm, and the juggling rhythm during walking from the body motion signal. Extract. Specifically, for example, the respiratory rhythm extraction unit 212 performs the processes (5) to (9) using a time constant for separating and extracting the respiratory rhythm from the body motion signal, thereby performing the body motion signal. Extract respiratory rhythm from Further, the respiratory rhythm extraction unit 212 performs the processes (5) to (9) by using the time constant for separating and extracting the juggling rhythm during walking from the body movement signal. To extract the rhythm of juggling while walking. Further, the exercise rhythm extraction unit 213 extracts the walking rhythm from the body motion signal by performing the processes (10) to (12) using the time constant for separating and extracting the walking rhythm from the body motion signal. To do.

  Various application programs for realizing each function of the central processing unit 21 provided in the information processing apparatus 10 are, for example, a flexible disk, a CD (CD-ROM, CD-R, CD-RW, etc.), a DVD (DVD). -ROM, DVD-RAM, DVD-R, DVD + R, DVD-RW, DVD + RW, HD DVD, etc.), recorded in a computer-readable recording medium such as a Blu-ray disc, magnetic disc, optical disc, magneto-optical disc, etc. Provided. Then, the computer reads the program from the recording medium, transfers it to the internal storage device or the external storage device, and uses it. The program may be recorded in a storage device (recording medium) such as a magnetic disk, an optical disk, or a magneto-optical disk, and provided from the storage device to the computer via a communication path.

  Various application programs for realizing the functions of a central processing unit (not shown) provided in the body motion signal detection apparatus 10 are, for example, a flexible disk, a CD (CD-ROM, CD-R, CD-RW, etc.), Recorded on a computer-readable recording medium such as a DVD (DVD-ROM, DVD-RAM, DVD-R, DVD + R, DVD-RW, DVD + RW, HD DVD, etc.), Blu-ray disc, magnetic disc, optical disc, magneto-optical disc, etc. Provided in different forms. Then, the computer reads the program from the recording medium, transfers it to the internal storage device or the external storage device, and uses it. The program may be recorded in a storage device (recording medium) such as a magnetic disk, an optical disk, or a magneto-optical disk, and provided from the storage device to the computer via a communication path.

  EXAMPLES Hereinafter, although an Example demonstrates this invention further in detail, this invention is not limited to a following example, unless it deviates from the summary.

As the body motion signal detection device 10, an acceleration recorder “Watching Gate” manufactured by Mitsubishi Chemical Corporation was put on a dedicated belt and wound around the abdomen of the subject, and body motion signals during walking and jogging on the treadmill were sampled at 100 Hz. The recorder (body motion signal detection device 10) was positioned at the center of the abdomen.
Exercise was performed continuously in the following order.

・ Walking gradually increasing speed from 4km / hour (4km / hr) to 6km / hour (6km / hr).
・ After jogging to gradually increase the speed from 6km / hr to 10km / hr, jogging to decrease again to 6km / hr.
-Walking at 6km / hr after walking gradually reducing the speed from 6km / hr to 4km / hr.

・ Jogging at a speed of 7.5km / hr. Repeatedly rhythmic breathing, inhaling twice and exhaling twice immediately after that, breathing once deeply and exhaling deeply once in the middle.
Of the measured data, the phase of the longitudinal acceleration signal was obtained by the Hilbert transform method, and the cycle ratio between the respiratory rhythm and the exercise rhythm was calculated. A result is shown in FIG. 16 with the time change of the step period (time interval of every step). Corresponding to the speed of walking and jogging, the step period also changes. The cycle ratio gradually decreases from the start of exercise until about 10:40 (denoted as 10:40 in the figure), but after that, it is stable at around 4. On the way, it is increased to 7 where deep breathing is performed. FIG. 17 is an enlarged view of a part of FIG. FIG. 17 shows that the cycle ratio is constant at the rhythmic breathing location where two inhalations and two exhalations occur, and the synchrony between the respiratory rhythm and the jogging rhythm is very good.

  In this embodiment, the evaluation of the synchrony between the respiratory rhythm and the exercise rhythm during exercise is performed quantitatively, continuously and accurately.

A 3-axis gyro sensor MP-G3-01B manufactured by Microstone as the body motion signal detection device 10 is installed in the center of the subject's abdomen, and the signal when walking around the room (with obstacles such as desks and chairs) is 100 Hz. Measured with. The subjects walked while being aware of breathing once every 6 steps and walking while being aware of breathing once every 4 steps.
Rhythms due to walking strides were extracted from the angular velocity signals around the vertical axis by the following process.
1. Perform high-pass filtering using a time constant of 0.88.
2. Integration is performed once on the high-pass filtered signal.
3. Again, high-pass filtering is performed using a time constant of 0.88.
4). Perform low-pass filtering with a time constant of 0.35.

Furthermore, it extracted from the angular velocity signal around the axis in the front-rear direction by processing below the respiratory rhythm.
1. High-pass filtering using time constant 2.55.
2. Integration is performed once on the high-pass filtered signal.
3. High-pass filtering using time constant 2.55.
4). Low-pass filtering using time constant 1.02.
5. A signal composed of an average of two envelopes connecting the maximum value and the minimum value of the signal after low-pass filtering is subtracted from the signal after low-pass filtering.

Phase information was obtained by autocorrelation using a 0.51 second reference wave for walking, and a 2.55 second reference wave for breathing, and the cycle ratio was calculated. The results are shown in FIG.
The cycle ratio fluctuates as small as 6 when breathing once every 6 steps and 4 when breathing once every 4 steps, and the synchrony between breathing and walking is well understood. I understand that.

DESCRIPTION OF SYMBOLS 1 System 10 Body motion signal detection apparatus 11 Body motion signal detection part 12 Interface part 20 Information processing apparatus 21 Central processing part 22 Storage part 23 Output part 24 Interface part 211 Time constant determination part 212 Respiration rhythm extraction part 213 Motion rhythm extraction part 214 Ratio determination unit 216 Determination unit 224 Phase calculation unit 234 Ratio calculation unit

Claims (23)

An extraction process for extracting a respiratory rhythm of the living body from the body movement signal by filtering a signal based on a repetitive rhythmic movement accompanying the voluntary movement of the living body (hereinafter referred to as a body movement signal) is provided. Information processing method. The extraction process performs filtering on the body motion signal using a time constant (hereinafter referred to as a first time constant) determined based on an output when a bandpass filter is applied to the body motion signal. The information processing method according to claim 1, wherein the information processing method is a process of extracting a respiratory rhythm of the living body. In addition to extracting the respiratory rhythm of the living body in the extraction process, a time constant different from the first time constant determined based on an output when a band pass filter is applied to the body motion signal 3. The information processing according to claim 2, wherein the rhythm of the voluntary movement is extracted by filtering the body motion signal using a second time constant (hereinafter referred to as a second time constant). Method. The time constant determination process which determines a said 1st time constant based on the output at the time of applying a band pass filter with respect to the said body motion signal is characterized by the above-mentioned. Information processing method described in 1.   The time constant determining step is a step of determining the second time constant based on an output when a band pass filter is applied to the body motion signal in addition to the determination of the first time constant. The information processing method according to claim 4, wherein: 6. The information processing method according to claim 4, wherein the first time constant is larger than the second time constant. A phase calculation process for obtaining a change amount of a phase in a certain time for each of the respiratory rhythm and the voluntary movement rhythm extracted by the extraction process;
A ratio calculating process for calculating a ratio between the phase change amount of the respiratory rhythm and the phase change amount of the rhythm based on the rhythm movement;
The information processing method according to claim 4, further comprising a determination step of determining whether the ratio of the phase change amounts is within a predetermined range over a predetermined time. . The information processing method according to claim 1, wherein the body motion signal is measured by an inertial sensor. The information processing method according to claim 1, wherein the voluntary movement is any one of walking, jogging, and running. A body motion signal detection unit that detects a signal based on a repetitive rhythmic motion accompanying a voluntary motion of a living body as a body motion signal;
An information processing system comprising: a rhythm extraction unit that extracts the respiratory rhythm from the body motion signal by performing filtering processing on the body motion signal.   The rhythm extraction unit uses a time constant (hereinafter referred to as a first time constant) determined based on an output when a bandpass filter is applied to the body motion signal, to the body motion signal. The information processing system according to claim 10, wherein a respiratory rhythm of the living body is extracted by performing a filtering process. When the rhythm extraction unit is different from the first time constant determined based on the output when applying a bandpass filter to the body motion signal in addition to extracting the respiratory rhythm of the living body The information processing system according to claim 11, wherein a rhythm of the voluntary movement is extracted by performing a filtering process on the body motion signal using a constant (hereinafter referred to as a second time constant). 13. A time constant determination unit that determines the first time constant based on an output when a band-pass filter is applied to the body motion signal is provided. Information processing system described in 1. The time constant determining unit determines the second time constant based on an output when a band pass filter is applied to the body motion signal in addition to the determination of the first time constant. The information processing system according to claim 13. The information processing system according to claim 10, further comprising an output unit that outputs a result extracted by the rhythm extraction unit. A rhythm extraction unit that extracts a respiratory rhythm of the living body from the body motion signal by performing filtering processing on a signal (hereinafter referred to as a body movement signal) based on a repetitive rhythm movement accompanying the voluntary movement of the living body; An information processing apparatus is characterized.   The rhythm extraction unit uses a time constant (hereinafter referred to as a first time constant) determined based on an output when a bandpass filter is applied to the body motion signal, to the body motion signal. The information processing apparatus according to claim 16, wherein a respiratory rhythm of the living body is extracted by performing a filtering process. When the rhythm extraction unit is different from the first time constant determined based on the output when applying a bandpass filter to the body motion signal in addition to extracting the respiratory rhythm of the living body The information processing apparatus according to claim 17, wherein a rhythm of the voluntary movement is extracted by performing a filtering process on the body motion signal using a constant (hereinafter, referred to as a second time constant). 19. A time constant determining unit that determines the first time constant based on an output when a band-pass filter is applied to the body motion signal is provided. The information processing apparatus described in 1. The time constant determining unit determines the second time constant based on an output when a band pass filter is applied to the body motion signal in addition to the determination of the first time constant. The information processing apparatus according to claim 19.   17. A display device comprising an output unit for outputting the result extracted by the information processing device according to claim 16. By causing a computer to execute an extraction process for extracting the respiratory rhythm of the living body from the body motion signal by performing a filtering process on a signal (hereinafter referred to as a body motion signal) based on a repetitive rhythm movement accompanying the voluntary movement of the living body. An information processing program characterized by the above. By causing a computer to execute an extraction process for extracting the respiratory rhythm of the living body from the body motion signal by performing a filtering process on a signal (hereinafter referred to as a body motion signal ) based on a repetitive rhythm movement accompanying the voluntary movement of the living body. A computer-readable recording medium on which an information processing program is recorded.

Images