JP2019181115A - Walking cycle variation measurement device - Google Patents

Abstract

To eliminate the need for a device fixation operation of equipment and a measurement assistant, achieve high precision measurement under the same measurement condition while being a simple measurement act so that an elderly person may repeat and continue by oneself on a daily basis, and enable a walking cycle variation coefficient being an important indicator of the syndrome at old age and the evaluation of the aging variation.SOLUTION: (1) Walking start is automatically determined to eliminate failure of measurement, and the measurement is made in the constant step count just after an acceleration period and measurement conditions are aligned to enable comparison with the standard and temporal comparison. (2) A measuring apparatus is contained in a pocket of the garment coming in contact with a thigh without being fixedly worn in a body. (3) Gravity direction acceleration is calculated from a value of an acceleration sensor of the measuring apparatus. (4) The time of arrival of an impact of the heel grounding that spreads without going by way of a hip joint is detected from the transitivity and high precision measurement of the walking cycle variation coefficient is achieved. (5) The measurement results capable of recognizing repeatability are extracted from multiple measurement results of the same timing, compared with the standard, or evaluated by catching a sequential change.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a method for measuring a walking cycle variation coefficient (hereinafter abbreviated as CV) that enables easy and stable high-accuracy measurement for the purpose of grasping changes with time for elderly people.

  Various types of sensors such as an acceleration sensor are mounted on smartphones that are already widely used.

CV is an index representing a sense of rhythm of walking, and is defined as the ratio of the standard deviation of the difference between the average walking cycle and each walking cycle to the average walking cycle when walking on a flat road surface at a constant speed.
A healthy person may have a very low CV of less than 1%. In existing studies, normal healthy adults: 1.6 ± 0.5%, normal elderly: 2.1 ± 0.8%, orthopedic patients: 2.8 ± 1.1%, cerebral infarction / cerebral hemorrhage patients Self-supporting group: 3.0 ± 1.7% and the same monitoring and assistance group 10.4 ± 10.0%.
CV is known to be an important risk evaluation index for various geriatric syndromes such as a fall risk, and becomes larger when it becomes weak at an old age or when cognitive function is lowered.
It is also affected by illness and may improve as a result of some kind of intervention.
Although it is an important index for elderly people, until now there has been no means for elderly people to easily measure on a daily basis, so most elderly people do not know their CV.

In order to know the state change of the person to be measured as described above using CV as an index, the measurement error is required to be sufficiently small, and is preferably less than 0.5% if possible.
The walking cycle in normal walking is about 900 to 1,350 milliseconds inside and outside for 1 second. This 0.5% is 4.5 to 6.8 milliseconds, and even 1% is 9.0 to 13.5 milliseconds. The measurement error of the walking cycle is suppressed to a level of several milliseconds, and the accuracy is stable. If measurement cannot be performed (hereinafter simply referred to as high-precision measurement), the improvement or deterioration of CV cannot be accurately evaluated. In addition, if measurement assistants, measurement equipment (hereinafter simply abbreviated as “terminal”), and fixing tools, etc. are required for measurement, tools for securing the elderly, It becomes a fatal obstacle.

The force propagation speed refers to the speed at which stress or strain when an impact force is applied to an object is transmitted as a stress wave in the object, and varies greatly depending on physical properties such as the rigidity of the object.
The fact that the velocity is present means that it takes time to propagate the stress wave, but the propagation velocity varies depending on the condition of the human body that is the propagation path. The impact of heel contact during walking propagates to the terminal through highly rigid bones and soft joints. Although sufficient data is not yet available for the propagation speed at this time, it is often the case that the propagation time is comparable or longer than that of the previous value of 4.5 to 13.5 milliseconds. .
In particular, when the terminal is held on the waist or thigh, the influence is not small because it passes through several joints having low rigidity. When the terminal is stored in a pocket in contact with the thigh, the impact on the same side of the heel passes through the ankle and knee joints, but the impact on the opposite side of the heel is in addition to the ankle and knee joints. Come through the hip joint. For this reason, the propagation time of the impact between the left and right eyelids is greatly different, and the inventors have verified that there is a difference of 100 milliseconds or more. Further, when passing through a number of joints, the shock is buffered at the joints, the waveform becomes gentle, the detection accuracy of the peak time is lowered, and the measurement error is easily increased.
In consideration of the above situation, the present invention eliminates the need for a fixing operation to the body and is a simple operation on the premise of the absence of a measurement assistant. Enables accuracy measurement. In addition, measurement conditions are aligned to enable analysis and evaluation of changes over time.
Note that the high-accuracy measurement referred to here is also described in the section 0004, but it is possible to stably maintain measurement accuracy sufficient to analyze the temporal change of the walking cycle fluctuation.

When wearing the terminal in the center of the abdomen, it is fixed firmly to prevent the displacement of the wearing position, so it is easy to receive a large reaction, and when the belt is loosely tightened, the reaction is mixed and the peak can not be properly captured Arise. As a result, the gait cycle fluctuation may be calculated to be larger than actual, and it is difficult for an elderly person who has started to weaken to realize stable high-accuracy measurement alone.
Since it is difficult to obtain a stable measurement result unless the terminal is firmly fixed to the body, until now, a method has been adopted in which a wearing position is determined and fixed with a dedicated belt for measurement.

  Conventional techniques have been developed to measure the CV of the elderly, primarily for research purposes. Therefore, since it is not based on the viewpoint that an elderly person uses it daily and grasps a time-dependent change of own CV, it does not have the requirements necessary for the purpose of the present invention. Specifically, familiar measurement means as requirements for the elderly to continue to measure, easy measurement operation and few measurement failures, level of measurement accuracy required to see changes over time There was no requirement to determine the accuracy and ensure stable accuracy. The present invention is intended to contribute to the maintenance and improvement of walking ability by satisfying these requirements and enabling an elderly person to grasp the temporal change of his / her CV in daily life.

The walking cycle measures the time interval of the load response period of walking (the ground contact period) using some sensor such as an acceleration sensor or a pressure sensor.
In Japanese Patent Application Laid-Open No. 2015-177925 (walking support device, gait measuring device, method and program, Nippon Telegraph and Telephone Corporation, hereinafter referred to as patent document), “using either one of three-axis acceleration data and angular velocity data, A half-walking period (time required for one step) that represents the time from when one foot of a pedestrian has landed until the other foot has landed is calculated. Based on the calculated half-walking cycle, the walking cycle fluctuation rate is calculated. The method of “calculating” is described. The purpose of the patent document is to “provide a walking support device and method and program that allow a pedestrian to improve the left-right balance during walking without the assistance of an expert”, and the measurement of walking cycle fluctuation is It is positioned as that process.
In the first embodiment of the specification of the patent document, a smartphone for acquiring sensor data is accommodated in a side pocket of trousers or attached to a waist belt, and the same thigh as the accommodation position of the present invention For example, an angular velocity sensor is selected when installed at a position close to the extremities, and an acceleration sensor is selected when installed at a position close to the trunk such as the waist.
In the inventor's measurement test, the left and right half-walking cycles are approximately the same time interval when worn in the waist, especially in the center of the abdomen, but the half-walking cycle measured is maximum when stored in the side pockets of the pants. It was found that about twice as much opening occurred. In order to avoid such measurement results, it is inferred in the patent literature that the angular velocity sensor is selected when stored in a side pocket of trousers. In addition, it has been found that the measurement accuracy is often very poor unless it is firmly fixed with some tool when it is attached to the center of the abdomen.
In the patent document, when detecting a walking cycle, a scalar value that is a vector length is used instead of a vector of the triaxial acceleration data itself. In the detection at, the gravity direction and the horizontal direction cannot be discriminated, and the reaction other than the gravity direction is also observed, so there is a concern that the measurement error of the walking cycle may be increased by mistaking the peak. Not suitable when accuracy is required.
There is also a statement that “the timing of contact with the ground is equal to the timing at which the peak value of acceleration or the extreme value of the angular velocity is integrated.” In reality, however, it takes time to propagate the impact during the ground contact period to the measuring device. Therefore, when viewed in milliseconds, there is a time lag between the ground contact period and the peak measurement time of the sensor acceleration. When looking at the extreme value of the integrated value of angular acceleration, that is, the peak of the posture angle of the terminal, the heel is grounded at the timing when it is first lifted and then lowered, and the living body bends (becomes), so the propagation of moment However, there is a time lag, and the timing is not the same when viewed in milliseconds. The generated time lag is not always constant, and becomes an increase factor of measurement error.
Therefore, although some CV value is calculated by the method of the patent document, high-precision CV measurement which is the object of the present invention cannot be realized. Measurement accuracy is different between acceleration measurement and angular velocity measurement, and it is necessary to unify them in order to evaluate the temporal change of the present invention.
In a walking cycle of 1 second (1,000 milliseconds), the 1% measurement error is only 10 milliseconds.
That is, in order to realize the measurement accuracy targeted by the present invention, a measurement method based on the difference in the measurement result due to the difference in the holding position of the terminal is required.
In addition, when wearing / fixing to the body, unless proper consideration is given to the mounting position and fixing method, the reaction time decays slowly, so the peak time is selected incorrectly, and the measurement error of the walking cycle exceeds the allowable range. It should be noted that there are many cases where this is exceeded. Furthermore, this fixing operation is troublesome for the elderly to repeatedly measure on a daily basis, and becomes a fatal obstacle to continued use.
The second embodiment of the patent document is omitted because it is not relevant.

Due to the above-mentioned problems in practical use, the CV measuring devices that have been commercialized at the present time are either floor-mounted using pressure sensors or optical sensors, or a portable terminal is fixed to a belt wrapped around the waist This is the type to measure. There are LSI Medience Corporation, Gate MG-M1110-HW of Mitsubishi Chemical Corporation, and Quality of Digital Standard Corporation.
The floor-standing type is not suitable for daily use by the elderly because the equipment is large and expensive and requires a measurer. All types that are fixed with a belt require a dedicated belt and a motion to wrap around the waist, which is not easy. Moreover, it is not suitable for the purpose of the present invention in terms of intervention of a measurer and measurement accuracy.
In some cases, a device for fixing a terminal to an ankle is used for an experiment.

The value of the acceleration sensor at the time of walking tends to draw a waveform close to a sine wave as in the example of FIG. 3-3, but actually varies greatly depending on the holding position and holding method of the measuring device.
When held in the center of the abdomen, it shows a sine wave waveform with a half-walking period as the basic cycle as shown in Fig. 3-3, but when stored in one side pocket of the pants in contact with the thigh Although the propagation of the impact on the other side of the heel contact is observed in the unit of one walking cycle as shown in 3-1, the time width is considerably shifted. When held on either the left or right flank, a waveform similar to that shown in FIG. The waveform of the observation results greatly differs depending on whether the terminal is held on the thigh side or the pelvis side across the hip joint.
The degree of impact received by the terminal varies depending on the hardness of the road surface and shoes, and the impact that propagates as the holding position moves away from the bag is buffered, and the value of the acceleration sensor varies greatly depending on how the terminal is held.

Means for solving the problem

In the present invention, the terminal is housed in a position in contact with the leg portion such as a pocket of a trouser, and measurement is performed without using a wearing tool such as a belt, and thus without fixing.
In this case, the terminal continues to move as if dancing in the pocket as it walks, during which time it frequently hits the pocket wall and receives a reaction, and the terminal's posture continues to change. As a result, it tends to be thought that the measurement accuracy deteriorates, but this is contrary to intuition, but in the inventor's test, this actually realized the most stable high-accuracy measurement. The point contrary to this intuition is considered to be one of the factors for which the contents of the present invention have not been filed so far.
The wall of the pocket is made of a soft cloth, and in many cases, the pocket itself is also free to move. Even if it collides with a soft cloth that has a high degree of freedom of movement, the recoil received by the terminal is immediately attenuated, so that only a large impact during the heel-contact period can be extracted clearly. Moreover, although it continues to move in the pocket, there is no possibility that a large position shift will occur.
A number of measurements were performed with a testing machine, and it was verified that the same waveform was generated in all measurement times, and CV could be measured stably with high accuracy. FIG. 4 shows a part of the measurement example.
Paradoxically, when the terminal is fixed at the center of the abdomen, it is necessary to fix it with a relatively hard material, so the reaction becomes large, and it is understood that if the fixing method is wrong, the measurement error is likely to increase. It was.

However, when measuring in a pocket, another problem that needs to be solved arises.
Unlike the measurement mode in which a hand can reach the terminal and operate and immediately walk, the measurement button operation (or measurement application activation operation) is performed with the terminal in hand, and then the terminal is placed in the pocket. Then follow the procedure of starting walking and measuring.
Although it can be seen that the measurement operation is actually performed, there are many cases where the terminal is caught on the edge of the pocket or the upper end of the inner pocket in the pocket, and it takes an unexpectedly long time to store. In particular, elderly people often take time. This is especially true when implementing double tasks, but it is also possible to spend time on mental preparation. For this reason, it is not possible to determine in advance the time from the measurement operation of the terminal until the terminal is put into the pocket and the stable walking motion is started.
If the time restriction until the terminal is stowed is severe, the subject will be impatient, and there may be a situation in which an erroneous measurement result is calculated by including a stop period or an acceleration period at the start of walking in the measurement target time. Conversely, if the measurement start time is delayed by loosening the constraints, the elderly who quickly stored must walk longer and the measurement conditions are not met as described later, and the convenience is lost accordingly.
There is a problem of aligning measurement conditions. When comparing CV with a reference value or looking at changes over time, the measurement time or the number of steps needs to be the same. When the measurement time is 10 seconds, 20 seconds, and 30 seconds, the difficulty level for realizing a low CV is different. In particular, even if an elderly person achieves a low CV by walking for 10 seconds, it cannot always achieve the same CV value by walking for 30 seconds. Long walks make it difficult to maintain a walking rhythm. Therefore, it is necessary to make the measurement time or the number of measurement steps constant.
It is also necessary to align the measurement timing. If one person measured in the first 10 seconds after starting stable walking and another person started measuring in 10 seconds after 15 seconds had passed after entering stable walking, the same condition could not be said.
Stress that urges elderly users to store in their pockets hinders continued use. Assuming that the elderly can measure by themselves without stress in daily life, this problem is fatal and a method for automatically realizing measurement under the same conditions is required.

A method for calculating CV with high accuracy using a terminal including a three-axis acceleration sensor will be described.
(1) CV is measured by consciously walking rhythmically on a flat walking surface.
The terminal is stored in a pocket of clothing that is in contact with the thigh, and triaxial acceleration data for a certain time or a certain number of steps during walking is obtained. When storing, a special fixing device is not required, and the posture of the terminal is arbitrary.
Although the acceleration data acquisition time interval is generally constant, for example, it is not 21.5 milliseconds or 18.3 milliseconds. Nevertheless, it has been confirmed that there is no problem in calculating CV through the following algorithm.
Note that the length of this approximate time interval can be selected by a program.
In addition, a method of storing the terminal in a pocket in contact with another body part instead of the thigh is not excluded, but measurement accuracy and accuracy stability are inferior depending on the part in contact. In actual trials, when the measurement is performed with the west part of the center of the abdomen of the trousers being fixed without being fixed, the position is likely to be displaced, so the calculated accuracy is often greatly lowered and is not stable. While an ankle pocket is desirable, there is not much such clothing.
{Circle around (2)} The acceleration in the direction of gravity is calculated from the value of this three-axis acceleration sensor.
One method is to use a three-axis magnetic sensor together and convert the value of the three-axis acceleration sensor to the world coordinate system to obtain vertical acceleration. For example, the Android terminal has a getRotationMatrix () method and a remapCoordinateSystem () method, which calculates the acceleration vector in the gravitational direction by calculating the tilt of the terminal in the world coordinate system from the 3-axis acceleration data and 3-axis magnetic intensity data. it can. However, it should be noted that the sensor value includes a measurement error, and that the error is likely to increase when the calculation is complicated due to magnetic strength or the like.
Another simpler method is to replace the gravity vector with the average value of a sufficient number of three-axis acceleration sensors for several seconds or more and calculate the gravity direction vector component at each time from the inner product theorem. . In this case, although there is a slight difference from the original gravity vector, it has been confirmed that no problem occurs in the calculation.
Acceleration vector at time Ti: (Xi, Yi, Zi) Gravity vector: (Xave, Yave, Zave)
However, Xave, Yave, and Zave are average values of Xi, Yi, and Zi of the three axes,
Gravity vector length: Lg = √ (Xave ² + Yave ² + Zave ² ) ≈9.8 m / ss
Acceleration vector length at time Ti: Li = √ (Xi ² + Yi ² + Zi ² )
Gravity direction vector length at time Ti: Lgi = Li * cos θ = (Xave * Xi + Yave * Yi + Zave * Zi) / Lg
Li * cosθ is a positive value when the acceleration vector at time Ti is in the same direction as the gravity vector, and a negative value when the acceleration vector is in the opposite direction, so the terminal is stored upside down. Even so, the measured waveform is the same.
In order to calculate the gravity direction vector length, most of the recoil when the impacted terminal collides with the side pocket wall is excluded. In a similar calculation, instead of using the direction of gravity, instead of simply using the vector length of the acceleration vector obtained from the three-axis acceleration sensor at each time, there is a method that cannot be recommended because it causes a reduction in measurement accuracy.
Further, although there is a method of supplementing the measurement error of the acceleration sensor by using a low-pass filter depending on the invention, the low-pass filter has a large influence on the peak time of the waveform, so that the error can be easily enlarged and this is not recommended.
(3) From the time-series transition of the vector length in the gravitational direction (waveform drawn by the gravitational direction vector length), the peak is extracted as the ground contact period.
That is, the time when a large vertical reaction force is received is defined as the heel contact period (load response period).
The observed waveform is not a waveform like a sine wave, but generally a waveform as shown in FIG. 3-1, but the period during normal walking is 750 milliseconds or more and less than 1,500 milliseconds (walking cycle number 40 to 40). 80 / min, cadence 80-160 steps / min), so that each walking cycle can be specified.
This specific method may be arbitrary, but as one method there is a method of determining an approximate cycle time using a discrete Fourier transform and collating with an actual data waveform. When one peak is set as the starting point, the next peak exists in the vicinity after the above-mentioned approximate cycle time, and the clear peak in the peak is narrowed down to one, so that the walking cycle can be specified accurately.
Note that the average period estimated by the discrete Fourier transform is only an approximate one, but there is no problem such as a decrease in measurement accuracy in the subsequent processing, and a half-walking period is extracted instead of one walking period. Note that it is necessary to add the process.
In addition, an arbitrary device that further increases the measurement accuracy, for example, a device that corrects an error caused by a time interval of acceleration data acquisition may be added.
{Circle around (4)} The walking cycle is calculated from each of the extracted contact times, and the standard deviation of the difference from the average value is obtained based on the definition of CV, and the ratio to the average value is defined as CV.

Next, a method for automatically realizing measurement under the same conditions when the terminal is housed in a pocket for measurement will be described. The same measurement condition is a measurement at a fixed number of seconds or a fixed number of steps in a stable walking section at a constant speed after an acceleration period immediately after entering a walk from a stop.
For example, when measuring 10 seconds, the CV can be calculated only with an even number of steps. Therefore, if it is determined by the number of seconds, an expression such as “an even number of steps within 10 seconds” becomes accurate.
Since the problem as described in item 0012 occurs when the measurement is started after a uniform time has elapsed after the measurement button operation or the application activation operation, the following functions are provided that automatically correspond to the situation that has occurred.
(1) After starting the application or operating the measurement button of the application, the walking start operation is automatically detected from the value of the acceleration sensor.
Hold the terminal in a stopped state and start walking through the action of putting it in your pocket. In the meantime, the position and posture of the terminal change greatly, but the acceleration received by the terminal until the start of walking continues to a value close to 9.8 m / ss until the start of walking, excluding the impact that is received when the pocket is finished.
The acceleration vector length √ (Xi * Xi + Yi * Yi + Zi * Zi) is calculated.
(2) When walking begins, the acceleration vector length starts to show a large value periodically.
Once it shows a large value, if it continues to return to a small value again, it will continue to show a large value periodically at intervals of a fixed time (acceleration period) such as about 3 seconds (around 1 second), at most 1.5 seconds It can be determined that it has entered stable walking. The threshold value for this determination is arbitrary.
In steps (1) and (2), a vertical acceleration vector may be used instead of the acceleration vector length, or a step count measuring function such as STEP_DETECTOR or STEP_COUNTER may be used.
(3) Measure in a certain section (time or number of steps) after entering stable walking.
Inform the person being measured that the terminal can be slowly put into the pocket before starting to walk, and instruct him to walk a certain number of steps when he starts walking.
This fixed number of steps is a number of steps that slightly exceeds the number of steps to be added, the number of steps to be measured, the number of steps in the previous acceleration period, and the number of steps in the deceleration section from the end of measurement to the stop. Even if “fixed time” is instructed, the measured person cannot grasp the time unless he / she is looking at the clock.
Although the time from the operation to the end of walking cannot be determined in advance on the terminal side, it can be dealt with by a method such as determining whether or not the vehicle stops every second by collecting acceleration data with a sufficient time width.
As described above, although the terminal storage operation time in the pocket varies depending on the person and sometimes, measurement under the same conditions becomes possible.

In the measurement, the person to be measured consciously walks rhythmically, but due to the road surface condition or some other factor, there is a case where a large CV value is measured due to loss of rhythm. Conversely, even if a very good CV value is measured only once, it may be difficult to reproduce. As described above, if a high or low measurement value that happens to be generated is arbitrarily selected, an appropriate evaluation cannot be performed.
In order to see the change in CV over time, whether it is deterioration or improvement, it is desirable to compare the reproducible CV value of the measurement subject with the high measurement accuracy.
The method of extracting the reproducible CV value may be arbitrary, but for example, in a certain number of measurement values, the average value of the measurement values excluding the lowest CV value of the upper fixed number is taken. A way to do this is conceivable.
In this way, it can be seen that the stability of the measurement accuracy is important as well as the high measurement accuracy. The method of the present invention can stably maintain high accuracy as seen in the measurement example of FIG. 4 and can grasp changes with time.
The measured CV data is transmitted to and stored in the 20 analyzer together with the measurement time information, and changes over time are analyzed at regular intervals. Analysis and evaluation methods are arbitrary.

  Prepare two terminals, store them in the pockets on both sides, and measure them simultaneously. If the left and right walking balances are equal, the two graphs similar to those in Fig. 3-1 should almost match, but if the balance is lost, the opposite side of the heel contact time observed during one walking cycle The degree of deviation does not match. The left / right balance can also be evaluated using the degree of this difference as an evaluation index.

The invention's effect

According to the present invention, a highly accurate CV can be easily measured even by an elderly person using only a smartphone equipped with the application of the present invention, and a highly reliable evaluation index can be obtained by easily repeating the measurement.
Differences in the case of applying some load such as a double task can be evaluated with high reliability.
As a result, the standing position of one's walking ability can be understood in comparison with the existing research standards in 0003. In addition, by continuously measuring, it is possible to evaluate changes over time, that is, deterioration due to age or disease, and improvement effects due to some kind of intervention. There are various interventions such as dual tasks, exercise, supplement intake, dietary improvement, and social activities.
If the CV suddenly deteriorates without knowing the reason, there is a possibility of some poor physical condition, which can be used as a countermeasure. Even though it is often overlooked in “somehow unwell”, if it can be confirmed by numerical values, it will lead to accurate judgment.
It is also significant as a means of measuring the degree of wandering when drunk.
Since it is possible to collect data for a large number of people at low cost under the same standard, research on the relationship between various diseases such as geriatric syndrome and CV can be promoted.

  Device configuration diagram   Measurement flow   Changes in acceleration in the gravitational direction due to differences in the holding position of measuring instruments   Examples of measurement results

A person to be measured (hereinafter simply referred to as a person to be measured) holds the 10 measurement apparatus in his / her hand and starts the application of the present invention or presses the measurement operation button of the application, and places the 10 measurement apparatus on the thigh. Store without fixing in the pocket of clothing that touches the part.
The 14 calculating unit automatically determines the start of walking of the person to be measured from the value of the 11 acceleration sensor, and the acceleration sensor for a certain time after the acceleration period of several seconds after the start of walking, or the values of the acceleration sensor and the magnetic sensor (hereinafter simply referred to as “acceleration sensor”) Get measurement data). Next, CV is calculated from the acceleration in the gravitational direction at each time based on the measurement data, and the CV data and the measurement time data are combined and transmitted from the 13 communication device to the 23 communication device. In the 20 analyzer, the received data is stored in the 21 storage unit, a reproducible value (hereinafter referred to as an evaluation index) is extracted from a plurality of CV data at the same time, and the evaluation index is stored in the 21 storage unit. 22 The evaluation unit evaluates the evaluation index and the transition.
The calculated evaluation index is transmitted to 13 communication devices and displayed on the screen of 10 measuring devices.

DESCRIPTION OF SYMBOLS 10 Measurement apparatus 11 Acceleration sensor 12 Magnetic sensor 13 Communication apparatus 20 Analysis apparatus 21 Storage part 22 Analysis part 23 Communication apparatus

Claims (1)

  By placing a measuring device equipped with an acceleration sensor in a pocket of clothing in contact with the leg, automatically detecting the start time of walking, and measuring a certain even number of steps after a certain period of time has elapsed from the start time of walking detection A method of aligning measurement conditions, automatically specifying each walking cycle from the value of the acceleration sensor in the measurement target, and calculating a walking cycle variation coefficient from each walking cycle. The certain even number of steps may be read as “maximum even number of steps within a certain time”.

Images