Posture control evaluation method and posture control evaluation device

Abstract

To evaluate how an evaluation target, such as a person, controls posture on the basis of acceleration measured in a trunk of the evaluation target.SOLUTION: Acceleration information is acquired for each unit time from a plurality of acceleration sensors mounted on a trunk of an evaluation target, evaluation acceleration information and evaluation instantaneous phase information are generated by removing a high frequency noise from the acceleration information collected along time series, at least one of a similarity of acceleration information between acceleration sensors and a degree of synchronous of instantaneous phases of acceleration sensors is calculated on the basis of the evaluation acceleration information and the evaluation instantaneous phase information, and evaluation information indicating how the evaluation target controls posture is generated and output.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] This invention relates to a posture control evaluation method and a posture control evaluation apparatus. [Background technology]

[0002] The posture of vertebrates such as humans (hereinafter referred to as "humans, etc.") is based on a complex regulatory function involving muscles, bones, nerves, and the brain, and it is believed that the brain functions highly to maintain balance.

[0003] In other words, a person's upright posture is maintained by controlling the body's center of mass within the base of support formed by the feet. In particular, because humans evolved from quadrupedal to bipedal locomotion, the position of the body's center of mass is higher relative to the narrow base of support. The trunk, consisting of the head, neck, chest, abdomen, pelvis, and tail, accounts for 50% to 60% of the body's mass. Therefore, the trunk, which has a high proportion of body mass, is positioned at the top, and the center of gravity is relatively high, which is a factor that destabilizes posture. Humans control this physical instability in their upright posture with extreme stability and precision.

[0004] On the other hand, the human body is a mechanically redundant structure because it is composed of a great many tissues, such as the skeleton, muscles, and ligaments. In particular, the trunk, from an anatomical perspective, has far more joints than the limbs, consisting of approximately 33 vertebrae and 24 ribs, and is continuously supported by soft tissues with various mechanical coefficients, such as ligaments, muscles, and skin. The spine also has a localized segmental structure and is classified into the cervical, thoracic, lumbar, sacral, and coccygeal vertebrae. Of these, the thoracic vertebrae form the rib cage with the ribs and sternum and are a highly rigid part of the skeletal structure. Structurally, the lumbar vertebrae are located between the highly rigid rib cage and the pelvis, and are thought to have skeletal structural redundancy in postural control and movement.

[0005] Furthermore, postural control strategies differ depending on sensory conditions. Regarding postural control strategies in a static standing position, it is known that the control strategies for head position and center of pressure are altered in postural control tasks with different sensory input conditions, such as eye-open or eye-closed.

[0006] To maintain an unstable standing posture, humans dynamically and highly control the trunk, which has a redundant mechanical structure. Therefore, investigating the role that the trunk, with its redundant structure, plays in controlling human standing posture is one of the most important factors.

[0007] In the field of biomechanics, the link-segment model is known as one of the commonly used methods for analyzing human standing posture.

[0008] The link-segment model is often used to divide the body into several link and segment parts for kinematic analysis of body segmentation. In most cases, the link-segment model only considers the joints of the limbs as links. However, the trunk anatomically has multiple joint structures. On the other hand, the range of motion at each joint in the trunk is smaller compared to the limbs. Therefore, it is difficult to clearly separate the joints of the trunk into link and segment parts in the link-segment model. Some previous studies have reported that a segment model with three or more segments can more appropriately represent trunk movement as an analysis method for trunk movement. However, there is much debate regarding the explanation of the link-segment model for the trunk, as different models are used depending on the research field.

[0009] Known methods for evaluating a person's resting standing posture include the COP recording method, which records the person's center of pressure (COP), and the cephalogram method, which records head sway.

[0010] In the COP recording method, for example, the subject is made to stand on a force plate and the sway of the center of load applied to the force plate is measured (see, for example, Patent Document 1). Patent Document 1 describes a system that performs training to maintain and improve standing posture balance function and comprehensively evaluates standing posture balance based on COP sway and trunk sway.

[0011] The standing posture balance evaluation training system described in Patent Document 1 comprises a balance board having a footboard on which a person can stand with both feet, which tilts in a predetermined direction due to the movement of the person's center of pressure on the footboard and detects the tilt, a head sway detector that detects the movement of the head of the person standing on the balance board, and a processing device that receives the detection results from the balance board and the head sway detector and evaluates the person's standing posture balance based on the tilt change of the balance board and the movement of the person's head.

[0012] In the cephalogram method, for example, a subject wears a cap with a marker indicating the top of the head, and the subject's head is photographed from above with a camera, and the movement of the marker is tracked to measure head sway (see, for example, Patent Document 2). Patent Document 2 describes a standing posture evaluation device that can ensure the accuracy of standing posture evaluation even if there is a displacement (tilt) in the position where the head marker is attached.

[0013] The standing posture evaluation device described in Patent Document 2 uses a 3D overhead camera to capture images of a head marker attached to the head of a subject in a standing position, and detects the position of the head's center of gravity projected onto the floor. The tilt of the head marker is detected by an acceleration sensor or the like, and the head's center of gravity is corrected. A body pressure sensor detects the position of the body's center of gravity projected onto the floor of the subject in a standing position. The control device evaluates the standing posture balance of the subject using the corrected head's center of gravity and the detected body's center of gravity.

[0014] Furthermore, Patent Document 3 describes a walking posture meter that can present to the user in a more easily understandable way the temporal changes in the quality of walking posture during continuous walking in daily life.

[0015] The walking posture meter described in Patent Document 3 includes an acceleration sensor attached to the midline of the waist of the person being measured, and an evaluation unit that repeatedly obtains an evaluation value that quantitatively represents the walking posture of the person being measured based on the output of the acceleration sensor for each predetermined unit period within a predetermined continuous walking period of 10 minutes or less, and a display processing unit that arranges the repeatedly obtained evaluation values in time series and displays them on a display screen.

Prior Art Documents

Patent Documents

[0016]

Patent Document 1

Patent Document 2

Patent Document 3

Non-Patent Documents

[0017]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0018] Although a person's standing posture appears to be stationary, it is known that each part of the body is constantly vibrating within a certain range. In the standing posture, the vibrations of each part of the body propagate between structures while increasing and decreasing. The vibration characteristics of the structures are less segmented in parts with similar vibration characteristics. On the other hand, the vibration characteristics of the structures are more segmented in parts with different vibration characteristics.

[0019] Non-patent document 1 shows that in the analysis of standing posture, where segmentation can be examined by evaluating the similarity of vibration characteristics, an acceleration sensor with high temporal resolution and sensitivity is useful for verifying subtle changes in vibration characteristics in posture control.

[0020] In vibration characteristic analysis using acceleration sensors, the magnitude and direction of instantaneous acceleration are often analyzed. In addition, analysis using phase information obtained from time-series changes in acceleration makes it possible to detect differences in vibration characteristics even when there is no difference in the magnitude of the vibration. Therefore, by analyzing the synchronization of the phase information of vibration characteristics, it is possible to detect similar motions over time.

[0021] This invention has been made in consideration of these circumstances, and its purpose is to provide a posture control evaluation method and posture control evaluation device that can evaluate how an object being evaluated controls its posture based on acceleration measured at the trunk of the object being evaluated, such as a person. [Means for solving the problem]

[0022] One aspect of the present invention is a posture control evaluation method for evaluating how a subject to be evaluated, such as a person, controls its posture, comprising: an acceleration information collection step of acquiring acceleration information from a plurality of acceleration sensors attached to the trunk of the subject to be evaluated at unit time intervals; a data processing step of removing high-frequency noise from the acceleration information collected in a time series in the acceleration information collection step and generating evaluation acceleration information, which is evaluation acceleration information from which high-frequency noise has been removed, and evaluation instantaneous phase information, which is the instantaneous phase information of the evaluation acceleration information; a data analysis step of calculating at least one of the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phase between acceleration sensors based on the evaluation acceleration information and evaluation instantaneous phase information generated in the data processing step, and generating evaluation information on how the subject to be evaluated controls its posture based on at least one of the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phase between acceleration sensors; and an evaluation information output step of outputting the evaluation information generated in the data analysis step.

[0023] One aspect of the present invention is that, in the acceleration information acquisition step of the above posture control evaluation method, acceleration information when the subject to be evaluated has its eyes open and acceleration information when the subject to be evaluated has its eyes closed are each collected in chronological order.

[0024] One aspect of the present invention is that, in the acceleration information acquisition step of the posture control evaluation method described above, a multi-axis acceleration sensor is used as the plurality of acceleration sensors to collect acceleration information in at least the left-right direction and acceleration information in the front-back direction, which are orthogonal to the vertical direction along the spine of the object to be evaluated in a standing posture, in a time series.

[0025] One aspect of the present invention is that, in the acceleration information acquisition step of the posture control evaluation method described above, a 3-axis acceleration sensor is used as the plurality of acceleration sensors to collect vertical acceleration information along the spine of the object to be evaluated in a standing posture, as well as acceleration information in the left-right direction and the front-back direction perpendicular to the vertical direction, each in a time series.

[0026] One aspect of the present invention is that, in the acceleration information acquisition step of the attitude control evaluation method described above, a 6-axis acceleration sensor is used as the plurality of acceleration sensors to acquire acceleration information in the vertical, left-right, and front-back directions, as well as acceleration information around the three axes, in a time-series manner.

[0027] In one aspect of the present invention, in the acceleration information acquisition step of the posture control evaluation method described above, the plurality of acceleration sensors are attached to the trunk of the body to be evaluated at equal intervals along the spine on the back side.

[0028] In one aspect of the present invention, in the acceleration information acquisition step of the posture control evaluation method described above, the plurality of acceleration sensors are attached to the back side along the spine at equal intervals from the head to a predetermined part of the trunk of the body being evaluated.

[0029] In one aspect of the present invention, in the acceleration information acquisition step of the above posture control evaluation method, the plurality of acceleration sensors are distributed and attached to a plurality of predetermined parts of the torso of the body to be evaluated.

[0030] In one aspect of the present invention, in the acceleration information acquisition step of the posture control evaluation method described above, the plurality of acceleration sensors are distributed and attached to three predetermined parts of the torso to be evaluated: the head, the chest, and the pelvis.

[0031] In one aspect of the present invention, in the acceleration information acquisition step of the attitude control evaluation method described above, additional acceleration sensors are further attached between the predetermined parts as the plurality of acceleration sensors.

[0032] In one aspect of the present invention, in the acceleration information acquisition step of the posture control evaluation method described above, the plurality of acceleration sensors are attached to a portion of the spine on the posterior side of the trunk of the body to be evaluated.

[0033] In one aspect of the present invention, in the data analysis step of the above attitude control evaluation method, the correlation coefficient between acceleration sensors is calculated as the similarity of acceleration information using the evaluation acceleration information.

[0034] One aspect of the present invention is that in the evaluation information output step of the above attitude control evaluation method, the evaluation information is output as a heat map and / or bar graph.

[0035] One aspect of the present invention is that, in the evaluation information output step of the above-described attitude control evaluation method, the evaluation information is output as a moving image that shows changes over time.

[0036] In one aspect of the present invention, in the data analysis step of the above attitude control evaluation method, at least one of the frequency power, the phase difference between acceleration sensors, and the cross-correlation coefficient between acceleration sensors is obtained for the acceleration information of the plurality of acceleration sensors, and this is used as evaluation information to determine how the object to be evaluated controls its attitude.

[0037] One aspect of the present invention is a posture control evaluation method described above, wherein a plurality of evaluation targets are grouped together as an evaluation target group, acceleration information is acquired from a plurality of acceleration sensors for each evaluation target in the evaluation target group at unit time intervals, evaluation acceleration information and evaluation instantaneous phase information are generated for each evaluation target in the evaluation target group, and statistical processing is performed on the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phases between acceleration sensors calculated for each evaluation target in the evaluation target group, and evaluation information is generated based on the results of the statistical processing.

[0038] One aspect of the present invention is a posture control evaluation device for evaluating how a subject to be evaluated, such as a person, controls its posture, comprising: a plurality of acceleration sensors attached to the trunk of the subject to be evaluated; an acceleration information collection unit that acquires acceleration information from the plurality of acceleration sensors at unit time intervals; a data processing unit that removes high-frequency noise from the acceleration information collected in a time series by the acceleration information collection unit and generates evaluation acceleration information, which is evaluation acceleration information from which high-frequency noise has been removed, and evaluation instantaneous phase information, which is the instantaneous phase information of the evaluation acceleration information; a data analysis unit that calculates at least one of the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phase between acceleration sensors based on the evaluation acceleration information and evaluation instantaneous phase information generated by the data processing unit, and generates evaluation information on how the subject to be evaluated controls its posture based on at least one of the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phase between acceleration sensors; and an evaluation information output unit that outputs the evaluation information generated by the data analysis unit. [Effects of the Invention]

[0039] According to the present invention, it is possible to evaluate how the subject of evaluation controls its posture based on the acceleration measured at the trunk of the subject of evaluation, such as a person. [Brief explanation of the drawing]

[0040] [Figure 1] This is a step diagram showing an example of a posture control evaluation method according to one embodiment. [Figure 2] This is a block diagram showing an example configuration of a posture control evaluation device according to one embodiment. [Figure 3] This figure schematically shows an example of a method for mounting multiple acceleration sensors according to one embodiment. [Figure 4] This figure schematically shows the relationship between the standing posture of the object being evaluated and the detection axis of the multi-axis acceleration sensor according to one embodiment. [Figure 5] This flowchart shows an example of the procedure for posture control evaluation processing according to one embodiment. [Figure 6] This flowchart shows an example of the acceleration data processing procedure according to one embodiment. [Figure 7] This flowchart shows an example of the procedure for acceleration data analysis processing and acceleration data analysis result output processing according to one embodiment. [Figure 8] This figure shows an example of the configuration of evaluation information for standing posture according to one embodiment. [Figure 9] This figure shows an example of the configuration of evaluation information for standing posture according to one embodiment. [Figure 10] This figure shows an example of the configuration of evaluation information for standing posture according to one embodiment. [Figure 11] This figure shows an example of the configuration of evaluation information for standing posture according to one embodiment. [Figure 12] This figure shows an example of the configuration of evaluation information for standing posture according to one embodiment. [Figure 13] This figure shows an example of the configuration of evaluation information for standing posture according to one embodiment. [Modes for carrying out the invention]

[0041] Embodiments of the present invention will be described below with reference to the drawings. It should be noted that the embodiments described below are not intended to unduly limit the scope of the present invention as described in the claims, and not all of the configurations described in these embodiments are necessarily essential as solutions to the present invention.

[0042] Figure 1 is a process diagram showing an example of the procedure for the posture control evaluation method according to this embodiment. Figure 2 is a block diagram showing an example of the configuration of the posture control evaluation device according to this embodiment. The posture control evaluation method according to this embodiment is a posture control evaluation method for evaluating how an evaluation target, such as a person, controls its posture, and has four steps, as illustrated in the process diagram of Figure 1: an acceleration information acquisition step ST1, a data processing step ST2, a data analysis step ST3, and an evaluation information output step ST4. Furthermore, the posture control evaluation method according to this embodiment is executed by the posture control evaluation device 100 illustrated in Figure 2.

[0043] In this embodiment, a person will be used as an example of the subject of evaluation.

[0044] Figure 3 is a schematic diagram showing an example of the arrangement of multiple acceleration sensors attached to the trunk of the body to be evaluated in the posture control evaluation method according to this embodiment. As illustrated in Figure 3, the multiple acceleration sensors are attached to the trunk 52 of the body to be evaluated, which consists of the head, neck, chest, abdomen, pelvis, and tail. In the example in Figure 3, 22 acceleration sensors S1, S2, ..., S22 are attached to the back side of the spine at equal intervals from the head 51 of the body to a predetermined part of the trunk 52 of the body to be evaluated 50.

[0045] Furthermore, "equal spacing" is not limited to strictly constant intervals; it may also mean substantially equal spacing. For example, the distances between all acceleration sensors may vary in length within an acceptable range relative to a constant length.

[0046] In Figure 1, in the acceleration information acquisition process ST1, acceleration information is acquired at unit time intervals from a sensor group 10 consisting of multiple acceleration sensors S1, S2, ..., S22 attached to the torso 52 of the body to be evaluated 50.

[0047] Next, in the data processing step ST2, high-frequency noise is removed from the acceleration information collected in chronological order in the acceleration information acquisition step ST1, and evaluation acceleration information (evaluation acceleration information) from which high-frequency noise has been removed, and instantaneous phase information (evaluation instantaneous phase information) of the evaluation acceleration information are generated.

[0048] Next, in the data analysis process ST3, the similarity of the acceleration information and the degree of synchronization of the instantaneous phase are determined from the evaluation acceleration information and evaluation instantaneous phase information generated in the data processing process ST2, and evaluation information (standing posture evaluation information) is generated to show how the evaluation target 50 controls its standing posture.

[0049] Next, in the evaluation information output process ST4, the evaluation information for standing posture generated in the data analysis process ST3 is output.

[0050] In Figure 2, the attitude control evaluation device 100 comprises a sensor group 10 consisting of multiple acceleration sensors S1, S2, ..., S22, an A / D conversion unit 20, a data acquisition unit 30 (data logger), and a calculation processing unit 40 (PC: personal computer).

[0051] Multiple acceleration sensors S1, S2, ..., S22 of the sensor group 10 are attached to the torso 52 of the evaluation target 50, as illustrated in Figure 3. The A / D conversion unit 20 generates acceleration information per unit time by digitizing the analog signals with signal levels corresponding to the acceleration output from the acceleration sensors S1, S2, ..., S22 of the sensor group 10. The data acquisition unit 30 collects the acceleration information generated per unit time by the A / D conversion unit 20 in a time series. The arithmetic processing unit 40 takes in the acceleration information per unit time collected in a time series by the data acquisition unit 30 and generates evaluation information (standing posture evaluation information) on how the evaluation target 50 controls its standing posture based on the acceleration information per unit time collected in a time series. The arithmetic processing unit 40 outputs the generated standing posture evaluation information.

[0052] In the attitude control evaluation device 100, in the acceleration information acquisition step ST1, the A / D conversion unit 20 digitizes analog signals with signal levels corresponding to the acceleration output from multiple acceleration sensors S1, S2, ..., S22 of the sensor group 10, thereby generating acceleration information for each unit of time. The data acquisition unit 30 collects the acceleration information generated for each unit of time in a time series, and the time series of acceleration information for each unit of time is input into the calculation processing unit 40. Next, in the data processing step ST2, the attitude control evaluation device 100 uses the calculation processing unit 40 to remove high-frequency noise from the time series of acceleration information for each unit of time, and generates evaluation acceleration information (evaluation acceleration information) from which high-frequency noise has been removed, and instantaneous phase information (evaluation instantaneous phase information) of the evaluation acceleration information (evaluation instantaneous phase information). Next, in the data analysis step ST3, the posture control evaluation device 100 uses the arithmetic processing unit 40 to determine the similarity of acceleration information and the degree of synchronization of instantaneous phase from the evaluation acceleration information and evaluation instantaneous phase information, and generates evaluation information (standing posture evaluation information) indicating how the evaluation target 50 controls its standing posture. Next, in the evaluation information output step ST4, the posture control evaluation device 100 uses the arithmetic processing unit 40 to output the standing posture evaluation information.

[0053] In the attitude control evaluation device 100, the A / D conversion unit 20 and the data acquisition unit 30 correspond to the acceleration information acquisition unit, and the calculation processing unit 40 corresponds to the data processing unit, the data analysis unit, and the evaluation information output unit.

[0054] Furthermore, each function of the arithmetic processing unit 40 of the attitude control evaluation device 100 may be realized by the arithmetic processing unit 40 being equipped with computer hardware such as a CPU (Central Processing Unit) and memory, and the CPU executing a computer program stored in memory. The arithmetic processing unit 40 may be configured using a general-purpose computer device, or it may be configured as a dedicated hardware device. For example, the arithmetic processing unit 40 may be configured using a server computer connected to a communication network such as the Internet. Also, each function of the arithmetic processing unit 40 may be realized by cloud computing. For example, the arithmetic processing unit 40 may be configured using a mobile terminal device such as a smartphone or a tablet computer (tablet PC). Also, the arithmetic processing unit 40 may be realized by a single computer, or the functions of the arithmetic processing unit 40 may be distributed and realized by multiple computers. Also, the arithmetic processing unit 40 may be configured to open a website using, for example, a WWW system.

[0055] In principle, the sensor group 10 in the posture control evaluation device 100 can evaluate how the subject 50 controls its standing posture based on the acceleration measured at the trunk 52 of the subject 50, by providing multiple single-axis acceleration sensors. On the other hand, the sensor group 10 in the posture control evaluation device 100 may also include multi-axis acceleration sensors S1, S2, ..., S22 (multi-axis acceleration sensors S1, S2, ..., S22) that measure acceleration information in at least the left-right direction and acceleration information in the front-back direction, respectively, which are perpendicular to the vertical direction along the spine.

[0056] Figure 4 is a schematic diagram showing the relationship between the standing posture of the evaluation subject 50, which is equipped with multiple multi-axis acceleration sensors S1, S2, ..., S22, and the detection axes of the multi-axis acceleration sensors S1, S2, ..., S22.

[0057] As shown in Figure 4, the spine and trunk 52 of a person in an upright posture are aligned along the Z-axis in the XYZ space, where the direction of gravity G (vertical direction) is the Z-axis, but they are gently curved in an S-shape in the Y-axis direction, which is the anterior-posterior direction of the trunk 52. For this reason, the posture control evaluation device 100 may be equipped with a sensor group 10 consisting of three-axis acceleration sensors S1, S2, ..., S22, which detect the X, Y, and Z axes, and use the acceleration information in the Z-axis direction of the acceleration sensors in the XYZ space to correct the orientation of the acceleration sensors and evaluate how the object to be evaluated 50 controls its upright posture. This makes it possible to evaluate the decrease in flexibility of the trunk 52 that occurs with aging.

[0058] In this embodiment, as a specific example, the acceleration sensors S1, S2, ..., S22 of the sensor group 10 are small 3-axis acceleration sensor modules MMA7361LC (manufactured by Strawberry Linux Co., Ltd.). The specifications of the acceleration sensors are as follows. Measurement range: ±58.8 m / sec 2 (±6G), Sensitivity: 206mV / G, Frequency Response: DC to 1500Hz, Noise: 350μG / √Hz (0.1Hz to 1000Hz), Module Size: 10mm (H) x 10mm (W) x 3.56mm (H), Weight: 2g.

[0059] The measurement sites are defined as 22 equally spaced points from the occipital protuberance to the sacrum (sacral vertebrae), and 22 acceleration sensors S1, S2, ..., S22 are attached at 22 equally spaced points on the skin on the back side from the head to the sacrum along the spine.

[0060] Furthermore, an A / D conversion board (NI USB-6225, manufactured by National Instruments) is used as the A / D conversion unit 20. Analog signals output from acceleration sensors S1, S2, ..., S22 are digitized at a sampling frequency of 1000 Hz. The data is then acquired by the arithmetic processing unit 40 (PC) via a data acquisition unit 30 using LabVIEW 2012 (manufactured by National Instruments). Finally, the data obtained from acceleration sensors S1, S2, ..., S22 is extracted as acceleration information for the X, Y, and Z axes.

[0061] Figure 5 is a flowchart showing the procedure for the attitude control evaluation process performed by the attitude control evaluation device 100 according to this embodiment.

[0062] First, in step SP1 (acceleration data acquisition process), the attitude control evaluation device 100 digitizes the analog signals output from acceleration sensors S1, S2, ..., S22 using the A / D conversion unit 20 at a sampling frequency of 1000 Hz, and acquires the acceleration data in chronological order into the arithmetic processing unit 40 via the data acquisition unit 30. Step SP1 (acceleration data acquisition process) corresponds to the acceleration information acquisition process ST1 in the attitude control evaluation method according to this embodiment.

[0063] Furthermore, when performing attitude control evaluation on multiple evaluation targets 50 in the attitude control evaluation device 100, in step SP1, acceleration data acquisition processing is sequentially executed for each of the multiple evaluation targets 50, so that the acceleration data from each of the multiple evaluation targets 50 is acquired by the calculation processing unit 40 in chronological order. Here, we will proceed with the explanation for one evaluation target 50.

[0064] Next, in step SP2 (acceleration data processing), the arithmetic processing unit 40 removes high-frequency noise from the acceleration data acquired in chronological order by the acceleration data acquisition process in step SP1 using a low-pass filter with a cutoff frequency of 20 Hz, thereby generating evaluation acceleration information (evaluation acceleration information) from which high-frequency noise has been removed. The arithmetic processing unit 40 also generates instantaneous phase information (evaluation instantaneous phase information) for the evaluation acceleration information.

[0065] Figure 6 is a flowchart of the acceleration data processing according to this embodiment. In Figure 6(A), the arithmetic processing unit 40 applies a low-pass filter (SP21) with a cutoff frequency of 20 Hz to the acceleration data acquired in time series by the acceleration data acquisition process in step SP1 to remove high-frequency noise and generate evaluation acceleration information from which high-frequency noise has been removed. Also, in Figure 6(B), the arithmetic processing unit 40 generates evaluation instantaneous phase information from the evaluation acceleration information from which high-frequency noise has been removed by the low-pass filter (SP21) by Hilbert transform processing (SP22). In this evaluation instantaneous phase information generation process, a Hilbert transform is applied to the time series data of the trunk 52 of the standing posture evaluation target 50 in the left-right (X-axis) direction and front-back (Y-axis) direction of the evaluation acceleration information to generate the instantaneous phase in the left-right (X-axis) direction (ML) at sampling time t. ML φ n (t) and instantaneous phase in the forward / backward (Y-axis) direction (AP) AP φ n (t) is calculated. Step SP2 (acceleration data processing) corresponds to the data processing step ST2 in the attitude control evaluation method according to this embodiment.

[0066] Next, in step SP3 (acceleration data analysis processing), the arithmetic processing unit 40 uses the evaluation acceleration information and evaluation instantaneous phase information generated by the acceleration data processing in step SP2 to perform acceleration similarity (AC) detection processing and instantaneous phase synchronization degree (PLV) calculation processing to generate evaluation information on how the evaluation target 50 controls its standing posture.

[0067] In step SP3, in the acceleration similarity (AC) detection process, the arithmetic processing unit 40 uses the acceleration information for evaluation to calculate the correlation coefficients between the acceleration sensors S1, S2, ···, S22 for the accelerations in the left-right (X-axis) direction and the front-back (Y-axis) direction of the trunk 52 of the standing posture evaluation target 50 as the similarity (AC) of the acceleration data between the acceleration sensors S1, S2, ···, S22.

[0068] Also, in the calculation process of the synchronization degree (PLV) of the instantaneous phase of acceleration, the synchronization degree (PLV: Phase Locking Value) of the instantaneous phase, which is used as an index for examining between measurement points in the analysis of oscillatory signals such as electroencephalograms, is adopted, and the synchronization degree (PLV) of the phases between all the acceleration sensors S1, S2, ···, S22 is obtained.

[0069] The calculation process of the synchronization degree (PLV) of the instantaneous phase of acceleration will be specifically described. The instantaneous phase ML φ n (t) at the sampling time t, which is the evaluation instantaneous phase information calculated in step SP2, and AP φ n (t) are used to represent the difference between the instantaneous phases j φ n (t) and j φ m (t) of the sensors n and m in the j direction at the sampling time t on the complex plane, and finally the average over all the sampling time numbers T is defined as shown in the following equation (1). j v[[ID=​​​​​​​​​​​​​​​​​​​nm We define it as shown in equation (3) below.

[0072]

number

[0073]

number

[0074] The synchronization index in equation (3) above. j V nm A value close to 1 indicates strong synchronization in phase, a value close to -1 indicates strong synchronization in opposite phase, and a value close to 0 indicates weak synchronization in either phase or opposite phase. The arithmetic processing unit 40 uses the synchronization index shown in equation (3) above as the instantaneous phase synchronization degree (PLV). j V nm Calculate.

[0075] In step SP3, for the acceleration of the trunk 52 of the standing posture evaluation target 50 in the left-right (X-axis) direction and the front-back (Y-axis) direction, a correlation coefficient is calculated as the similarity (AC: Acceleration Correlation) of the evaluation acceleration information between acceleration sensors S1, S2, ..., S22, and the degree of instantaneous phase synchronization (PLV: Phase Locking Value) is determined, and this is used as evaluation information (standing posture evaluation information) of how the evaluation target 50 controls its standing posture. Step SP3 (acceleration data analysis processing) corresponds to the data analysis process ST3 in the posture control evaluation method according to this embodiment.

[0076] Next, in step SP4 (acceleration data analysis result output processing), the arithmetic processing unit 40 generates heatmap display data and bar graph display data using the standing posture evaluation information generated in step SP3. This heatmap display data and bar graph display data are for displaying the correlation coefficients and instantaneous phase synchronization degrees between acceleration sensors S1, S2, ..., S22 included in the standing posture evaluation information as a heatmap and bar graph. Step SP4 (acceleration data analysis result output processing) corresponds to the evaluation information output process ST4 in the posture control evaluation method according to this embodiment.

[0077] The attitude control evaluation device 100 may perform attitude control evaluation on multiple evaluation targets 50. First, in step SP1, the attitude control evaluation device 100 sequentially executes acceleration data acquisition processing for each of the multiple evaluation targets 50, thereby acquiring each acceleration data from the multiple evaluation targets 50 in chronological order into the arithmetic processing unit 40. Next, in step SP2, the arithmetic processing unit 40 uses the acceleration data collected from the multiple evaluation targets 50 to generate evaluation acceleration information and evaluation instantaneous phase information for each evaluation target 50 through acceleration data processing. Next, in step SP3, the arithmetic processing unit 40 uses the evaluation acceleration information and evaluation instantaneous phase information for each evaluation target 50 to calculate a correlation coefficient as the similarity (AC) of the evaluation acceleration information between acceleration sensors S1, S2, ..., S22, and to determine the degree of synchronization (PLV) of the instantaneous phase. Furthermore, the arithmetic processing unit 40 performs statistical processing on the correlation coefficient and the degree of synchronization to obtain the mean (AAC: Average Acceleration Correlation, APLV: Average Phase Locking Value) and the standard deviation (SDAC: Standard Deviation Acceleration Correlation, SDPLV: Standard Deviation Phase Locking Value). The arithmetic processing unit 40 outputs the mean (AAC, APLV) and the standard deviation (SDAC, SDPLV) as evaluation data for the group being evaluated. The evaluation data for the group being evaluated may also be heatmap display data or bar graph display data for displaying in a heatmap or bar graph.

[0078] Figure 7 is a flowchart of the acceleration data analysis process and acceleration data analysis result output process according to this embodiment. In Figure 7(A), the arithmetic processing unit 40 calculates a correlation coefficient as the similarity (AC) of the evaluation acceleration information between acceleration sensors S1, S2, ..., S22 for each evaluation target 50, from the evaluation acceleration information for each evaluation target 50 generated by the acceleration data processing in step SP2 (step SP31). Next, the arithmetic processing unit 40 creates display data for the correlation coefficient between acceleration sensors S1, S2, ..., S22 for each evaluation target 50 (step SP32). Next, the arithmetic processing unit 40 performs statistical processing on the correlation coefficients obtained for each evaluation target 50 to obtain the mean (AAC) and standard deviation (SDAC) for all evaluation targets 50 (step SP33). Next, the arithmetic processing unit 40 creates and outputs display data for displaying the mean (AAC) and standard deviation (SDAC) as evaluation data for the evaluation target group (step SP34).

[0079] In Figure 7(B), the arithmetic processing unit 40 calculates the degree of instantaneous phase synchronization (PLV) between acceleration sensors S1, S2, ..., S22 for each evaluation target 50 from the evaluation instantaneous phase information generated by the acceleration data processing in step SP2 (step SP35). Next, the arithmetic processing unit 40 creates display data for the degree of instantaneous phase synchronization (PLV) between acceleration sensors S1, S2, ..., S22 for each evaluation target 50 (step SP36). Next, the arithmetic processing unit 40 calculates the average value (APLV) and standard deviation (SDPLV) for all evaluation targets 50 by performing statistical processing on the degree of synchronization (PLV) obtained for each evaluation target 50 (step SP37). Next, the arithmetic processing unit 40 creates and outputs display data for displaying the average value (APLV) and standard deviation (SDPLV) as evaluation data for the evaluation target group (step SP38).

[0080] Next, we will explain the results of an experiment in evaluating posture control using the posture control evaluation device 100 described above. In this experiment, 10 healthy adults (10 males, age: 20.9 ± 0.7 years, height: 171.22 ± 4.6 cm, weight: 68.9 ± 6.3 kg) were the subjects 50 being evaluated. Through this experiment, we were able to examine the behavior of the trunk 52, a redundant structure under standing posture control conditions with different levels of stability, from multiple points and in a data-driven rather than model-based manner, and obtain the following evaluation results regarding how the subjects 50 control segmentation.

[0081] In this experiment, we evaluated how structural redundancy in the trunk 52 is controlled in a task of maintaining an upright posture under two conditions: with eyes open and with eyes closed.

[0082] All 50 participants evaluated (10 individuals) were confirmed to have no history of orthopedic disease. The experiment was conducted in accordance with the guidelines proposed in the Declaration of Helsinki, the research protocol was approved by the Ethics Committee of Teikyo University of Science (Approval No. 20A018), and all participants, including the 50 participants evaluated, gave their informed consent to participate in the experiment.

[0083] The 50 subjects were instructed to stand still barefoot on a flat surface with the insides of both feet touching the surface. The task consisted of two conditions: standing with eyes open and standing with eyes closed. Each condition was performed randomly. The thinking time for each task was 20 seconds, and two trials were measured. During measurement, subjects were instructed to fix their gaze on a point at eye level 2 meters in front of them. In the closed-eye condition, subjects were also instructed to fix their gaze on the target before closing their eyes. Measurement began only after confirming that the subjects' gaze and static standing posture were stable.

[0084] For each of the 50 subjects being evaluated, the correlation coefficient between each acceleration sensor S1, S2, ..., S22 was calculated as the acceleration similarity (AC) using evaluation acceleration information obtained from the acceleration data of the measured acceleration sensors S1, S2, ..., S22. In addition, the average value (AAC) for 10 people evaluating the 50 subjects was calculated for each of the acceleration sensors S1, S2, ..., S22, and the standard deviation of the similarity (AC) (SDAC) was calculated as the variability of the acceleration similarity (AC) among the 50 subjects being evaluated, thereby evaluating the variability among the 10 people evaluating the 50 subjects.

[0085] Furthermore, using the instantaneous phase information obtained for each of the 50 evaluation subjects, the degree of instantaneous phase synchronization (PLV) between each acceleration sensor S1, S2, ..., S22 was calculated. For each of the acceleration sensors S1, S2, ..., S22, the average APLV for 10 evaluation subjects was calculated, and the standard deviation of the degree of instantaneous phase synchronization (PLV) (SDPLV) was also calculated to evaluate the variability occurring among the 50 evaluation subjects for 10 people.

[0086] In this experiment, acceleration similarity (AC) was used as one indicator to represent the degree of segmentation. (1) A score of 0.6 or higher but less than 0.8 indicates "a region with strong similarity and reduced segmentation," (2) A score of 0.4 or higher but less than 0.6 indicates "a region with moderate similarity and a certain degree of reduced segmentation," (3) A value of less than 0.4 indicates "a region where the similarity is weak and the segmental relationship is clearly high." This was used as an indicator. Furthermore, in this experiment, the degree of instantaneous phase synchronization (PLV) was used as one indicator to represent the degree of segmentation. (1) A value of 0.6 or higher but less than 0.8 indicates "a region with a strong degree of synchronization and reduced segmentation," (2) A value of 0.4 or higher but less than 0.6 indicates "a region with moderate synchronization and a certain degree of reduced segmentation," (3) A value of less than 0.4 indicates "a region where synchronization is weak and segmentation is clearly high." This was used as an indicator. Furthermore, an index using both acceleration similarity (AC) and instantaneous phase synchronization (PLV) may be used to represent the degree of segmentation.

[0087] Furthermore, the mean and standard deviation of acceleration similarity (AC) and instantaneous phase synchronization (PLV) are shown in heatmaps.

[0088] MATLAB® was used to calculate the similarity of acceleration and the degree of synchronization of instantaneous phase, and SPSS (Statistics Desktop 22.0) was used for statistical processing. The significance level was set to 0.05 for all calculations.

[0089] The similarity (AC) of acceleration between each acceleration sensor S1, S2, ..., S22 in the 50 evaluation subjects from 10 individuals was evaluated as shown in Figure 8.

[0090] Figure 8 is a heatmap showing an example of evaluation results regarding acceleration similarity (AC). Figure 8 shows an Average Heatmap for 50 evaluation subjects from 10 individuals, where the average value of acceleration similarity (AC) (AAC(a,b,c,d)) between acceleration sensors S1, S2, ..., S22 is used as evaluation information, and the heatmap is displayed in grayscale for each condition (ML (Medial Lateral): left-right direction, AP (Anterior Posterior): front-back direction, EO: eyes open condition, EC: eyes closed condition). Figure 8 also shows an SD Heatmap for 50 evaluation subjects from 10 individuals, where the standard deviation of acceleration similarity (AC) (SDAC(e,f,g,h)) between acceleration sensors S1, S2, ..., S22 is used as evaluation information, and the heatmap is displayed in grayscale for each condition (ML: left-right direction, AP: front-back direction, EO: eyes open condition, EC: eyes closed condition).

[0091] In Figure 8, the heatmap of the average acceleration similarity (AC) (AAC(a,b,c,d)) shows that the left vertical axis and horizontal axis represent sensor numbers, and the intensity of each section within the heatmap indicates the similarity between the two acceleration sensors corresponding to each section. The relationship between the intensity of the sections and the similarity is shown in the bar graph displayed alongside the heatmap, with the vertical axis of the bar graph representing the similarity. Alternatively, the sections could be color-coded instead of using intensity.

[0092] Furthermore, in Figure 8, in the heatmap of standard deviation (SDAC(e,f,g,h)), the left vertical axis and horizontal axis represent sensor numbers, and the intensity of each section in the heatmap indicates the standard deviation value of the similarity between the two acceleration sensors corresponding to each section. The relationship between the intensity of the sections and the standard deviation value is shown in the bar graph displayed alongside the heatmap, and the vertical axis of the bar graph represents the standard deviation value.

[0093] Furthermore, in Figure 8, the sections showing the accelerometer sets that showed particularly high average values ​​(AAC) under the eye-open condition EO are circled in white, and the sections showing the accelerometer sets that showed low standard deviations (SDAC) under the eye-open condition EO are circled in black.

[0094] In both the left-right and front-back directions of eye opening and closing, sections showing a continuous AAC of 0.4 or higher were extracted for combinations of acceleration sensors S8-S15.

[0095] The acceleration similarity (AC) for the left-right ML under eye-open EO conditions was as follows:

[0096] The combinations of acceleration sensors with a similarity of 0.8 or higher were S9-S10, S10-S11, and S11-S12.

[0097] The combinations of acceleration sensors with a similarity of 0.6 or higher and less than 0.8 were S1-S2, S5-S6, S7-S8, S8-S9, S12-S13, S13-S14, S14-15, S15-S16, and S16-S17.

[0098] The combinations of acceleration sensors with a similarity of 0.4 or higher and less than 0.6 were S4-S5, S17-S18, S20-S21, and S21-S22.

[0099] The combinations of acceleration sensors with a similarity of less than 0.4 were S2-S3, S3-S4, S6-S7, S18-S19, and S19-20.

[0100] The acceleration similarity (AC) for the anterior-posterior direction AP under eye-opening condition EO was as follows:

[0101] The combinations of acceleration sensors with a similarity of 0.8 or higher were S9-S10, S10-S11, and S11-S12.

[0102] The combinations of acceleration sensors with a similarity of 0.6 or higher and less than 0.8 were S1-S2, S5-S6, S7-S8, S8-S9, S12-S13, S13-S14, S14-S15, S15-S16, S16-S17, and S17-S18.

[0103] The combinations of acceleration sensors with a similarity of 0.4 or higher and less than 0.6 were S2-S3 and S21-S22.

[0104] The combinations of accelerometers with a similarity of less than 0.4 were S3-S4, S4-S5, S6-S7, S18-S19, S19-S20, and S20-S21.

[0105] The acceleration similarity (AC) for the left-right ML under closed-eye EC conditions was as follows:

[0106] No combinations of accelerometers with a similarity score of 0.8 or higher were found.

[0107] The combinations of acceleration sensors with a similarity of 0.6 or higher and less than 0.8 were S8-S9, S9-S10, S10-S11, S11-S12, S13-S14, and S14-S15.

[0108] The combinations of acceleration sensors with a similarity of 0.4 or higher and less than 0.6 were S4-S5, S6-S7, S12-S13, S16-S17, S17-S18, S18-S19, S20-S21, and S21-S22.

[0109] The combinations of acceleration sensors with a similarity of less than 0.4 were S1-S2, S2-S3, S3-S4, S5-S6, S7-S8, S15-S16, and S19-S20.

[0110] The acceleration similarity (AC) for the anterior-posterior direction AP under closed-eye condition EC was as follows:

[0111] No combinations of accelerometers with a similarity score of 0.8 or higher were found.

[0112] The combinations of acceleration sensors with a similarity of 0.6 or higher and less than 0.8 were S9-S10, S10-S11, S11-S12, S13-S14, S14-S15, and S17-S18.

[0113] The combinations of acceleration sensors with a similarity of 0.4 or higher and less than 0.6 were S1-S2, S6-S7, S8-S9, S12-S13, S16-S17, and S18-S19.

[0114] The accelerometer sensor combinations with a similarity of less than 0.4 were S2-S3, S3-S4, S4-S5, S5, S7-S8, S15-S16, S19-S20, S20-S21, and S21-S22.

[0115] The acceleration similarity (AC) (AAC(a,b,c,d)) decreased in both the left-right ML direction and the anterior-posterior AP direction in the closed-eye condition EC compared to the open-eye condition EO.

[0116] The tendency for acceleration similarity (AC) (AAC(b,d)) to decrease under closed-eye EC conditions was more pronounced in the anterior-posterior AP direction compared to the lateral-lateral ML direction.

[0117] In the 50 evaluation subjects from 10 individuals, the standard deviation (SDAC(e,f,g,h)) indicating the variability of acceleration similarity (AC) between each acceleration sensor S1, S2, ... S22 was as follows:

[0118] The standard deviation (SDAC(e,f,g,h)) of the acceleration similarity (AC) between each acceleration sensor S1, S2, ..., S22 was lower when eyes were open than when eyes were closed.

[0119] Furthermore, under the eye-opening condition EO, the combination of acceleration sensors S8-S12 showed a particularly low standard deviation (SDAC(f,h)) of the similarity of acceleration (AC) among the 50 subjects being evaluated.

[0120] The tendency for the standard deviation (SDAC(f,h)) to be lower in the open-eye EO condition was more pronounced in the lateral ML direction compared to the anterior-posterior AP direction.

[0121] The combination of acceleration sensors S1-S2 showed a higher standard deviation (SDAC(f,h)) in the left-right direction (ML) and the front-back direction (AP) under the eye-open condition (EO) compared to the eye-closed condition (EC).

[0122] The combination of acceleration sensors S11-S12 showed a lower standard deviation (SDAC(f,h)) in the left-right direction (ML) and the front-back direction (AP) under the eye-open condition (EO) compared to the eye-closed condition (EC).

[0123] Furthermore, the average value (APLV) of the instantaneous phase synchronization (PLV) between each acceleration sensor S1, S2, ... S22 in the 50 evaluation subjects from 10 individuals was determined as shown in Figure 9.

[0124] Figure 9 is a heatmap showing an example of evaluation results regarding the degree of instantaneous phase synchronization (PLV). Figure 9 shows an Average Heatmap for 50 evaluation subjects from 10 individuals, showing the average value (APLV(a,b,c,d)) of the degree of instantaneous phase synchronization (PLV) of acceleration between acceleration sensors S1, S2, ... S22, as evaluation information, categorized by condition (ML: left-right direction, AP: front-back direction, EO: eyes open condition, EC: eyes closed condition). Figure 9 also shows an SD Heatmap for 50 evaluation subjects from 10 individuals, showing the standard deviation (SDPLV(e,f,g,h)) of the degree of instantaneous phase synchronization (PLV) of acceleration between acceleration sensors S1, S2, ... S22, as evaluation information, categorized by condition (ML: left-right direction, AP: front-back direction, EO: eyes open condition, EC: eyes closed condition).

[0125] In Figure 9, the heatmap of the average instantaneous phase sync rate (PLV) (APLV(a,b,c,d)) shows that the left vertical axis and horizontal axis represent sensor numbers, and the intensity of each section in the heatmap indicates the degree of sync between the two accelerometers corresponding to each section. The relationship between the intensity of the sections and the degree of sync is shown in the bar graph displayed alongside the heatmap, with the vertical axis of the bar graph representing the degree of sync. Alternatively, the sections may be color-coded instead of using intensity.

[0126] Furthermore, in Figure 9, the heatmap of standard deviation (SDPLV(e,f,g,h)) shows that the left vertical axis and horizontal axis represent sensor numbers, and the intensity of each section in the heatmap indicates the standard deviation value of the synchronization between the two acceleration sensors corresponding to each section. The relationship between the intensity of the sections and the standard deviation value is shown in the bar graph displayed alongside the heatmap, with the vertical axis of the bar graph representing the standard deviation value.

[0127] Furthermore, in Figure 9, the sections showing the sets of accelerometers that showed particularly high average values ​​(APLV) under the eye-open condition EO are circled in white, and the sections showing the sets of accelerometers that showed low standard deviations (SDPLV) under the eye-open condition EO are circled in black.

[0128] For both the longitudinal AP and lateral ML directions, the instantaneous phase synchronous degree (PLV) was such that sections showing a continuous APLV of 0.4 or higher were extracted for combinations of acceleration sensors S8-S15.

[0129] The degree of synchronization of the instantaneous phase of acceleration (PLV) in the left-right direction ML under eye-open EO conditions was as follows:

[0130] No combinations of accelerometers with a synchronization degree of 0.8 or higher were observed.

[0131] The combinations of acceleration sensors with a synchronization degree of 0.6 or higher and less than 0.8 were S9-S10, S10-S11, S11-S12, S13-S14, and S14-S15.

[0132] The combinations of acceleration sensors with a synchronization degree of 0.4 or higher and less than 0.6 were S1-S2, S8-S9, S12-S13, S16-S17, S17-S18, and S18-S19.

[0133] The combinations of accelerometers with a synchronization degree of less than 0.4 were S2-S3, S3-S4, S4-S5, S5-S6, S6-S7, S7-S8, S15-S16, S19-S20, S20-S21, and S21-S22.

[0134] The average instantaneous phase sync rate (PLV) (APLV) was lower in both the left-right ML and anterior-posterior AP directions under closed-eye EC compared to open-eye EO.

[0135] The tendency for the average value of instantaneous phase synchronization (PLV) (APLV) to decrease under closed-eye EC conditions was more pronounced in the anterior-posterior AP direction compared to the lateral ML direction.

[0136] In the 50 evaluation subjects from 10 individuals, the standard deviation (SDPLV(e,f,g,h)) indicating the variation in the degree of synchronization (PLV) of instantaneous phase of acceleration between each acceleration sensor S1, S2, ... S22 was as follows:

[0137] The standard deviation (SDPLV(e,f,g,h)) of the degree of synchronization (PLV) of instantaneous phase of acceleration between each acceleration sensor S1, S2, ... S22 was lower when eyes were open than when eyes were closed.

[0138] Furthermore, the combination of acceleration sensors S8-S12 showed a particularly low standard deviation of the degree of synchronization (PLV) (SDPLV(e,f,g,h)).

[0139] The tendency for the standard deviation of synchronization (PLV) (SDPLV(e,f,g,h)) to be lower under eye-open EO conditions was more pronounced in the lateral ML direction compared to the anterior-posterior AP direction.

[0140] The combination of acceleration sensors S1-S2 resulted in a higher standard deviation of synchronization degree (PLV) (SDPLV(e,f,g,h)) in the left-right ML direction and the front-back AP direction compared to the closed-eye condition EC.

[0141] The combination of acceleration sensors S11-S12 resulted in a lower standard deviation of synchronization degree (PLV) (SDPLV(e,f,g,h)) in the left-right direction (ML) and the front-back direction (AP) under the eye-open condition (EO) compared to the eye-closed condition (EC).

[0142] Figure 10 is a bar graph illustrating the redundancy of trunk posture control based on the average value (APLV) of the instantaneous phase synchronization (PLV) of acceleration. Figure 10 shows the APLV (a) for the left-right ML direction and the APLV (b) for the front-back AP direction of adjacent acceleration sensors.

[0143] In Figure 10, the horizontal axis of the graph shows the APLV values ​​between adjacent accelerometers, separately for eye-open condition EO and eye-closed condition EC. The APLV values ​​for 21 pairs of accelerometers, from accelerometer S1 to S22, are shown from top to bottom. Combinations of accelerometers showing high APLV values ​​are shown in dark gray, and combinations showing low APLV values ​​are shown in light gray. Alternatively, color coding could be used instead of varying shades.

[0144] Compared to EO under the open-eye condition, a decrease in APLV, mainly in the thoracic region, was observed under EC under the closed-eye condition.

[0145] From the results of the attitude control evaluation experiment conducted using the attitude control evaluation device 100 described above, multiple clusters of acceleration sensors were identified on the heat map with high average values ​​of acceleration similarity (AAC) and instantaneous phase synchronization (APLV) in both the left-right direction ML and the front-back direction AP, under both the open-eye condition EO and the closed-eye condition EC. Measurement sites with high APLV and AAC can be evaluated as being areas with low segmentation.

[0146] Furthermore, the acceleration sensor groups that showed particularly high values ​​of 0.6 or higher for the average acceleration similarity (AC) (AAC) and the average instantaneous phase symmetry (PLV) (APLV) were those attached to the head, thoracic cage, and lumbar spine. The spine, which is the bone that makes up the vertebral column, is classified into the cervical, thoracic, lumbar, sacral, and coccygeal vertebrae and has a localized segmental structure. The thoracic spine is a highly rigid part of the skeletal structure, forming the rib cage with the ribs and sternum. On the other hand, the lumbar spine, which is located between the highly rigid thoracic cage and the pelvis, showed high APLV and AAC values ​​despite being a highly flexible part of the skeletal structure. It is thought that the rigidity of the lumbar spine increased because the muscles of the lower back were activated to support the instability in the skeletal structure of the lumbar spine.

[0147] Furthermore, from the results of the posture control evaluation experiment conducted using the posture control evaluation device 100 described above, the redundancy of trunk posture control based on the average value of the instantaneous phase synchronization (PLV) of acceleration (APLV) is shown in Figure 10 as a horizontal bar graph. A comparison of APLV under open-eye condition EO and closed-eye condition EC revealed that the control strategy for structural redundancy of the trunk 52 in a standing posture can change depending on the conditions.

[0148] Furthermore, from the results of the posture control evaluation experiment conducted using the posture control evaluation device 100 described above, the group of acceleration sensors located from the thoracic cage to the lumbar spine (accelerometers S8-S15), which showed high values ​​for both the lateral ML and anterior-posterior AP in the open-eye condition EO, showed low values ​​in the more unstable closed-eye condition EC. In addition, from the results of the standard deviations (SDAC, SDPLV) of the acceleration similarity (AC) and the anterior-posterior AP, it became clear that the variability among the evaluation subjects 50 in the trunk 52 increased in the closed-eye condition EC compared to the open-eye condition EO, for both the lateral ML and anterior-posterior AP.

[0149] As described above, this embodiment provides the effect of being able to evaluate how the person being evaluated (50) controls its standing posture based on the acceleration measured at the trunk (52) of the person being evaluated (50).

[0150] In the embodiments described above, the standing posture evaluation information was displayed using a heatmap or bar graph, but the invention is not limited to this. For example, the standing posture evaluation information may be output as a video or other format that shows changes over time.

[0151] Figures 11-13 are screen capture data showing an example of a video of standing posture evaluation information. In Figures 11-13, the screen area 200 shows acceleration information obtained from each of the 22 acceleration sensors S1, S2, ..., S22 in chronological order for each acceleration sensor. Also in the screen area 200, the acceleration information of each acceleration sensor S1, S2, ..., S22 at time t (seconds) is indicated by circles. The similarity (AC) of acceleration between each acceleration sensor S1, S2, ..., S22 is indicated by the intensity of these circles. Acceleration sensors with similar intensity of circles have a high degree of acceleration similarity (AC). In addition, the screen area 210 shows the intensity of the circles for each acceleration sensor S1, S2, ..., S22 to make the intensity of the circles easier to recognize.

[0152] Figure 11 is a screen capture at time "0.001 seconds", Figure 12 is a screen capture at time "5.001 seconds", and Figure 13 is a screen capture at time "10.001 seconds". As shown in Figures 11-13, the arithmetic processing unit 40 generates video display data for displaying a video in which the acceleration information of each acceleration sensor S1, S2, ..., S22 and the acceleration similarity (AC) between each acceleration sensor S1, S2, ..., S22 at each time t are continuously displayed on the screen in a time series using circles and the intensity of the circles, and outputs this video display data as evaluation information for standing posture. When this video display data is played back and displayed on the screen, the temporal changes in acceleration and acceleration similarity (AC) of each part of the trunk 52 of the evaluation target 50 (the part where each acceleration sensor is attached) can be clearly presented. Note that instead of varying the intensity of the circles, the circles may be color-coded.

[0153] Furthermore, in the embodiment described above, as illustrated in Figure 3, 22 acceleration sensors S1, S2, ..., S22 were attached to the back of the subject under evaluation 50 at equal intervals along the spine from the head 51 to a predetermined part of the trunk 52 (for example, the sacrum in the above experiment), but the method of attaching the acceleration sensors can be changed as appropriate.

[0154] For example, the number of acceleration sensors is not limited to 22 as illustrated in Figure 3 above. By installing three or more acceleration sensors, it is possible to calculate the similarity of acceleration information (AC) and the degree of instantaneous phase synchronization (PLV) based on the acceleration information obtained from them, and generate evaluation information on how the object under evaluation 50 controls its posture.

[0155] Furthermore, for example, three acceleration sensors may be attached to the back of the head, chest, and pelvis of the 50 subjects being evaluated, for a total of nine sensors. In each of these areas (head, chest, and pelvis), the three acceleration sensors may or may not be spaced equally apart. In addition, an additional acceleration sensor may be attached to the back between the head and chest. Also, an additional acceleration sensor may be attached to the back between the chest and pelvis.

[0156] Furthermore, when using multiple acceleration sensors, it is preferable to space them equally to make it easier to grasp the characteristics of acceleration in each part of the torso 52.

[0157] Furthermore, the preferred location for attaching the accelerometer is on the spine on the back. This is because there are virtually no muscles on the spine on the back, which prevents the accelerometer from detecting acceleration information caused by muscle movement, thereby improving the accuracy of the analysis. However, if the evaluation is to be conducted in a way that allows the accelerometer to detect acceleration information caused by muscle movement, the location for attaching the accelerometer does not have to be on the spine on the back. For example, the accelerometer may be attached to a location shifted to the left or right of the spine on the back.

[0158] Furthermore, in the above-described embodiment, an example of the sensor group 10 being equipped with three-axis acceleration sensors S1, S2, ...S22 was explained. However, by equipping the sensor group 10 with six-axis acceleration sensors S1, S2, ...S22 that also detect acceleration (angular velocity or rotational acceleration) in the directions θx, θy, and θz around the X, Y, and Z axes, as shown in Figure 4, it is possible to evaluate torso twisting and other such movements.

[0159] Furthermore, by using a 9-axis accelerometer that also detects direction based on the Earth's magnetic field, it becomes possible to automate the calibration process for each accelerometer S1, S2, ... S22 attached to the torso 52 of the subject 50 under evaluation along the spine, relative to the origin orientation in three-dimensional space.

[0160] The arithmetic processing unit 40 may calculate at least one of the acceleration similarity (AC) and the degree of synchronization of the instantaneous phase of acceleration (PLV). For example, the arithmetic processing unit 40 may calculate only the acceleration similarity (AC), or only the degree of synchronization of the instantaneous phase of acceleration (PLV). For example, the arithmetic processing unit 40 may calculate both the acceleration similarity (AC) and the degree of synchronization of the instantaneous phase of acceleration (PLV).

[0161] Furthermore, the arithmetic processing unit 40 may further determine at least one of the following from the acceleration information of acceleration sensors S1, S2, ..., S22: frequency power, phase difference between acceleration sensors, and cross-correlation coefficient between acceleration sensors, and use this as evaluation information to determine how the object under evaluation 50 controls its posture.

[0162] Furthermore, the posture control evaluation device 100 may also be equipped with sensors for measuring the center of gravity of the object to be evaluated 50. The calculation processing unit 40 generates evaluation information indicating which part of the trunk 52 (which part of which acceleration sensor is attached) is involved in the center of gravity sway of the object to be evaluated 50, based on the time-series data of the center of gravity sway of the object to be evaluated 50 obtained from the measurement results of the center of gravity of the object to be evaluated 50 and the time-series data of acceleration information obtained from acceleration sensors S1, S2, ..., S22 attached at equal intervals to the trunk 52 of the object to be evaluated 50.

[0163] In the embodiments described above, humans were used as an example of the subject of evaluation, but the method is also applicable to vertebrates other than humans.

[0164] As described above, according to this embodiment, the posture control evaluation device 100 acquires acceleration information from a plurality of acceleration sensors attached to the trunk of the subject to be evaluated at unit time intervals, removes high-frequency noise from the acceleration information collected in a time series, generates evaluation acceleration information, which is the acceleration information for evaluation from which high-frequency noise has been removed, and evaluation instantaneous phase information, which is the instantaneous phase information of the evaluation acceleration information, calculates at least one of the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phase between acceleration sensors based on the generated evaluation acceleration information and evaluation instantaneous phase information, generates evaluation information on how the subject to be evaluated controls its posture based on at least one of the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phase between acceleration sensors, and outputs the generated evaluation information. This provides the effect of being able to evaluate how the subject to be evaluated controls its posture based on the acceleration measured on the trunk of the subject to be evaluated, such as a person.

[0165] Furthermore, by examining the behavior of the trunk, a redundant structure under different standing posture control conditions (eyes open, eyes closed), from multiple perspectives and in a data-driven rather than model-based manner, it is possible to evaluate how the subject controls segmentation. In this regard, as is clear from the experimental results described above, when visual input is lost and it becomes necessary to control an even more unstable standing posture, a response of increasing links in the rib cage is observed. This suggests that, in order to control the structural redundancy of the trunk in a human standing posture, it is important to make the rib cage a certain degree of "cohesion" and adjust its acceleration. According to the experiments described above, in a task of maintaining a standing posture under two conditions, eyes open and eyes closed, it is possible to evaluate how the subject controls structural redundancy in the trunk using multiple acceleration sensors.

[0166] Furthermore, in the experiment described above, 10 healthy adults were evaluated to assess how they control their posture, but the evaluation group of 50 is not limited to healthy adults. For example, multiple evaluation groups with similar postures could be used.

[0167] Furthermore, by accumulating posture evaluation information from the evaluation groups, it becomes possible to compare and evaluate posture evaluation information between evaluation groups, and to assess which evaluation group's posture evaluation information an individual's is closest to. For example, it is possible to evaluate how factors such as aging, growth, disease, martial arts, dance, and tea ceremony experience affect the posture of the evaluation group.

[0168] Furthermore, the posture control evaluation method and posture control evaluation device according to this embodiment can be applied to various fields. For example, the standing posture evaluation information obtained from the evaluation target 50 can be used to determine diseases, mind-body balance, etc. of the evaluation target 50. For example, evaluation target groups can be formed for each sport, and the standing posture evaluation information obtained from each evaluation target group for each sport can be used to analyze body control, etc., in each sport. For example, the standing posture evaluation information obtained from evaluation target groups can be used in the development of robots that work in a standing posture.

[0169] Alternatively, a computer program for realizing the functions of the attitude control evaluation device described above may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be loaded into a computer system and executed. The term "computer system" here may include hardware such as an operating system and peripheral devices. Furthermore, "computer system" shall also include the homepage provisioning environment (or display environment) if a WWW system is being used. Furthermore, "computer-readable recording media" refers to writable non-volatile memory such as flexible disks, magneto-optical disks, ROMs, and flash memory, portable media such as DVDs (Digital Versatile Discs), and storage devices such as hard disks built into computer systems.

[0170] Furthermore, "computer-readable recording media" also includes volatile memory (such as DRAM (Dynamic Random Access Memory)) within computer systems that act as servers or clients when programs are transmitted via networks such as the Internet or communication lines such as telephone lines, which retain programs for a certain period of time. Furthermore, the above program may be transmitted from a computer system that stores the program in a memory device or the like to another computer system via a transmission medium or by transmission waves within the transmission medium. Here, the "transmission medium" used to transmit the program refers to a medium that has the function of transmitting information, such as a network (communication network) like the Internet or a communication line (communication line) like a telephone line. Furthermore, the above program may be intended to implement some of the functions described above. It may also be a so-called differential file (differential program) that can implement the aforementioned functions in combination with programs already recorded in the computer system.

[0171] Although embodiments of the present invention have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments, and design modifications and the like are also included within the scope of the gist of the present invention. [Explanation of Symbols]

[0172] 10...Sensor group, 20...A / D conversion unit, 30...Data acquisition unit, 40...Calculation processing unit, 100...Attitude control evaluation device, S1, S2, ..., S22...Accelerometer

Claims

1. A posture control evaluation method for evaluating how a subject of evaluation, such as a person, controls its posture, An acceleration information acquisition step is performed to acquire acceleration information from multiple acceleration sensors attached to the torso of the body being evaluated at regular intervals. A data processing step which removes high-frequency noise from acceleration information collected in a time series in the acceleration information acquisition step and generates evaluation acceleration information, which is evaluation acceleration information from which high-frequency noise has been removed, and evaluation instantaneous phase information, which is instantaneous phase information of the evaluation acceleration information. A data analysis step which calculates at least one of the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phase between acceleration sensors based on the evaluation acceleration information and evaluation instantaneous phase information generated in the data processing step, and generates evaluation information on how the object to be evaluated controls its posture based on at least one of the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phase between acceleration sensors, An evaluation information output step that outputs the evaluation information generated in the data analysis step, A method for evaluating posture control, including the following:

2. In the acceleration information acquisition process, acceleration information is collected in chronological order when the subject to evaluation has its eyes open and acceleration information is collected when the subject to evaluation has its eyes closed. The posture control evaluation method according to claim 1.

3. In the acceleration information acquisition step, a multi-axis acceleration sensor is used as the plurality of acceleration sensors to collect acceleration information in at least the left-right direction and acceleration information in the front-back direction, which are perpendicular to the vertical direction along the spine of the subject being evaluated in a standing posture, in a time series. The posture control evaluation method according to claim 1.

4. In the acceleration information acquisition process, a three-axis acceleration sensor is used as the plurality of acceleration sensors to collect vertical acceleration information along the spine of the subject being evaluated in a standing posture, acceleration information in the left-right direction perpendicular to the vertical direction, and acceleration information in the front-back direction, each in a time series. The posture control evaluation method according to claim 1.

5. In the acceleration information acquisition process, using a six-axis acceleration sensor as the plurality of acceleration sensors, acceleration information in the vertical, left-right, and front-back directions, as well as acceleration information around the three axes, is collected in a time-series manner. The posture control evaluation method according to claim 4.

6. In the acceleration information acquisition process, the plurality of acceleration sensors are attached to the back of the torso of the body being evaluated at equal intervals along the spine. The posture control evaluation method according to claim 1.

7. In the acceleration information acquisition process, the plurality of acceleration sensors are attached to the back side along the spine at equal intervals from the head to a predetermined part of the trunk of the body being evaluated. The posture control evaluation method according to claim 1.

8. In the acceleration information acquisition process, the multiple acceleration sensors are distributed and attached to multiple predetermined parts of the torso of the body being evaluated. The posture control evaluation method according to claim 1.

9. In the acceleration information acquisition process, the multiple acceleration sensors are distributed and attached to three predetermined parts of the torso being evaluated: the head, chest, and pelvis. The posture control evaluation method according to claim 1.

10. In the acceleration information acquisition step, additional acceleration sensors are further mounted between the predetermined parts as the plurality of acceleration sensors. The posture control evaluation method according to any one of claims 8 or 9.

11. In the acceleration information acquisition process, the plurality of acceleration sensors are attached to a portion of the spine on the back side of the torso being evaluated. The posture control evaluation method according to claim 1.

12. In the data analysis step, the correlation coefficient between acceleration sensors is calculated as the similarity of acceleration information using the evaluation acceleration information. The posture control evaluation method according to claim 1.

13. In the evaluation information output step, the evaluation information is output as a heat map and / or bar graph. The posture control evaluation method according to claim 1.

14. In the evaluation information output step, the evaluation information is output as a moving image that shows changes over time. The posture control evaluation method according to claim 1.

15. In the data analysis step, further, at least one of the following is determined for the acceleration information of the multiple acceleration sensors: frequency power, phase difference between acceleration sensors, and cross-correlation coefficient between acceleration sensors, and this is used as evaluation information to determine how the object being evaluated controls its posture. The posture control evaluation method according to claim 1.

16. Multiple of the aforementioned evaluation targets are designated as an evaluation target group. In the acceleration information acquisition step, for each evaluation target group, acceleration information is acquired from the multiple acceleration sensors at unit time intervals. In the data processing step described above, for each evaluation target in the evaluation target group, the evaluation acceleration information and the evaluation instantaneous phase information are generated. In the data analysis step, statistical processing is performed on the similarity of acceleration information between acceleration sensors and the degree of instantaneous phase synchronization between acceleration sensors, which are calculated for each evaluation target group, and the evaluation information is generated based on the results of the statistical processing. The posture control evaluation method according to claim 1.

17. A posture control evaluation device for evaluating how a subject of evaluation, such as a person, controls its posture, Multiple acceleration sensors attached to the torso of the subject being evaluated, An acceleration information acquisition unit that acquires acceleration information from the aforementioned multiple acceleration sensors at unit time intervals, A data processing unit removes high-frequency noise from acceleration information collected in a time series by the acceleration information collection unit and generates evaluation acceleration information, which is evaluation acceleration information from which high-frequency noise has been removed, and evaluation instantaneous phase information, which is instantaneous phase information of the evaluation acceleration information. A data analysis unit calculates at least one of the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phase between acceleration sensors based on the evaluation acceleration information and evaluation instantaneous phase information generated by the data processing unit, and generates evaluation information on how the object to be evaluated controls its posture based on at least one of the similarity of acceleration information between acceleration sensors and the degree of synchronization of instantaneous phase between acceleration sensors. An evaluation information output unit that outputs evaluation information generated by the data analysis unit, A posture control evaluation device equipped with the following features.

Images