JP2003111646A - System for measuring and analyzing characteristic of body pressure distribution and system for predicting and evaluating change with time of body pressure distribution - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system for measuring and analyzing characteristics of body pressure distribution by which the measurement and analysis with time are performed and the results are classified. SOLUTION: In the system for measuring displaying and storing in a computer the effective body pressure distribution by a sensor a body pressure distribution measuring means by using the sensor which is arranged so as to cover the sitting face and the back face of a chair or a seat and can effectively detect sitting and back pressures of a human body. The sitting and back pressures are displayed on the computer. A body characteristic analyzing on a definite number of examees are performed and differences of sex and body shapes are measured. In a posture analyzing means, differences of various postures are obtained from the sitting pressure. In a means for analyzing shape and function of the chairs the differences in cushion raw materials and shapes and differences in functions of the chairs in backward inclined postures are measured. In a means for analyzing the characteristic the lapse of time, the lapse of time after seating is measured.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to measurement, analysis and evaluation of body pressure distribution showing correspondence between physical quantity and sensory quantity of sitting comfort of a chair or the like.

[0002]

2. Description of the Related Art Recently, by using an ultra-thin flexible pressure-sensitive sensor sheet having a thickness of 0.5 mm or less, it is possible to design a chair or an automobile seat with a feeling very close to actual use. The new body pressure distribution measuring device is used. FIG. 47 is a view showing the structure of such a sensor sheet, which is a packing sheet 1 such as a polyester film.
A row electrode 102 and a column electrode 104 are provided between 00 and 105, and these electrodes are coated with a pressure-sensitive special ink 103 and silver paste (silver color) of the electrodes and printed in a grid pattern. . The intersection 106 of the row and column electrodes acts as the sensing point.

When this sensor sheet is spread over a chair or a seat of an automobile and a person sits down, pressure is applied to the sensing point 106 shown in FIG. 47, and as shown in the graph of FIG. The resistance value changes, and this change in resistance value is transmitted to the detection circuit (not shown) as a change in the current in the row direction and the column direction, and the value is A / D converted into 8 bits to be corrected and converted into pressure. On the other hand, the output voltage is converted so that it has linearity, and the data is captured as body pressure distribution data in a personal computer or the like. Sensing points usually exist in the unit of several thousand points, and the taken body pressure distribution data is processed by image processing software and displayed as a body pressure distribution image on a personal computer, which is used for chair and seat development data. .

[0004]

However, in the conventional technique, the sensor sheet is used as one of the methods for grasping the perception / sensory stimulus received from the seat surface / back surface of the chair from the contact state between the chair and the human body. Since the body pressure distribution is measured and it is possible to know the strength of the support position when supporting the lumbar spine and what effect it has on the shape and function of the chair, the body pressure distribution Evaluation is a basic method for developing chairs and seats for furniture manufacturers and automobile seat manufacturers, but the types of collected data of body pressure distribution are limited, and the analysis method is also different from that of the experimenter. Judgment based on experience subjectivity is the main subject,
There was a problem that almost no time series analysis was performed on changes in body pressure distribution over a long period of time.

Therefore, the first aspect of the present invention is to measure and analyze the basic characteristics of the body pressure distribution with respect to the body characteristics, posture, chair shape and function, and the time elapsed after sitting. Aim to acquire a wide variety of data by adding consideration to some physical parameters (statistical values) such as pressure values, and to classify objective findings from the results. It aims to provide a basic characteristic measurement / analysis system for pressure distribution.

Further, in the second invention of the present invention, regarding the analysis of the body pressure distribution data, the physical quantity relating to the sensor sheet is evaluated by the direct multivariate analysis, the time series analysis and the analysis evaluation by the neural network.
It is an object of the present invention to provide a system for predicting / evaluating changes in body pressure over time, which predicts and evaluates the relationship between the physical quantity of a sensor seat and the sitting comfort, and associates the physical quantity of the sitting comfort with the functional quantity.

[0007]

In order to achieve the above object, the invention according to claim 1 is a body pressure distribution in which a body pressure distribution is measured by a sensor capable of measuring an effective pressure and is displayed and stored in a computer. In this measurement system, the seat or back surface of the chair or seat is placed so as to cover the seat surface and the back surface, and the seat pressure and back pressure of the human body are effectively measured using a sensor seat, and the seat pressure of the subject and the A body pressure distribution measuring means for displaying back pressure, a body characteristic analyzing means for measuring a body pressure distribution by the body pressure distribution measuring means for a certain number of subjects and considering differences in sex and body type, and for a certain number of subjects, The body pressure distribution is measured by the body pressure distribution measurement means, and the posture analysis means mainly considers the difference in various postures from the seat pressure, and the body pressure distribution for a certain number of subjects. The body pressure distribution is measured by the cloth measuring means, and the difference between the cushion material and the shape, and the difference in the function of the chair in the backward tilted posture are considered. And a time-dependent change analyzing means for measuring a body pressure distribution and considering a time course after sitting.

According to the second aspect of the present invention, the body pressure distribution measuring means displays the distribution measurement data of the seat pressure and the back pressure of the subject on the screen of the computer, and the number of sensors used and the average pressure value are used. It is characterized by performing statistical processing of various measurement data including standard deviation, variance, and maximum pressure value and screen display. Further, in the invention according to claim 3, the body characteristic analyzing means averages various data including the pressure values of the seat pressure / back pressure, the number of activity sensors, and the sensor coordinates in the distribution of the seat pressure by gender of the subject for each sex. Characterized by extracting gender characteristics as data and comparing various data of sitting pressure / back pressure classified by body type such as lean type, standard type and overweight type in addition to gender There is.

Further, in the invention according to claim 4, the posture analyzing means corresponds various data including a pressure value of body pressure on the seat surfaces of all the subjects, the number of activity sensors, and sensor coordinates to various sitting postures. It is characterized by considering various statistically processed data. In the invention according to claim 5, when the shape / function analyzing means of the chair takes a difference in cushion material and shape of the chair and takes a backward tilted posture,
It is characterized by distinguishing the difference in chair function from changes in seat pressure and back pressure. Further, in the invention according to claim 6, the time-dependent change analyzing means sets a sitting condition at each measurement of the body characteristic, posture, chair shape and function, and observes for a long time, and at every predetermined time. It is characterized by measuring body pressure distribution and extracting changes.

According to the present invention, the body pressure distribution is measured by a sensor capable of measuring the effective pressure, displayed and stored on a computer, and the correspondence between the physical quantity and the sensory quantity is analyzed and evaluated. In a pressure distribution evaluation system, a time series for viewing changes in pressure rhythm using a principal component analysis means for reducing statistical values of body pressure distribution and an AR (autoregressive) model for time series analysis based on the result of the principal component analysis. Analyzing means, burden degree predicting means for estimating partial autocorrelation coefficient as time series analysis and predicting burden rate by multiple regression analysis with burden degree, partial autocorrelation coefficient and rating by burden site Performs a canonical correlation analysis with and estimates the burden level for each burden site by means of the burden site predictor, and the neural network is used to create a self-organizing map by applying the body pressure distribution data to perform cluster classification to extract the body pressure pattern. Is characterized in that a body pressure pattern definition means for defining.

Further, in the invention according to claim 8, the body pressure pattern defining means extracts and defines a body pressure pattern in which the cluster and the burden evaluation are associated with each other by a radar chart of statistical values divided into clusters and a body pressure distribution. It is characterized by doing.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION A first embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a configuration diagram of a body pressure distribution basic characteristic measuring system according to a first embodiment of the present invention. FIG. 2 is a diagram showing an example of body pressure distribution data measured by the measurement system shown in FIG.

In FIG. 1, 1 is a chair to be measured,
Reference numeral 2 denotes a sensor sheet that is spread over the seat surface and the back surface of the chair 1 and that detects the body pressure distribution of the subject 3 sitting on the chair 1. Reference numeral 4 is a pelvic tilt angle sensor that detects the pelvic tilt angle. Reference numeral 5 is an A / D converter for A / D converting the detection current of the sensor sheet 2, and 6 is an A / D converter for A / D converting the output of the pelvic tilt angle sensor. 7 is A /
It is a computer that captures and processes the output of the sensor sheet 2 input via the D converter 5. 8 is A / D
It is a computer that takes in the output of the pelvic tilt angle sensor 4 via the converter 6 and processes it. Reference numeral 9 is a video camera for photographing the posture of the human body, and 10 is a work device for imposing keyboard input work on the subject.

In FIG. 2, 20 is seat pressure data displayed on the computer 7, and 21 is back pressure data. A color bar 22 displays a color corresponding to the pressure value (unit: mmHg). 23 is the number of used sensors, pressure average value, standard deviation, variance, and maximum pressure value displayed as statistical values (detection data) in the lower part of the drawing.

Next, the operation will be described. First, in measuring the body pressure distribution, as shown in FIG. 3, the pressure value increases due to an increase in conductivity due to factors such as temperature rise of the sensor 2 as shown in FIG. Compensate by unifying the time to improve reliability.

In the following, the body pressure distribution data for the body characteristics, posture, chair shape / function, and change over time will be measured and processed / held as computer software data. First of all, regarding physical characteristics, students who are physically and mentally healthy 10
1 person (5 males, 5 females) was used as a test subject to measure the body pressure distribution by the system shown in FIG. 1, and the measurement results were used 10 seconds after re-seating and classified by gender and body type. The data was measured. The body is visually thin,
The measurement was performed by classifying into a standard type and a fat type.

FIG. 4 is a diagram of the seat pressure distribution seen in men, and FIG. 5 is a diagram of the seat pressure distribution seen in women. In FIGS. 4 and 5, a straight line is drawn from the ischial tuberosity (upper part of the drawing) to the thigh (lower part of the drawing) with reference to the peak position of the pressure, and the pressure values on the straight line are graphed. Is obtained. Figure 6
4 is a graph of the linear pressure value drawn in FIG. 4 (for men), and FIG. 7 is a graph of the linear pressure value drawn in FIG. 5 (for women). In addition, this FIG.
The straight line indicates the average value of the pressure applied to the seat surface.

FIGS. 4 and 6 are men's seat pressure distribution charts, and FIGS. 5 and 7 are women's seat pressure distribution charts. From now on, men tend to support at the ischial tubercle and lower thigh, and the woman tends to support ischial tubercle. There is a tendency to support it as a whole. Further, as shown in the comparison diagram of the maximum pressure value of FIG.
The maximum pressure value tends to be high. Further, FIG. 9 is a comparison diagram of the maximum pressure value depending on the body type, and the pressure values are compared for the lean type, the standard type and the fat type. It was confirmed that larger men tended to have a body pressure distribution similar to that of women.

Next, regarding the posture, three students (two males)
The body pressure distribution was measured by the system shown in FIG. Regarding the measurement results, the data 10 seconds after reseating was used, and the pressure mainly applied to the sciatic tubercle was observed.

FIG. 10 is a graph showing the average of the maximum pressure values on the seat surface of three test subjects, and the maximum pressure value on the seat surface indicates the pressure at the ischial tuberosity. In this case, the posture in which the foot is placed forward or backward, the sacral sitting posture, and the reclining posture show low values. For postures with the foot placed forward or backward, the height from the ground to the knee changes as the position of the foot changes, and the pressure on the thigh increases and the pressure on the ischial tuberosity decreases accordingly. Is.

Further, in the sacral sitting or reclining posture, as shown in the diagram showing the measurement conditions in FIG. 11, the upright sitting posture is affected by the increase in the pressure applied to the backrest and the angle of the pelvis. it is conceivable that. From the measurement data of the pelvic tilt angle, it is considered that the pressure of the sciatic tubercle is reduced by the backward rotation of the pelvis. This is
It is also clear from the upright posture, the forward leaning posture, and the high pressure of the ischial tuberosity when seated in front of the chair.

Next, regarding the shape and function of the chair, three subjects (two males and one female) had a forward leaning posture, which is often seen during VDT (video display terminal) work.
The body pressure distribution was measured by the system of FIG. 1 for five chairs in the upright posture and the reclining posture.
With regard to the results, we used 10 seconds after re-seating, and examined the difference in cushion material and shape, and the difference in the function of the chair when taking a backward tilted posture.

FIG. 12 is a diagram showing the difference between the cushion material of the chair and the distribution of body pressure. There is a difference between the foam cushion (left) used this time and the mesh cushion (right) as shown in the figure. As shown in the explanatory view of FIG. 13, the foamable cushion sinks in a wide range around the contact portion, whereas the mesh-like cushion expands the contact portion, which causes tension. For this reason, there is little subsidence and the central region between the two points of the ischial tuberosity does not come into contact with the seat surface, so it is considered that no pressure is exerted.

FIG. 14 is an explanatory view of the body pressure distribution in the backward tilted posture. In the case of the chair (hereinafter referred to as a reclining mechanism) in which only the back surface in FIG. , The pressure at the thighbone region increases as the pressure at the ischial tuberous region decreases, and the upper body tends to be supported around the upper part of the backrest. Further, in the case of a chair (hereinafter referred to as a tilt mechanism) in which the seating surface is inclined together with the back surface shown in FIG. 14, there is little change in seating pressure, and there is a tendency to support the upper body around the waist. Further, FIG. 15 shows the body pressure distribution of the chair of the tilt mechanism in which the backrest has a strong horizontal curvature. In this case, it is considered that the upper body cannot be supported centering on the lumbar region because the pressure of the shoulder bone is strong.

Next, regarding the change over time, when the user sits on a chair or a seat for a long time, the feeling of the hips and waist changes,
The posture may be changed such as sitting down again. This time, the data of the changes in body pressure distribution over a long time are also included. 10 students (5 men, 5 women)
Body pressure distribution was measured for 20 minutes. During that time, subjects were given keyboard work.

FIG. 16 is a diagram showing changes in body pressure distribution over time. As an example, the height is 176 cm, and the weight is 59 kg.
3 shows the body pressure distribution every 5 minutes when the work is performed in the upright posture when the male subject is the subject. The change at this time is: 1. The pressure value rises with the passage of time. 2. The pressure value decreases due to re-seating and posture changes. 3. With the passage of time, changes in posture and displacement of the body occur, and pressure is transferred to the thigh. Such changes over time are shown.

Further, as shown in the graph of FIG. 17 showing the time series change of the pressure applied to the seat surface, the pressure value increases with time, but this is largely due to the influence of the sensor characteristics. However, the rapid change in the pressure value is due to a change in posture such as sitting down again. In addition, as shown in the time series change of the pelvis inclination angle in FIG. 18, the time series change of the pelvis inclination angle is relatively stable, but the pelvis tends to rotate forward every time the person is seated again. Can be seen.

FIG. 19 shows an extracted body pressure distribution when re-seating. When comparing the body pressure distribution before re-seating with the body pressure distribution after that, the pressure becomes lower after the posture change than before. This is also clear from FIG. Further, FIG. 20 is a diagram showing a displacement of the body and changes in the body pressure distribution.
In the case of 10, the pressure of the ischial tuberosity decreased and the pressure of the lower thigh increased for many subjects from the 10 minutes after the start, but this was due to body displacement as shown in FIG. It is considered that the applied pressure is transferred to the thigh. It is considered that the displacement of the body as shown in FIG. 20 causes the pelvis to rotate backward as shown in FIG. 18 and the pressure at the ischial tuberosity portion to decrease.

As described above, according to the present embodiment, the following classification data can be obtained from the objective knowledge about the basic characteristics of the physical pressure distribution "physical characteristics, posture, chair shape / function, time passage". It was 1. Regarding physical characteristics: a. Among men and women, men tend to have higher maximum pressure values. b. The back pressure distribution for men is wide, whereas the seat pressure for women tends to be wide. c. Large men tend to show similar body pressure distribution to women. d. A person with a thin body tends to have a small contact area and a high pressure. e. Overweight people tend to have lower maximum pressure values.

With respect to 2, posture, f, the position on which the foot is placed affects the seat pressure. g. As the pelvis rotates, the pressure applied to the ischial tuberosity changes.

3. Regarding the shape and function of the chair, there was a difference in body pressure distribution depending on h and the cushion material. i. The chair of the reclining mechanism tends to support the upper body centering on the upper part of the backrest by decreasing the pressure at the ischial tuberosity and increasing the pressure at the thigh. j. Tilt mechanism chairs show little change in seat pressure and tend to support the upper body centering on the waist. k, the stronger the back curve, the higher the pressure on the shoulder bones.

With respect to time 4, the pressure value increases with time m. n, pressure decreases due to re-seating and posture changes. o As time elapses, posture changes and body displacement occur, which causes the pressure in the ischial tuberosity to decrease and the pressure in the lower thigh to increase.

As described above, according to the present embodiment, the relationship between the body pressure distribution and "body characteristics, posture, chair shape / function, time passage" is influenced by individual differences such as body characteristics and posture. Since it is large, it is difficult to generalize, but for the purpose of selecting and producing a chair that suits the individual, it is necessary to extract the characteristics of the individual, so in that case, the above data is the CAD data of the chair / seat. It can be used as one of the important evaluation indexes by putting it in a database as such.

Next, a second embodiment of the present invention will be described with reference to the drawings. FIG. 21 is a flowchart of the process of the system for predicting / evaluating changes in body pressure distribution over time according to the second embodiment of the present invention. In the previous embodiment, various body pressure distribution data were measured, but in the second embodiment, for the sake of simplicity of explanation, two male and female university students were asked to sit on the chair for a long time and the angle of the chair was adjusted. Body pressure distribution data was measured under two conditions A and B, and three cushion materials from soft to hard, s, h, and m, in order, and the measured data was used as an objective evaluation method. ) Direct multivariate analysis, time series analysis,
It is a sensory evaluation such as sitting comfort by performing analysis / evaluation by neural network processing or the like.

In this case, one frame for measuring the body pressure distribution is 30 seconds, and the total experimental frame is 210, which is from 1 hour and a half to 2
It's time. At the same time, as a sensory evaluation, a burden evaluation is performed by a subject questionnaire using a burden evaluation form as shown in FIG. The burden rating is 5 steps, with 5 being the highest burden score, and 8 points from shoulder to thigh.

Details of the evaluation process will be described below with reference to FIG. First, as shown in the statistical value explanatory diagram of FIG. 23, as the statistical value of the body pressure distribution data, the number of sensors on the seat surface or the back surface (for example, the number of sensing points of the sensor sheet), the average pressure value, the standard deviation, Using 10 data of rate of change and maximum pressure value (S100), principal component analysis was used for these multivariate analysis to reduce and organize the statistical values, and a scatter plot was drawn with the reduced variables. The tendency of the chair for each condition is extracted (S101).

Specifically, in the principal component analysis for summarizing multidimensional data, as shown in FIG. 24, the 10-dimensional data shown in FIG. 23 is focused on a variable having a large absolute value of factor loading. It is reduced to two-dimensional data by the method of obtaining the principal component. This is seen from the diagram of the factor loading amount showing the result of the principal component analysis in FIG. 25. The pressure-related values of the first factor are 0.779, 0.
855 and 0.846, 0.7, and the number of sensors of the third factor 0.458, 0.542, the factor contribution ratio 56.
Since it reaches 1%, it is reduced to two-dimensional data of the average pressure value and the number of sensors.

Using the two-dimensional variables reduced as described above, a scatter diagram (horizontal axis is pressure average value, vertical axis is sensor number related value) of each condition (angle condition and hardness of material) is shown. FIG. 26-
It shows in FIG. The trends seen in these scatter plots are: a: long on the vertical axis, b: rising to the right, c: falling to the right,
d: bipolarization. First, in the case of a, the average pressure maintains a constant range regardless of the number of sensors. This is a material with a hard cushioning property, the angle condition is A, the condition A-h
This tendency can be seen in the chair shown in FIG.

In the case of b, the average pressure value increases as the number of sensors increases, and the plot points extend to the right. This is a condition of the soft material having a cushioning property, and the condition of the chair in FIG. 26 corresponds to the As chair, and this relationship is established by the cushioning property when the chair is supported wide and continues to be in a sitting posture for a long time.

In the case of c, the average pressure tends to decrease as the number of sensors increases, which corresponds to the chair of Am shown in FIG. 28, which is the material having the hardest cushioning property. This is considered to be a good sitting posture. Finally, in the case of d, the angle condition is B regardless of the cushioning property.
9 to 31, the average pressure decreases as the number of sensors increases, and it is considered that a good sitting posture can be achieved.

Next, as a result of the principal component analysis as described above, pressure values, which are reduced variables, are time-series analyzed (S10).
2). First, the AR model is used to convert it into a spectrum and observe changes in pressure value and pressure rhythm (S103). A
R model is Autoregressive
Model) model, which is a method that is routinely used for linear prediction processing of time series data. The results of time series analysis are shown in FIG. 32 to FIG. 35 in which the horizontal axis represents the period and the vertical axis represents the pressure value. Shows. Further, FIGS. 36 to 39 show the case where the pressure difference is taken in the same time series cycle (S104).
Note that FIG. 40 shows a burden evaluation of the chair in each case in this case.

The pressure value spectra shown in FIGS. 32 to 35 depend on the subjects 1 and 2 and the angle conditions and the cushioning characteristics (A-s, B-s, A-m) in this order, and the spectra are for 1 hour. No significant changes were observed during the experimental period of half to 2 hours.
When the unit of the pressure value is changed by taking the step difference of the pressure values in this time series cycle, the fluctuation can be seen under each condition as shown in FIGS. 36 to 39. Even if it looks stationary, it is non-stationary in the minute parts. This state is called a pressure rhythm, and the burden is predicted based on the presence or absence of the pressure rhythm and the burden evaluation result of FIG. 40 (S105). However, it is difficult to predict the degree of burden with this, so for the evaluation algorithm of the pressure rhythm, we leave it as reference data at this point and leave room for association with other evaluation indexes, and the system will be able to predict the pressure by such learning in the future. The rhythm evaluation algorithm shall be placed so that it can be updated and evolved in time series.

Therefore, the partial autocorrelation coefficient (PARCOR) is obtained as another calculated value in the time series analysis (S10).
6). Since the partial autocorrelation coefficient is defined as the correlation coefficient between the forward prediction error and the backward prediction error, it does not depend on the order of the model, and the periodic fluctuation between the data in the time series becomes clearer. A burden level prediction formula similar to the pressure rhythm is established by using the multiple regression analysis according to the fluctuation cycle (S10).
7).

The obtained formula is Y = φ ₁ × (−2.3) + φ ₃ × (−8.0) + φ ₄ ×
4.1 + 5.5 However, Y: burden rating φ: partial autocorrelation coefficient R ² = 0.8887 (coefficient of determination), and φ corresponds to a constant of the multiple regression model.

Substituting the partial autocorrelation coefficient of the new data as a verification for checking the validity of the equation, in the case of the subject 1 and the chair type A / s as shown in FIGS. As a result of substitution of the correlation coefficient, a predicted value Y = 3.5 was obtained.
Since the measured value was 3.1, the predicted value was valid. The above is the processing of the burden degree prediction means.

Further progressing the time series analysis, the burden site (Fig. 2
Canonical correlation analysis was performed using the partial autocorrelation coefficient and the score according to the burden site in order to investigate whether the burden level could be predicted for each of the 8 sites of 2 (shoulder to lower leg) (S108). Results are shown in FIG. 42. In the table of FIG. 42, X1 to X5 are partial autocorrelation coefficients from the 1st order to the 5th order, and Y6 to Y13 are the load ratings of the load parts in order from the top of the load evaluation sheet of FIG. F and G are composite functions of X and Y. The canonical correlation analysis aims to maximize the correlation between the canonical variables, and from this table, it is possible to extract the burden on the back, waist, and buttocks in the fourth-order partial autocorrelation coefficient.

Further, using a neural network, a self-organizing map as shown in FIG. 43 is created from the two-dimensional data of the number of sensors and pressure values obtained by contracting multidimensional data by principal component analysis. S109). The self-organizing map characteristically reproduces high-dimensional information on a two-dimensional plane visible to humans by learning neurons, and the self-organizing map in FIG. 43 is also created by learning about 1000 times.

In the self-organizing map of FIG. 43, set neurons are gathered in the y-axis direction. This represents the characteristics of the data of the body pressure distribution. From this result, as shown in FIG. 44, firing neurons are divided into three clusters. As a result, classification processing by the neural network is performed. This is subject 1 by principal component analysis,
2, the characteristics of the chairs A-s, A-h, A-m, etc. (FIG. 26-). This result is shown on the radar chart and the body pressure pattern at that time is extracted and viewed (S
110). An example thereof is shown in FIGS. 45 and 46 together with a radar chart of statistical values after clustering. However, the cluster 2 is omitted.

When the clusters in FIGS. 45 and 46 are associated with the burden evaluation, the buttocks are burdened in the cluster 1. Further, with respect to the cluster 3, the load on the buttocks, the waist, and the back is large. When a typical body pressure pattern in that case is extracted, it is true that the former has a high buttock pressure value, and the latter has a similarly high buttock pressure, which defines a pattern in which the back surface is not used (S11).
1).

With regard to the neural network, after the body pressure is classified (S110), there is still room for reference in association with other evaluation indexes in the future, and the prediction / evaluation algorithm also uses time series for each learning. Added and updated,
To be able to evolve. Thus, S100 to S111
In the prediction / evaluation algorithm up to, the pressure rhythm is limited to reference data in the three points of verification of pressure rhythm by AR model, prediction evaluation by partial autocorrelation coefficient, and prediction evaluation by neural network.
Predictive evaluation by partial autocorrelation coefficient is the main axis, and (AN
The mutual confirmation system is configured in the form of D) / or (OR) and prediction evaluation by a neural network, and comprehensive evaluation is performed (S112).

As described above, according to the second embodiment,
When evaluating the correspondence between physical quantities of body pressure distribution and sensory quantities such as sitting comfort, it becomes possible to reduce the statistical value of body pressure distribution by principal component analysis and handle the data. The body pressure data related to the chair was analyzed by replacing it with the pressure rhythm, and the multiple regression equation for predicting the degree of burden was obtained by the partial autocorrelation coefficient obtained by the time series analysis to improve the reliability of prediction. In addition, a canonical correlation analysis is performed using the partial autocorrelation coefficient and the burden, and prediction is performed for each burden site. The body pressure data is also clustered by a neural network, and the body pressure pattern is also extracted to determine the burden site. By performing another prediction and mutual confirmation, it becomes possible to perform highly reliable evaluation in the sitting posture for a long time using the physical quantity of the body pressure distribution and the psychological quantity of the subjective evaluation. It should be noted that although the description has been made so far on chairs and seats, the present invention can be applied to measurement of the soles of shoes and the like, and medical applications such as beds.

[0052]

As described above, according to the invention described in claim 1, the body pressure distribution measuring system for measuring the body pressure distribution by the sensor capable of measuring the effective pressure and displaying / storing it on the computer. In order to measure the seat pressure and back pressure of the human body by using a sensor seat that can be placed so as to cover the seat surface and back surface of the chair or seat, the seat pressure and back pressure of the subject can be measured on a computer. Body pressure distribution measuring means for displaying, body pressure analyzing means for measuring body pressure distribution for a fixed number of subjects by means of body pressure distribution measuring means, and considering differences in sex and body type, body pressure distribution measuring means for a fixed number of subjects The body pressure distribution is measured by the body pressure distribution measurement method and the posture analysis means mainly considers the difference in various postures from the seat pressure, and the body pressure distribution measurement means is used for a certain number of subjects. Sit by performing cloth measurement and considering the difference in cushion material and shape, and the function of the chair in the backward tilted posture, and the shape and function analysis means of the chair and the body pressure distribution measurement means for a certain number of subjects. Since it is equipped with means for analyzing changes over time afterwards, it is possible to widely measure body pressure distribution data as multivariate statistical values, and select and create items such as chairs and seats that suit individuals. In this case, the characteristics of the individual can be easily extracted, so that there is an effect that it can be used as a database as an important evaluation index.

According to the second aspect of the present invention, the body pressure distribution measuring means displays the distribution measurement data of the seat pressure and the back pressure of the subject on the screen of the computer, and the number of sensors used and the pressure average are used. Since various measurement data including values, standard deviations, variances, and maximum pressure values are statistically processed and displayed on the screen, there is an effect that observation, analysis, and evaluation of body pressure distribution data can be facilitated interactively. Also,
According to the invention of claim 3, the body characteristic analyzing means comprises:
Pressure values of sitting pressure and back pressure in the subject's gender pressure distribution,
The gender characteristics are extracted as average data of various data including the number of activity sensors and sensor coordinates for each gender, and various types of sitting pressure and back pressure are classified according to body type such as lean type, standard type and overweight type in addition to gender. Since the data are analyzed and classified by comparing the data, there is an effect that it is possible to easily identify the sex and body type by the body pressure distribution data.

Further, according to the invention described in claim 4, the posture analyzing means is configured so that the pressure value of the body pressure on the seat surfaces of all the subjects,
Since various data including the number of activity sensors and sensor coordinates are considered by various statistical processing data corresponding to various sitting postures, there is an effect that it is possible to determine a posture such as a forward leaning posture to sit on a chair from the body pressure distribution data. Further, according to the invention described in claim 5, the chair shape / function analyzing means determines the difference between the chair cushion material and the shape of the chair and the difference in the function of the chair when the seat is tilted backward. Since the determination is made based on changes in back pressure, there is an effect that the difference in chair shape and function can be easily identified from the body pressure distribution data.
Further, according to the invention described in claim 6, the time-dependent change analyzing means sets the sitting condition at each measurement such as body characteristics, posture, chair shape and function, observes for a long time, and at every predetermined time. Since the body pressure distribution is measured and the change is extracted, it is possible to analyze the change in the body pressure distribution over a long period of time.

According to the invention described in claim 7, the body pressure distribution is measured by a sensor capable of measuring the effective pressure, displayed and stored in a computer, and the correspondence between the physical quantity and the sensory quantity is analyzed and evaluated. In the body pressure distribution evaluation system, the principal component analysis means for reducing the statistical value of the body pressure distribution,
By means of multiple regression analysis with time series analysis means for observing changes in pressure rhythm using AR (autoregressive) model of time series analysis from the result of principal component analysis and partial autocorrelation coefficient as time series analysis. A burden level predicting unit that predicts the burden level by establishing a prediction formula, and a burden unit-based predictor that predicts the burden level by the burden site by performing a canonical correlation analysis using the partial autocorrelation coefficient and the score by the burden site. , A body pressure pattern defining means for extracting a body pressure pattern by defining a self-organizing map by multiplying the neural network with the body pressure distribution data and performing cluster classification. The effect is that a burden level evaluation system for a long sitting posture can be constructed.

According to the eighth aspect of the present invention, the body pressure pattern defining means defines the extraction and definition of the body pressure pattern for correlating the cluster with the burden evaluation based on the radar chart of statistical values divided into clusters and the body pressure distribution. Since this is performed, there is an effect that the body pressure pattern of the body pressure distribution at each burden site can be extracted and defined.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a body pressure distribution basic characteristic measuring system according to a first embodiment of the present invention.

FIG. 2 is a diagram showing an example of body pressure distribution measurement data by the measurement system shown in FIG.

FIG. 3 is a diagram showing an example of a rise in pressure value over time in the measurement system shown in FIG.

FIG. 4 is a diagram showing an example of the seat pressure shown in FIG. 2 in the case of a male.

FIG. 5 is a diagram showing an example of the seat pressure shown in FIG. 2 in the case of a woman.

FIG. 6 is a graph showing the seat pressure shown in FIG.

FIG. 7 is a graph showing the seat pressure shown in FIG.

8 is a diagram showing a bar graph comparing the maximum pressure values between men and women in the measurement shown in FIG.

9 is a bar graph showing a comparison of maximum pressure values according to body types in the measurement shown in FIG.

10 is a bar graph showing the difference in the pressure applied to the ischial tuberosity portion depending on the posture in the measurement shown in FIG.

11 is an explanatory diagram showing a difference due to an inclination angle in the measurement shown in FIG.

FIG. 12 is a diagram showing a difference between a cushion material of a chair and a body pressure distribution in the measurement shown in FIG.

FIG. 13 is an explanatory diagram showing a difference in cushion material shown in FIG. 12.

FIG. 14 is a diagram showing a case where a backward tilted posture is taken in the measurement shown in FIG.
It is a figure which shows the difference in the function of a chair, and body pressure distribution.

FIG. 15 is a diagram showing a body pressure distribution of the chair of the tilt mechanism shown in FIG. 14 in which the backrest is strongly curved in the horizontal direction.

16 is a diagram showing a body pressure distribution showing a change over time in the measurement shown in FIG.

FIG. 17 is a diagram showing a time-series change of the pressure value with the lapse of time shown in FIG.

FIG. 18 is a diagram showing a time-series change in the pelvic tilt angle with respect to the time-series change shown in FIG.

FIG. 19 is a diagram showing a change in body pressure distribution at the time of reseating while measuring the change with time shown in FIG. 16.

20 is a diagram showing a change in body pressure distribution due to a displacement of the body that occurs during the measurement of the change with time shown in FIG.

FIG. 21 is a flowchart of the process of the system for predicting / evaluating changes in body pressure distribution over time according to the second embodiment of the present invention.

22 is a diagram showing a burden evaluation form used in the evaluation system shown in FIG. 21.

23 is an explanatory diagram of statistical values used in the principal component analysis processing shown in FIG. 21.

24 is an explanatory diagram of the principal component analysis processing shown in FIG.

FIG. 25 is a diagram showing a factor loading amount as a result of the principal component analysis shown in FIG. 21.

FIG. 26 is a scatter diagram for the chair under the condition As of the principal component analysis shown in FIG. 21.

FIG. 27 is a scatter diagram for the chair under the condition Ah of the principal component analysis shown in FIG. 21.

28 is a scatter diagram for the chair under the condition A-m of the principal component analysis shown in FIG.

FIG. 29 is a scatter diagram for the chair under the condition Bs of the principal component analysis shown in FIG. 21.

FIG. 30 is a scatter diagram for the chair under the condition Bh of the principal component analysis shown in FIG. 21.

FIG. 31 is a scatter diagram for the chair under the condition B-m of the principal component analysis shown in FIG. 21.

32 is a diagram showing a spectrum in the case of subject 1 chair As in the time series analysis shown in FIG. 21.

FIG. 33 is a diagram showing a spectrum in the case of subject 2 chairs As in the time series analysis shown in FIG. 21.

FIG. 34 is a diagram showing a spectrum in the case of subject 1 chair B-s in the time series analysis shown in FIG. 21.

35 is a diagram showing a spectrum in the case of subject 1 chair A-m in the time series analysis shown in FIG. 21.

FIG. 36 is a diagram showing a case where a difference in pressure value is taken in the spectrum shown in FIG. 32.

FIG. 37 is a diagram showing a case where a difference in pressure value is taken in the spectrum shown in FIG. 33.

38 is a diagram showing a case where a difference in pressure value is taken in the spectrum shown in FIG. 34.

FIG. 39 is a diagram showing a case where a difference in pressure value is taken in the spectrum shown in FIG. 35.

FIG. 40 is a diagram showing a table of burden evaluation for each condition of the chair used in the time series analysis shown in FIG. 21.

41 is a diagram showing a partial autocorrelation coefficient shown in FIG. 21. FIG.

42 is a diagram showing the result of the canonical correlation analysis shown in FIG. 21.

FIG. 43 is a diagram showing a self-organizing map in the body pressure pattern extraction shown in FIG. 21.

FIG. 44 is an explanatory diagram of clusters associated with the self-organizing map shown in FIG. 43.

45 is a diagram showing a radar chart and body pressure of the cluster 1 shown in FIG. 44.

FIG. 46 is a diagram showing a radar chart and body pressure of another cluster shown in FIG. 44.

FIG. 47 is a diagram showing a structure of a conventional sensor sheet.

48 is a diagram showing sensitivity characteristics of the sensor sheet shown in FIG. 47.

[Explanation of symbols]

1 chair 2 sensor sheet 3 subjects 4 Pelvic tilt angle sensor 5, 6 A / D converter 7, 8 PC 9 video cameras 10 keyboard 20 Seat pressure 21 Back pressure 22 Color bar 23 Statistics

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Ryoko Miyake             3-14-16 Nakaochiai, Shinjuku-ku, Tokyo (72) Inventor Goro Fujimaki             48 17 Kudamachi, Kodaira-shi, Tokyo F-term (reference) 2F051 AA17 AB07 BA07 BA08                 4C038 VA20 VB40 VC20

Claims (8)

[Claims] 1. A body pressure distribution measuring system in which a body pressure distribution is measured by a sensor capable of measuring an effective pressure and is displayed / stored on a computer, wherein the chair or seat is arranged so as to cover a seat surface and a back surface. Body pressure distribution measuring means for displaying the seat pressure and back pressure of the subject on a computer by measuring the seat pressure and back pressure of the human body using a sensor sheet, and the body pressure for a certain number of subjects. The body pressure distribution is measured by the distribution measurement means and the difference in gender and body type is considered, and the body pressure distribution is measured by the body pressure distribution measurement means for a certain number of subjects, and the differences in various postures are mainly discussed. The posture analysis means to be considered from the pressure, and the body pressure distribution measurement by the body pressure distribution measurement means for a certain number of subjects to measure the cushion material and shape. Difference, and the difference in the function of the chair in the backward leaning posture, and the shape / function analysis means of the chair and the body pressure distribution measuring means for the body pressure distribution measuring means of a certain number of subjects, and the time course after sitting is considered. A basic characteristic measurement / analysis system for body pressure distribution, characterized by comprising: a time-dependent change analysis means. 2. The body pressure distribution measuring means displays the measurement data of the seat pressure and the back pressure of the subject on a screen of a computer, and the number of sensors used, average pressure value, standard deviation, variance and maximum pressure value. The statistical processing and screen display of various measurement data including the above are performed.
The system for measuring and analyzing the basic characteristics of body pressure distribution described in. 3. The body characteristic analyzing means extracts gender characteristics as data obtained by averaging various data including pressure values of seat pressure / back pressure, number of activity sensors, and sensor coordinates in a subject's gender-specific seat pressure distribution by gender. However, the data is analyzed and classified by comparing various data of sitting pressure and back pressure classified by body type such as lean type, standard type and overweight type in addition to the gender. Item 2. A basic characteristic measurement / analysis system for body pressure distribution according to item 2. 4. The posture analysis means considers various data including pressure values of body pressures on the seating surfaces of all subjects, the number of activity sensors, and sensor coordinates by various statistical processing data corresponding to various sitting postures. Claims 1-3 characterized by the above.
The basic characteristic measurement / analysis system for body pressure distribution according to any one of 1. 5. The chair shape / function analysis means determines the difference in chair material and shape between the chair and the difference in function of the chair when the chair is in a backward tilted position, based on changes in seat pressure and back pressure. The basic characteristic measuring / analyzing system for body pressure distribution according to any one of claims 1 to 4, wherein 6. The secular change analysis means sets a sitting condition at each measurement of the body characteristic, posture, chair shape and function, observes for a long time, and measures a body pressure distribution at every predetermined time. A change is extracted by extracting the change.
Basic characteristic measurement of body pressure distribution according to any one of 5
Analysis system. 7. A body pressure distribution evaluation system for measuring and displaying on a computer a body pressure distribution by a sensor capable of measuring an effective pressure, and analyzing / evaluating the correspondence between a physical quantity and a sensory quantity. Principal component analysis means for reducing statistical values of distribution, time series analysis means for observing changes in pressure rhythm using AR (autoregressive) model of time series analysis from the result of the principal component analysis, and time series analysis A canonical correlation analysis is performed by means of a burden degree predicting means for predicting the burden degree by establishing a prediction equation by multiple regression analysis with the partial autocorrelation coefficient and the burden degree, and the partial autocorrelation coefficient and the rating according to burden portion A body for predicting the degree of burden for each burdened part and a body for which a body pressure pattern is extracted by defining a self-organization map by applying the body pressure distribution data to a neural network to create a self-organizing map. Change over time prediction and evaluation system of the body pressure distribution is characterized in that a pattern definition means. 8. The body pressure pattern defining means extracts and defines a body pressure pattern that associates a cluster with a burden evaluation, using a radar chart of statistical values divided into clusters and a body pressure distribution. A system for predicting and evaluating changes in body pressure distribution over time according to.