JP3594574B2 - System and method for quantitative evaluation of walking motion - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention observes a walking motion that represents a basic motor function of a human, compares the case where a healthy person walks in a natural state with the walking of a subject with a walking function disorder, and quantifies the walking motion. Technical fields that require a comprehensive evaluation method.
[0002]
[Prior art]
Conventionally, in order to obtain walking data by observing a walking motion, a miniature bulb, a reflecting bulb, or the like is attached as a marker to each joint part of the subject, and the image is digitally video-walked by walking on a preset reference line. There is a system in which a camera captures and observes and records the locus of light spots caused by a miniature bulb attached to each joint.
In this system, it is necessary to walk in a relatively dark room in order to obtain a sufficient amount of light, and the image is taken by a home NTSC (National Television System Committee) digital video camera, and this image is taken into a personal computer. For each frame of the image, the coordinates of the light spot were input by manually clicking the mouse.
[0003]
[Problems to be solved by the invention]
In the conventional walking motion data input method, since the coordinates are manually input, it takes a long time to complete the data, and real-time processing cannot be performed.
In addition, since the photographing format is the NTSC system, the sampling cycle is only 30 Hz, and it was not possible to reproduce even minute movements such as trembling of the body of the subject.
Furthermore, since it is necessary to take an image in a relatively dark room, there is a risk that a subject having a problem with the walking function may fall.
[0004]
[Means for Solving the Problems]
The present invention has been made in order to solve the above-described problem, and performs walking data input in real time by automatic input, increasing a sampling cycle to reproduce even fine movements such as body tremors, and a relatively bright room. But it allows data entry.
That is, the sampling cycle is accelerated by a combination of a marker LED (light emitting diode) and a PSD (position detecting element), the LEDs are sequentially turned on one point at a time, and the states of when the LEDs are turned on and when they are not turned on are detected, and the extraneous light is detected. Get rid of the effects.
In addition, the lighting of the LEDs is not simply lit, but is modulated to include identification information of the LED of which channel is lit, demodulated on the PSD side, and distributed to each point. Output.
[0005]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of a quantitative evaluation system of a walking motion according to the present invention, and FIG. 2 is a circuit diagram showing a configuration of a PSD processing unit.
FIG. 3 is an explanatory diagram showing a relationship between a subject having a marker LED fixed to each joint site and a PSD sensor unit for measuring the walking motion. FIGS. 4 and 5 show a sequence of signal waveforms accompanying LED lighting. Is shown.
[0006]
As shown in FIG. 1, the quantitative evaluation system of walking motion according to the present invention measures the walking motion of a subject by a PSD sensor unit 4 and performs signal processing by a PSD 3 and a signal processing circuit 2 on a change in the light spot position of a marker LED. It consists of a gait measurement system that measures and measures data, a gait measurement data processing system that pre-processes and displays and graphs the measured data, and a gait measurement data modeling system that analyzes and discriminates the pre-processed data. I have.
The walking measurement data processing system and the walking measurement data modeling system are stored in the personal computer 12 as programs.
[0007]
In FIG. 3, marker LEDs 16 _{1 to} 168 are attached to _{eight positions} of the subject's top, shoulders, elbows, hands, hips, knees, heels, and feet, and the subject walks on a preset reference line.
The PSD sensor unit 4 is installed at a position perpendicular to the central portion on the reference line, and measures the walking motion of the subject walking on the reference line from the lateral direction.
The measurement of the walking motion is performed in a walking range where tracking of the PSD sensor unit 4 is not required by the wide-angle lens.
[0008]
The marker LED is composed of a group of a plurality of LEDs in order to obtain sufficient brightness and wide directivity, and can be fixed to each joint of the body by Velcro tape or the like.
The control unit with a built-in LED control circuit 17 having a function of blinking sequentially one by one light marker LED (not shown), being adapted to be fixed on the waist or the like by a belt or the like, each marker LED 16 ₁ between to 16 ₈ are connected by a dedicated cable (not shown).
[0009]
4, there is shown a signal waveform CH ₁ to CH ₈ by spot output from the eight markers LED which is fixed for each joint portion, by applying a modulation to each signal waveform, a channel identification It consists of bits and bits for outputting a continuous signal for position measurement.
In addition, since the marker LEDs are sequentially turned on in a time-division manner with a time lag, the movement of each joint part is clearly displayed sequentially as a locus of light spots, and the signal waveform sequence is as shown in FIG.
[0010]
FIG. 2 is a circuit diagram showing a circuit configuration of the PSD unit.
The blinking of the light spot of the LED via the lens 1 provided in the PSD sensor unit 4 is captured by the PSD (position detecting element) 3 and becomes two-dimensional coordinate data (analog). Since the sampling frequency of the PSD is 150 to 300 Hz, by setting the frame rate to 300 frames / second, it is possible to display a more detailed walking activity state than that of the NTSC video camera (30 frames / second), and it is possible to display fine movements such as trembling of the subject's body. Can be reproduced. The LED lighting cycle and the PSD cycle are configured to be synchronized.
[0011]
The output signal from PSD3 is input to the X-coordinate detecting circuit ₆₁ and the Y-coordinate detecting circuit 6 _2, respective output signals (analog) is converted into a digital signal is input to the A / D conversion circuit 7 ₁ and 7 ₂ You.
Furthermore, the quantity of received light outputted from the PSD is input to the light quantity detecting circuit 5, the output signal is input to the X-coordinate detecting circuit ₆₁ and the Y-coordinate detecting circuit 6 ₂ via the decoder 9, the register 10 ₁ to 10 type _8, samples the digital signal output from the a / D conversion circuit 7 ₁ and 7 _2, to determine how many channel-th LED output, automatically output in real time.
[0012]
The output signal from the register ₁₀ 1 to 10 ₈ is transmitted to the personal computer 12 through the interface 11.
The personal computer 12 includes a gait measurement data processing system that pre-processes data obtained from the gait measurement system to display and graph the data, and analyzes the gait measurement data obtained by the pre-processing to obtain a healthy person and a person with a gait disorder. For each of the cases, an analysis program comprising a gait measurement data modeling system for modeling data and quantifying physical disorders is stored.
[0013]
The gait measurement data processing system is a program that preprocesses digital data obtained from the gait measurement system, and displays and graphs the data.
This program includes a calculation unit, a drawing unit, and a processing unit.
The calculation unit consists of a Fourier transform function for obtaining a power spectrum for characterizing walking (as comparison data with others) for each individual, and a multivariate analysis for obtaining a correlation matrix for finding features in an individual. Has functions.
The drawing unit has a function of displaying the result of the calculation unit as a power spectrum graph and a matrix of coefficients (corresponding to a human body measurement position).
The processing unit is responsible for the whole program progress, such as reading digital data, inputting it to the calculation unit, and outputting the output from the calculation unit to the drawing unit and the gait measurement data modeling system.
[0014]
The gait measurement data modeling system is a program that performs multivariate analysis of digital data of the gait measurement data processing system and quantifies an apparent or latent physical disability. In addition, we perform analysis to construct quantification.
This program includes an input interface unit, a multivariate analysis unit, and an output interface unit.
In multivariate analysis, principal component analysis and discriminant analysis are mainly performed, and from walking data arranged as a correlation matrix, data is modeled for each case of a healthy person and a person with gait disability, and the characteristics are analyzed. The identification of the failure is statistically performed by comparing the extracted evaluation parameters.
[0015]
Although the above description is an example using a single PSD sensor unit, three-dimensional measurement can be performed by using a plurality of PSD sensor units, and the possibility of losing the marker LED can be reduced. it can.
Further, by using the infrared LED and the infrared filter 18 for the lens, the influence of the background light can be reduced.
[0016]
【The invention's effect】
As described above, the system and method for quantitatively evaluating walking motion according to the present invention can automatically and in real time obtain walking motion data, and increase the sampling frequency of the body to increase the sampling frequency. It has the effect of reproducing even small movements such as tremors.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a quantitative evaluation system of a walking motion according to the present invention.
FIG. 2 is a circuit diagram showing a configuration of a PSD processing unit.
FIG. 3 is an explanatory diagram showing a relationship between a subject having a marker LED fixed to each joint site and a PSD sensor unit for measuring a walking motion of the subject.
FIG. 4 is a sequence of a signal waveform accompanying LED lighting.
FIG. 5 is a signal waveform sequence associated with LED lighting.
[Explanation of symbols]
1 lens 2 signal processing circuit 3 PSD
4 PSD sensor unit 5 Light amount detection circuits 6 ₁ , 6 ₂ X coordinate, Y coordinate detection circuits 7 ₁ , 7 ₂ A / D conversion circuit 9 Decoders 10 ₁ to 10 ₈ Register 11 Interface 12 Personal computer 13 Walking measurement data processing system 14 Walking measurement data modeling system 15 tripod ₁₆ 1 to 16 ₈ marker LED
17 Marker LED control circuit 18 Filter

Claims (4)

A marker LED fixed to each joint site of the subject walking on the reference line, and a control unit carried by the subject, which controls lighting of the marker LED;
A PSD (Position Detecting Element) that two-dimensionally captures a change in the position of a light spot by the marker LED obtained by measuring the walking activity of the subject, and converts an analog signal output by the PSD into a digital signal. A walking measurement system configured by a PSD sensor unit installed at a right angle at the center of the reference line, which is converted and transmitted to a personal computer;
A gait measurement data processing system that pre-processes and displays and graphs digital data from the PSD sensor unit, and analyzes gait measurement data obtained by the gait measurement data processing system, and analyzes the gait measurement data of healthy persons and persons with gait disorders. A personal computer that stores a gait measurement data modeling system that models each case and quantifies specific obstacles as an analysis program,
Quantitative evaluation system for walking motion, characterized by comprising: Step of sequentially blinking the lighting of the marker LED fixed to each joint site of the subject walking on the reference line, and displaying the walking activity of the subject as a light spot locus;
The subject's walking activity is measured by the PSD sensor unit installed at a right angle at the center of the reference line, and the change in the light spot position of the marker LED is captured two-dimensionally by the PSD built in the PSD sensor unit. Computing the output data of the PSD, A / D converting and bundling the data, and transmitting the digital data to a personal computer via an interface;
Reading the input digital data into a calculation unit, obtaining a power spectrum based on the Fourier-transformed data, connecting an output from the calculation unit to a drawing unit, and displaying a calculation result in a power spectrum or a matrix,
Multivariate analysis of the correlation matrix data obtained by processing the gait measurement data from the above step to obtain a discriminant function or an evaluation function, modeling of data for each case of a healthy person and a gait disability person, A step of statistically enabling the identification of a failure by comparing the extracted evaluation parameters with each other. A method for quantitatively evaluating a walking motion, the method comprising: Marker LEDs fixed to each joint part of the subject are sequentially lit in a time-division manner with a staggered time, and each marker LED is modulated to include a channel identification bit in the signal waveform from each LED. The method for quantitatively evaluating a walking motion according to claim 2, wherein The walking motion of the subject is simultaneously measured by a plurality of PSD sensor units equipped with a PSD, and the trajectory of the light spot position of the marker LED captured by each PSD sensor unit is input to one personal computer for data processing. The method for quantitatively evaluating a walking motion according to claim 2, wherein:

Images