TWI581757B - System and method for evaluating the quality of joint mobility - Google Patents

Description

Joint activity quality estimation system and method

The present invention relates to a joint activity estimation system and method, and more particularly to a system and method for estimating joint activity quality.

It is known that the estimation of joint activity quality is applied to the field of class determination or rehabilitation of limb disorders, which involves measuring the quality of joint activities and evaluating the joint function of the affected part.

In the evaluation of the joint function of the affected part, the hand function evaluation is taken as an example, including the Minnesota test and the Jebsen test. The test content of the Minnesota test is that the test subject uses one hand (dominant or non-dominant hand). Turn over 60 pieces and place them in the groove of the board. After turning over 60 pieces, record the time of the roll. You must test twice and add the time of the two flips. Look up the table to classify the patient's hand function damage assessment. The sub-project of the Jaberson test is to use five hands to flip five 3 cm by 5 cm cards and record the time spent in order to analyze the hand movement sensitivity of the patient's thinner parts.

However, the existing test methods are not able to quickly and effectively know the quality of the joint activity and evaluate it.

Accordingly, it is an object of the present invention to provide a joint activity quality estimation system and method that overcomes the deficiencies of the prior art.

Therefore, the joint activity quality estimation system of the present invention is applied to a joint to be tested which is interposed between two limbs and rotates in an extended state and a flexed state, the system comprising at least two inertial sensing modules, a recording unit and a processing unit.

Each of the inertial sensing modules is respectively disposed on two limbs connected to the joint to be tested to sense joint motion between the limbs to generate a multi-axis inertial sensing data. The recording unit records a reference material relating to the angle of movement of the joint to be tested. The processing unit is electrically connected to the inertial sensing module, and when the joint to be tested is flexed or stretched, the buckling or stretching activity of the joint to be tested relative to the extended state is converted according to each of the multi-axis inertial sensing data. An activity angle, and interacting with the reference material according to the activity angle to calculate an evaluation data related to the activity quality of the joint to be tested.

The joint activity quality estimation method of the present invention is applied to a joint to be tested which is rotated between two limbs and rotated in an extended state and a flexed state, and the method comprises the following steps: (a) obtaining respectively connected to the joint to be tested At least two inertial sensing modules on the two limbs sense a joint motion of each of the limbs and respectively generate a multi-axis inertial sensing data; (b) when the joint to be tested is subjected to flexion or stretching activities, Each of the multi-axis inertial sensing data to convert an active angle of the flexion or extension activity of the joint to be tested relative to the extended state; and (c) providing a reference material relating to an active angle of the joint to be tested, and The activity angle interacts with the reference material to calculate a joint activity related to the joint to be tested. Dynamic quality assessment data.

The effect of the invention is that the joint activity quality estimation system of the invention can automatically sense the activity of the joint to be tested and correspondingly the activity angle when the activity is output and can automatically evaluate the joint function, which can not only accurately measure but also save manpower and time. cost.

100‧‧‧ Joint Activity Quality Estimation System

2‧‧‧Computer host

3‧‧‧Inertial Sensing Module

31‧‧‧First inertial sensing module

32‧‧‧Second inertial sensing module

33‧‧‧The third inertial sensing module

10‧‧‧Processing unit

11‧‧‧Wireless transmission unit

5‧‧‧ joints to be tested

51, 52‧‧‧ knuckles

Other features and effects of the present invention will be apparent from the embodiments of the present invention, wherein: FIG. 1 is a system diagram illustrating an embodiment of the joint activity quality estimation system of the present invention; FIG. 2 is a schematic view The three-axis tilt angle of the single inertial sensing module is illustrated; FIG. 3 is a schematic diagram illustrating the tilt angle ψ and the angle θ _xe between the X-axis direction and the horizontal plane of the inertial sensing module; FIG. 4 is a schematic diagram illustrating the finger Figure 5 is a schematic diagram illustrating the angular range of the tilt angle θ _{atan 2} of the different inertial sensing modules; FIG. 6 is a schematic diagram illustrating the inward flexion of the finger relative to the extended state FIG. 7A to FIG. 7D are schematic diagrams illustrating the calculation manner of the movement angle of the finger during movement; and FIGS. 8 and 9 are waveform diagrams illustrating the motion angle uncompensated filtering processing and the motion subjected to the compensation filtering processing. Angle timing waveform.

Referring to FIG. 1 , in the embodiment of the present invention, the joint activity quality estimation system 100 includes a plurality of sensing modules 3 , a processing unit 10 , and a wireless transmission unit 11 .

The number of the inertial sensing modules 3 is a plurality, and the first inertial sensing module 31 and the first one are measured by taking the joint 5 to be tested between the knuckles 51 and 52 of one of the fingers as an example. The two inertial sensing modules 32 are respectively disposed on the knuckles 51, 52 connecting the joints to be tested 5 to sense joint motion between the phalanxes 51, 52 to generate a multi-axis inertial sensing data.

The processing unit 10 is electrically connected to the inertial sensing modules 3, and the multi-axis inertia of the first inertial sensing module 31 and the second inertial sensing module 32 is used when the joint to be tested 5 is flexed or stretched. The measured data converts an active angle of the joint to be tested 5 relative to the flexion or extension of the joint in an extended state, wherein the extended state is defined as a state in which the two phalanxes or the limbs are in a straight line. In addition, in addition to the first inertial sensing module 31 and the second inertial sensing module 32, other third inertial sensing modules 33, such as a third inertial sensing module 33, and the second inertial sensing module 32 can also be used for multi-axis inertial sensing data. The calculation results of the processing unit 10 are similarly obtained to obtain a plurality of sets of estimation results relating to the moving angle and the joint activity quality, and are not limited by the first inertial sensing module 31 and the second inertial sensing module 32.

In this embodiment, the inertial sensing modules 3 are disposed on a glove, and the number thereof is sixteen. Each of the inertial sensing modules 3 may be, but not limited to, a six-axis inertial sense using the model LSM330DLC. The module 3 is tested, and each of the inertial sensing modules 3 is disposed on each glove on the glove corresponding to the hand joints of the important joints, and in addition to the finger knuckles, the front of the hand and the front of the palm An inertial sensing module 3 is also disposed at the arm.

It should be noted that although the present embodiment is described by taking a single finger joint activity quality as an example, it is known to those skilled in the art that the estimation of the activity quality of other types of limb joints, such as: wrist joints, various Finger activity should also fall within the scope of the present invention.

In this embodiment, each inertial sensing module 3 includes an accelerometer that can output three-axis acceleration, a gyroscope that can output three-axis angular velocity, and/or a magnetometer that can output a three-axis magnetic direction. (magnetometer), therefore, each inertial sensing module 3 can output data such as acceleration, angular velocity and/or magnetic direction of a plurality of axes. It is to be noted that it is known to those skilled in the art that other types of inertial sensing modules 3 including three-axis, six-axis or nine-axis may not be limited by the components of the model of the embodiment, as long as The sensing data includes acceleration, angular velocity, magnetic direction, angle, and the like, all of which fall within the scope of the present technology.

The processing unit 10 may be, but is not limited to, transmitting the data sensed by the inertial sensing module 3 to the computer host 2 via the Bluetooth wireless transmission unit 11 by using the model of the MSP 430, and performing the algorithm by the computer host 2 Calculation and processing.

Referring to FIG. 2, the method calculates the Acc _total value according to the sensing data of the three-axis accelerations Acc_X, Acc_Y, and Acc_Z of the inertial sensing module 3, wherein the Acc _total value represents the total vector of the acceleration three axes, as in Equation 1. Show.

Next, use the trigonometric function to calculate the tilt angle of the triaxial acceleration (Acc_X, Acc_Y, Acc_Z) to facilitate the calculation of the tilt angle of the three axes. , ρ , θ , as shown in Equation 2 to Equation 4.

Referring to FIG. 3, using the tilt angle of the three-axis acceleration, the motion angle is calculated by the projection vector. First, Equation 8 is derived using Equation 5 to Equation 7, and the tilt angle of the previously calculated triaxial acceleration is added to Formula 8. calculation, and calculates θ _xe, θ _xe inertia sensing module is the angle X-axis direction and the horizontal plane.

Next, the atan2 function is used instead of the arctan function of Equation 8, because the atan2 function is used for calculation, wherein the value range of the atan2 function is (-π, π), and the range of values in this mode is large, and the angle can be reduced. The calculated error, the relationship between the θ _{atan 2} function and arctan is shown in Equation 9.

Referring to FIG. 4, the angle relationship between θ _xe and θ _{atan 2} is shown, θ _xe is the angle between the X-axis direction of the inertial sensing module and the horizontal line; θ _{atan 2} is the X-axis direction and the line segment (L) of the inertial sensing module. Angle.

Referring to FIG. 5, a positive or negative value of the angular range of θ _{atan 2} is defined as: the angles of the quadrants I and II are greater than 0, and are 0 degrees to plus 180 degrees; the angles of the quadrants III and IV (angle) It is less than 0 and is 0 degrees to minus 180 degrees.

Referring to FIG. 6 , the first inertial sensing module 31 and the second inertial sensing module 32 are respectively disposed on the knuckles 51 , 52 of the joint 5 connecting the fingers to be tested to sense the between the phalanx 51 , 52 . The joint 5 is in an active state with respect to the extended state with respect to the finger, and the movement angle when the finger is flexed inward as shown in the figure is θ _bend . It should be emphasized that the definition of the angle range of the phalanx 51, 52 of the joint 5 is different, and the motion angle measured by this method ( θ _bend ) is a medical range of motion, which can replace the finger. The joint protractor is used for measurements.

7A to 7D, and in conjunction with the positive or negative value of the angular range of the tilt angle θ _{atan 2} of FIG. 5, when the finger is moving, the motion angle ( θ _bend ) is calculated as shown in Equation 6, and its motion is as shown in Equation 6. calculating an angle of the foregoing need to use the calculated first inertia sensing module 31 and the second inertia sensing module 32 respective θ _{atan 2,} the following formula will be θ ₁ represents a first inertia sensing module 31 is [theta] _The angle of the _{atan 2} is θ ₂ representing the angle θ _{atan 2 of the} second inertial sensing module 32, and the calculation of the angle of motion ( θ _bend ) is performed by dividing the combination of θ ₁ and θ ₂ into four states. To perform analytical calculations.

See Figure 7A, when θ ₁ 0 and θ ₂ >0, the formula 10 is applied.

θ _bend = Abs ( θ ₂ - θ ₁ ) Equation 10

See Figure 7B, when θ ₁ 0 and θ ₂ 0, the formula 11 is applied.

θ _bend = Abs ( θ ₁ - θ ₂ ) Equation 11

Referring to Fig. 7C, when θ ₁ > 0 and θ ₂ > 0, Equation 12 is applied.

θ _bend = Abs ( θ ₂ - θ ₁ ) Equation 12

Referring to Figure 7D, when θ ₁ >0 and θ ₂ 0, applies Equation 13.

θ _bend =360-( θ ₁ - θ ₂ ) Equation 13

When the finger joint is active, due to the error of its angle, in the method, the compensation angle filter is used to correct the movement angle to make the calculated angle more precise. The main purpose of the integrator is It is a calculation method for integrating the angular velocity into an angle; a low-pass filter is a passage that allows a low-frequency signal to pass, but attenuates (or reduces) a signal whose frequency is higher than the cut-off frequency; a high-pass filter (High-pass filter) It is a filter that allows high-frequency signals to pass, but weakens (or reduces) the passage of frequencies below the cutoff frequency.

It should be noted in advance that other compensation methods may be used in addition to the aforementioned compensation methods, and are not limited by those mentioned in the embodiment. For example, when the angle between the plane of the motion angle and the horizontal angle is not lower than the predetermined angle, the movement angle is calculated by using one of an accelerometer, a gyroscope, and a magnetometer, and the angle between the plane of the motion angle and the horizontal angle is inclined. When the degree is lower than the predetermined angle, the angular velocity calculation of the other of the accelerometer, the gyroscope, and the magnetometer is utilized. The angle of motion.

The compensation of this embodiment is as shown in Equation 14.

Angle ( n )=0.9 *( Angle ( n -1)+( Gyrorate * interval _ time ))+0.1 * Acc _angle formula 14

When the inclination of the movement angle exceeds 60 degrees, the angular velocity (gyro value) is used to calculate the integral of the angular velocity to measure the motion angle, as in Equation 14.

Angle ( n )=0.9 *( Angle ( n -1)+( Gyrorate * interval _ time ))+0.1 * Acc _angle formula 15

In Equation 14 and Equation 15, Acc _angle represents the motion angle before filtering, Angle (n) represents the current motion angle after compensation filtering; Angle (n-1) represents the motion angle calculated before the compensation filter; Gyrorate represents the motion angle. In the calculation, the difference between the angular velocity values (gyro values) of the first inertial sensing module 31 and the second inertial sensing module 32; the interval time at which the interval_time reads the data.

Referring to FIG. 8 and FIG. 9 , the moving angle timing waveform of the continuous motion angle uncompensated filtering processing is as shown in FIG. 8 and the motion angle timing waveform processed by the compensation filtering is as shown in FIG. 9 , and it can be clearly seen that the compensation filtering processing can be obtained well. effect.

The method can perform an evaluation of the hand function test based on the obtained angle of motion, including the following methods.

First, the statistical data comparison: collect a large number of hand function data of normal subjects, establish the normal pattern of each hand function test, and normal joint and patient joint activity data and normal The patterns are compared to assess the dissimilarity of the two.

Using the joint activity quality estimation system 100 of the present invention in the Minnesota test to measure the hand function of a normal subject and the hands of an abnormal subject (eg, minor injury, stroke, or Parkinson's disease leading to hand dysfunction, etc.) The motion angle timing waveform of the function can be used to objectively evaluate the hand function by subsequent signal processing such as frequency domain analysis, and also achieve the effect of fast and effective evaluation.

Second, self-data comparison: compare the data of each test action of the testee with the average of the previous or past three data of the test subject, if the test data and past data Similarly, it can be judged as normal joint activity; on the contrary, each time the joint activity data has different differences, it can be judged to be abnormal joint activity.

The method adopts the Dynamic Time Warping (DTW) method in data evaluation. The dynamic time adjustment method mainly compares the similarity of the waveforms of the two curves and gives a distance value if the waveforms of the two curves are extremely similar. The distance value calculated by the dynamic time adjustment method will be smaller; conversely, if the waveforms of the two curves are very different, the distance value calculated by the dynamic time adjustment method will be larger.

As shown in Table 1, the result of this evaluation is the result of flipping the chess piece 20 times in the Minnesota test. The observed finger angle data is the first knuckle part of the index finger, and the analysis method is based on the normal person's movement angle. The timing waveform uses the dynamic time adjustment method to compare the similarity of the waveforms of different curves. Among them, the algorithm 1 compares each curve with the normal mode curve, and the algorithm 2 compares each curve with the first three average curves, and Algorithm 3 is a comparison of each curve with the previous curve, because the movement angle changes more stable, The calculated distance plus the total value is small.

As shown in Table 2, the result of this evaluation is the result of flipping 20 pieces in the Minnesota test. The observed finger angle data is the first knuckle part of the index finger, and the analysis method is based on hand dysfunction. The time series waveform of the patient's motion angle uses the dynamic time adjustment method to compare the similarity of the waveforms of different curves. Among them, the algorithm 1 compares the curve of each curve with the normal mode, and the algorithm 2 compares each curve with the first three averages. The comparison of the curves, and the third algorithm is the comparison of each curve with the previous curve. Because the movement angle of the hand dysfunction is unstable, the total distance calculated by the comparison table 1 is larger.

In summary, the effect of the present invention is that the joint activity quality estimation system 100 of the present invention can automatically sense the activity of the joint to be tested 5 and correspondingly the angle of motion when the activity is output, and can automatically evaluate the joint activity quality, which can not only accurately The measurement can also save labor and time costs, so it is indeed possible to achieve the object of the present invention.

However, the above is only the embodiment of the present invention, and the scope of the present invention is not limited thereto, that is, the simple equivalent changes and modifications made by the patent application scope and the patent specification of the present invention are still It is within the scope of the patent of the present invention.

100‧‧‧ Joint Activity Quality Estimation System

2‧‧‧Computer host

3‧‧‧Inertial Sensing Module

31‧‧‧First inertial sensing module

32‧‧‧Second inertial sensing module

33‧‧‧The third inertial sensing module

10‧‧‧Processing unit

11‧‧‧Wireless transmission unit

5‧‧‧ joints to be tested

Claims (8)

A joint activity quality estimation system is applied to a joint to be tested that is rotated between two limbs and rotated in an extended state and a flexed state. The system comprises: at least two inertial sensing modules, each of the inertial sensing modules Arranging respectively on the two limbs connected to the joint to be tested to sense joint motion between the limbs to generate a multi-axis inertial sensing data; a recording unit recording a moving angle of the joint to be tested And a processing unit electrically connecting the inertial sensing modules, and when the joint to be tested performs flexion or stretching, converting the joint to be tested relative to the extended state according to each of the multi-axis inertial sensing data An activity angle of the flexing or stretching activity, and interacting with the reference material according to the activity angle to calculate an evaluation data related to the activity quality of the joint to be tested; wherein the processing unit calculates the evaluation data by The timing waveform of the active angle is similar to the waveform of the waveforms included in the reference data by the dynamic time adjustment method. Method calculated from the sum value represented by a low quality joint activities preferred. The joint activity quality estimation system according to claim 1, wherein the recording unit records a plurality of reference materials related to an activity angle corresponding to at least one joint function; the processing unit interacts the activity angle with the reference materials A self-assessment data relating to the joint activity quality of the joint to be tested is calculated. The joint activity quality estimation system according to claim 1, wherein the recording unit is a reference material for recording an activity angle related to a normal joint function; and the processing unit associates the activity angle with the reference material Evaluation data on whether the joint activity quality of the joint to be tested is abnormal. The joint activity quality estimation system of claim 1, wherein each inertial sensing module comprises an accelerometer, a gyroscope, and a magnetometer, and the processing unit determines the angle of motion. When the inclination of the plane and the horizontal angle is not lower than a predetermined angle, the movement angle is calculated by using one of the accelerometer, the gyroscope, and the magnetometer; and the processing unit determines the plane and level of the motion angle When the angle of inclination of the angle is lower than the predetermined angle, the angle of motion is calculated using the accelerometer, the gyroscope, and the angular velocity of the other of the magnetometers. A joint activity quality estimation method is applied to a joint to be tested that is rotated between two limbs and rotated in an extended state and a flexed state, and the method comprises the following steps: (a) obtaining respectively connected to the joint to be tested At least two inertial sensing modules on the two limbs sense a joint motion of each of the limbs and respectively generate a multi-axis inertial sensing data; (b) when the joint to be tested is subjected to flexion or stretching activities, Each of the multi-axis inertial sensing data to convert an active angle of the flexion or extension activity of the joint to be tested relative to the extended state; and (c) providing a reference to the activity angle of the joint to be tested And calculating an evaluation data related to the activity quality of the joint to be tested according to the activity angle, and calculating the evaluation data by comparing the time series waveform of the activity angle with a dynamic time adjustment method The degree of similarity of the waveforms of the plurality of curves included in the reference data indicates that the joint activity quality is better if the distance plus the total value calculated by the dynamic time adjustment method is lower. The joint activity quality estimation method according to claim 5, wherein the step (c) is to record a plurality of reference materials related to the activity angle corresponding to the at least one joint function to interactively calculate the activity angle with the reference materials. A self-assessment data relating to the quality of joint activity of the joint to be tested. The joint activity quality estimation method according to claim 5, wherein the step (c) is to provide a reference information about an activity angle of the normal joint function to interact the activity angle with the reference material. Evaluation data on whether the joint activity quality of the joint to be tested is abnormal. The joint activity quality estimation method according to any one of claims 5 to 7, wherein each inertial sensing module comprises an accelerometer, a gyroscope, and a magnetometer. And step (a) is further configured to calculate the movement angle by using one of the accelerometer, the gyroscope, and the magnetometer when determining that the angle between the plane and the horizontal angle of the motion angle is not lower than a predetermined angle; And when the processing unit determines that the angle between the plane of the motion angle and the horizontal angle is lower than the predetermined angle, the motion angle is calculated by using the accelerometer, the gyroscope, and the angular velocity of the other of the magnetometers.