AU2022201230A1

AU2022201230A1 - Method and system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision

Info

Publication number: AU2022201230A1
Application number: AU2022201230A
Authority: AU
Inventors: Guanglin Li; Ke Li; Leitong LIN; Na WEI; Jianhong Zhang
Original assignee: Shandong University
Current assignee: Shandong University
Priority date: 2021-02-23
Filing date: 2022-02-23
Publication date: 2022-09-08
Anticipated expiration: 2042-02-23
Also published as: AU2022201230B2; CN112998654A; CN112998654B

Abstract

The present disclosure provides a method and system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision, including a mirror system, a three-dimensional reflective marker system, a motion capture system, and an analysis system. When a subject performs a reach-to-grasp task on a shaded side of a mirror, a line of sight of the subject may observe a virtual target object under refraction of the mirror, but real-time postures of the arm and hand during grasping cannot be observed due to mirror shading. By using the mirror system, visual supervision is separated during reach-to-grasp. In addition, a sensorimotor function of the subject is evaluated by accurately recording kinematics parameter information of the finger. This system has important application value for neurophysiological tests, neurodevelopmental tests, hand function tests, and quantitative evaluation of sensorimotor functions with a variety of neuromuscular lesions. 1/4 5 6 4 /1 1 Y+7 13 12 FIG. 1 14 15 16 ) (P2) -------- - :~~!:~P2} J/~ 17 -i 18 FIG. 2

Description

1/4

5 6 4

/1 1

Y+7

13 12

FIG. 1 14 15

16

) (P2) -------- -

:~~!:~P2} J/~

17 -i

18

FIG. 2

METHOD AND SYSTEM FOR TESTING AND ANALYZING REACH-TO-GRASP KINEMATIC COORDINATION WITH AND WITHOUT VISUAL SUPERVISION TECHNICAL FIELD

The present disclosure belongs to the field of coordination test analysis technologies, and specifically relates to a method and system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision.

BACKGROUND

The description in this section merely provides background information related to the present disclosure and does not necessarily constitute the prior art. Reaching to grasp an object is one of the most common and important behaviors in daily life. This seemingly simple action is the result of a combination of multiple complex sensorimotor control mechanisms. Central nervous system decision-making is the core of action control and information processing. Therefore, the function state of the central nervous system can be learned by examining the reach-to-grasp kinematic coordination. During motion control of reach-to-grasp, vision plays a key role in observation and judgment on the position of a target object, and adjustment of a motion trajectory and a posture of the hand. The visual cortex in the central nervous system has a wide range of close structural correlations and functional pathways with a sensorimotor area and a primary motor cortex area, and forms a visuomotor mechanism, so that complex motion of the hand can be effectively coordinated, and fine motor of the upper limb and the hand is adapted to internal and external environments, a task target, and the like. When visual information is interfered during reach-to-grasp, a finger spacing increases abnormally, a correlation between a maximum spacing of the fingers and a target appearance disappears, and multiple peaks appear in a distance graph between fingers in the motion process, which is no longer a typical graph of a single peak in a normal visual condition. These phenomena indicate that weakening visual information during reach-to-grasp greatly affects kinematic coordination of the hand. This impact may become particularly prominent in various conditions of central or peripheral neuromuscular lesions, because neuromuscular lesions are often accompanied by sensorimotor impairment. Therefore, during reach-to-grasp, hand motion control is more dependent on compensation of visual information. However, a current test system cannot separate the visual information during reach-to-grasp, thereby making it impossible to highlight features of a dysfunction caused by neuromuscular lesions.

SUMMARY

To resolve the foregoing problems, the present disclosure proposes a method and system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision. The present disclosure may separate visual information during reach-to-grasp, so as to highlight features of a dysfunction caused by neuromuscular lesions, and provide an objective and accurate basis for diagnosing and evaluating multiple neuromuscular diseases. According to some embodiments, the following technical solutions are used in the present disclosure: A system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision, including: a mirror system, a three-dimensional reflective markers system, a motion capture system, and an analysis system, where: the mirror system includes a platform and a mirror vertically disposed on the platform, multiple markers are disposed in the mirror, a reflective surface and a shaded surface of the mirror form a symmetrical structure, and a side of the reflective surface of the mirror is configured to dispose a physical object; the three-dimensional reflective markers system includes a marker-cluster disposed on each of a thumb tip and an index finger tip, a marker attached to a joint of each of the thumb and the index finger, a marker-cluster on the hand back, and a single marker on the wrist, so as to represent a motion trajectory of each of the thumb and the index finger; the motion capture system includes multiple cameras disposed around the platform, which are configured to obtain a real target position, a mirror position, position information of a marker at a certain moment, and a motion trajectory of a hand marker-cluster in a process in which a subject grasps a virtual target; the analysis system is configured to perform conversion between a hand coordinate system and a mirror coordinate system, and establish a mathematical model of three-dimensional digital kinematics during grasping of the thumb and the index finger based on the converted motion trajectory; and extract an initial position and an end position in the converted motion trajectory, determine an optimal motion trajectory from the initial position to the end position by using a minimum-jerk trajectory model, and evaluate grasp kinematic coordination by comparing the mathematical model with the optimal motion trajectory. In an optional implementation, the hand marker set includes a nail marker-cluster used to mark a distal phalanx of a thumb and a distal phalanx of an index finger, markers used to mark an index finger joint and a thumb joint, a hand back marker-cluster used to mark a second metacarpal bone, and a wrist single marker used to mark a styloid process of a ulna. In a further definition, the nail marker-cluster includes a base, a cluster including three non-collinear markers is connected to the base by using a connector, and the base can be fastened to the nail by using a fastener, where the marker -clusters are non-collinear, and are used to define a local three-dimensional coordinate system fastened to each segment, and x, y, and z axes point to a right end, an upper end, and a proximal end respectively from a back angle. In an optional implementation, the single marker at each of the joint of the index finger and the joint of the thumb is used to determine a position of the following joint: an interphalangeal joint and a metacarpophalangeal joint of the thumb, and a distal interphalangeal joint, a proximal interphalangeal joint, and a metacarpophalangeal joint of the index finger. In an optional implementation, the motion capture system has fixed capture frequency. In an optional implementation, the analysis system includes a joint rotation center position recognition module, configured to specify a reference coordinate system from a marker-cluster fixed on a nail, and observe relative motion between markers on adjacent segments, where a conversion matrix of the reference coordinate system is defined by a rotation matrix fixed on a selected coordinate origin on the mirror coordinate system, and the conversion matrix is applied to collected marker position data, to perform spherical fitting and calculate a three-dimensional coordinate system of the joint center to obtain a joint center position on the mirror coordinate system. In an optional implementation, the analysis system includes a coordinate system conversion module, configured to define a hand coordinate system and a mirror coordinate system, and the two coordinate systems have a same origin to separately describe a unit vector in the hand coordinate system and the mirror coordinate system, convert a unit vector in the hand coordinate system into a vector in the mirror coordinate system, and determine a conversion matrix between the two. In an optional implementation, the analysis system includes a mathematical model construction module, configured to separately establish a mathematical model of three-dimensional digital kinematics during grasping of the thumb and the index finger based on degrees of freedom, a segment length, a center position, and a segment coordinate system of each fingerjoint. In an optional implementation, the analysis system includes a motion trajectory analysis module, configured to: set motion to start and end at a zero speed and a zero acceleration, construct a hand motion trajectory model based on a start position coordinate, an end position coordinate, and a movement time, determine, by using a dynamic optimization theory, a unique motion trajectory that produces optimal performance, and compare an acquired subject motion trajectory and the unique motion trajectory to evaluate a sensorimotor function of the subject. In a further definition, a specific process of comparing an acquired subject motion trajectory and the unique motion trajectory to evaluate a sensorimotor function of the subject includes: when the hand moves from an initial position to a final position within a given time, minimizing a time integral of a square of a jerk amplitude as the unique motion trajectory with the optimal performance, and comparing a collected motion trajectory with the unique motion trajectory to determine a response speed of the subject's hand grasp action and accuracy of the motion trajectory. Compared with the prior art, the present disclosure has the following beneficial effects: In the test analysis system provided in the present disclosure, when a subject performs a reach-to-grasp task on a shaded side of a mirror, a line of sight of the subject may observe a virtual target object under refraction of the mirror, but real-time postures of the arm and hand during grasping cannot be observed due to mirror shading, thereby separating visual supervision during reach-to-grasp. In the present disclosure, a sensorimotor function of the subject is evaluated by accurately recording kinematics parameter information of the finger. This system has important application value for neurophysiological tests, neurodevelopmental tests, hand function tests, and quantitative evaluation of sensorimotor functions with a variety of neuromuscular lesions.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings constituting a part of the present disclosure are used to provide further understanding of the present disclosure. Exemplary embodiments of the present disclosure and descriptions thereof are used to explain the present disclosure, and do not constitute an improper limitation to the present disclosure. FIG. 1 is a structural diagram of a test analysis system according to an embodiment. FIG. 2 is a coordinate system attached to each digital segment and a joint center corresponding to each digital segment according to an embodiment. FIG. 3(a) is a sagittal bitmap of an index finger according to the present disclosure, and FIG. 3(b) is a parametric index finger model diagram according to the present disclosure. FIG. 4 is a diagram of point-to-point motion trajectory comparison between a patient with a sensorimotor dysfunction and a healthy subject. FIG. 5 is a test flowchart according to an embodiment.

DETAILED DESCRIPTION

The present disclosure is further described below with reference to the accompanying drawings and embodiments. It should be noted that the following detailed descriptions are all exemplary and are intended to provide a further description of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by a person of ordinary skill in the art to which the present disclosure belongs. It should be noted that terms used herein are merely for describing specific implementations and are not intended to limit exemplary implementations according to the present disclosure. As used herein, the singular form is also intended to include the plural form unless the context clearly dictates otherwise. In addition, it should further be understood that, terms "comprise" and/or "include" used in this specification indicate that there are features, steps, operations, devices, components, and/or combinations thereof. In the present disclosure, orientation or positional relationships indicated by terms such as "upper", "lower", "left", "right", "front", "rear", "vertical", "horizontal", "side", "bottom" and the like are orientation or positional relationships shown in the drawings, merely relational words determined for ease of recitation of structural relationships of various components or elements of the present disclosure, are not specific to any component or element of the present disclosure, and are not to be construed as a limitation on the present disclosure. In the present disclosure, terms such as "fixedly connected", "interconnection", and "connection" should be understood in a broad sense. The connection may be a fixing connection, an integral connection or a detachable connection; or the connection may be a direct connection, or an indirect connection by using an intermediary. Relevant scientific research or technical personnel in the art may determine the specific meanings of the foregoing terms in the present disclosure according to specific situations, and such terms should not be construed as a limitation on the present disclosure. As described in the background, in an existing test system, visual information during reach-to-grasp cannot be separated. This embodiment is intended to provide a system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision. The system mainly includes a precisely calibrated mirror system, a set of three-dimensional reflective markers systems, a motion capture system, and an analysis system. When a subject performs reach-to-grasp task on a shaded side of a mirror, a line of sight of the subject may observe a virtual target object under refraction of the mirror, but real-time postures of the arm and hand during grasping cannot be observed due to mirror shading. By using the mirror system, visual is separated during reach-to-grasp. The analysis system evaluates a sensorimotor function of the subject by accurately recording kinematics parameter information of the finger. This system has important application value for neurophysiological tests, neurodevelopmental tests, hand function tests, and quantitative evaluation of sensorimotor functions with a variety of neuromuscular lesions. The key points of this test analysis system include the following aspects: First, a mirror test system is used. A key of the mirror system is to form a symmetrical structure on a reflective surface and a shaded surface of the mirror. A virtual target is formed in the mirror by using a mapping principle of the mirror. When a line of sight of a subject focuses on the virtual target, the subject reaches to complete grasp motion on a side of the shaded surface. In this case, the arm cannot be observed in the line of sight, and visual supervision is separated. The mirror system also cooperates with an optical motion capture system. Multiple cameras are arranged around the mirror system, and a specially specified three-dimensional reflective marker set is attached to a key part of a finger to precisely identify a motion trajectory of the finger. A capture target point and a positioning reference of the mirror include reflective markers. When the system is used to analyze reach-to-grasp motion, information such as a real-time posture of the hand, a capture target point, and a mirror position can be collected simultaneously. As shown in FIG. 1, there is a desktop, a mirror, a grasped object, and a reflective marker set, and a customized platform has a central slit to accommodate a high-resolution mirror and maintain a vertical position. The mirror is used to separate visual supervision to perform a reach-to-grasp task without visual feedback from a moving hand. When the mirror is configured, the grasped object is placed on a reflective side of the mirror. A subject can clearly gaze at a mirror reflective position of a target, so as to instruct the subject to regard a target reflected in the mirror as a virtual target. The subject pre-determines a spatial position of the virtual target when the position of the moving hand is not visible, and completes a reach-to-grasp task. In the figure, 1 is a real grasped object used to generate a virtual target, 2 is an experimental table, 3 is a central slit for placing a mirror, 4 is a mirror reflective surface, 5 is a mirror height, 6 is a mirror local marker (Mi, M 2 , and M 3 (M 3 is directly behind M 2 )), 7 is a grasped virtual target, 8 is an index fingernail marker, 9 is an index finger skin marker, 10 is a wrist joint marker, 11 is a hand back marker, 12 is a thumb skin marker, and 13 is a thumb nail marker. Certainly, in another embodiment, a position, a number, and the like of the mirror local marker may be changed according to a specific test environment and requirement. In addition, a mathematical model is established during the reach-to-grasp task. In this embodiment, thumb and index finger digital kinematics are comprehensively and precisely measured by using the hand marker set and the motion capture system. Then, a digital kinematics solution model is established during the reach-to-grasp task, and a three-dimensional position and a direction of each segment of the finger during the reach-to-grasp task are calculated. The mathematical model uses fewer markers, essentially reducing the impact of passive motion errors. This is detailed in the following: A mathematical model is established for each finger, and a finger includes three joints. The index finger is assumed to have 2 degrees of freedom at the joint of MCP2, 1 degree of freedom at the joint of PIP2, and 1 degree of freedom at the joint of DIP2, and the thumb is assumed to have 3 degrees of freedom at the joint of CMC1, 2 degrees of freedom at the joint of MCP1, and 1 degree of freedom at the joint of IP. Known and unknown kinematics quantities are parsed as index finger models. Coordinate systems attached to the hand

{MC2} and the distal phalanx of the index finger {DP2} are clearly known from the

labeled data, and are used as references for estimating OCP2 and 0 ODP2, as the method

mentioned in the first part. The lengths of the proximal (LPP 2 ) and middle (LMP 2 )

phalanx bone segments are also pre-estimated. The remaining unknown kinematics quantities

. . OPI2. PP2 IMP2) are the joint center positionPIP2 and segment coordinate systems and M.

FIG. 2 shows a coordinate system attached to each digital segment and a joint center

corresponding to each digit, where 14 is a distal phalangeal joint center ODIP2, 15 is a middle

phalangeal joint center 0PIP2, 16 is a proximal phalanx OMCP2, 17 is a distal phalangeal joint

center 01P of the thumb, and 18 is a proximal phalangealjoint center 0,P1 ofthethumb. In this embodiment, inverse kinematics and optimization techniques are used to calculate kinematics data of unknown kinematics quantities. The unknown kinematics quantities

include 21 solution elements (formed by a three-dimensional position of OPP2 and three

three-dimensional rotational axes ' of (PP2) and ( MP2) s= [OPIP2XMP2YMP2ZMP2XPP2 YPP2ZPP2

By connecting adjacent joint center positions, z of each segment is defined to be equal to a unit vector pointing to the segment in a longitudinal direction.

= ij - Oi+J 0i - Oi+||1

where is a segment, J denotes a direction vector of ,MC2 and "I-1"and"T"

denote "proximal" and "distal" respectively. By definition, axis x and axis Y are unit

vectors perpendicular to Z. xi y xJ = i _,y=_,xi = yi x zi, yi = zi x xi xi

It can be learned from rigid body kinematics and non-translational joint properties that

the previously specified segment lengths Lp2 and LP2 are invariant. The remaining system equations specify the degree of freedom of rotation at each joint and are considered as constraints met within a tolerance range, to calculate the joint rotation

angle by aligning ordered (x-y-z) Euler angles between coordinate systems with adjacent segments. It is assumed that the aligned rotation axis corresponds to structure

bending/elongation (X , extension/retraction (Y,±Y), and axial rotation ( (+z is considered as clockwise). Therefore, the constraint equation of the degree of freedom of the

joint at point 0 is:

xi = xi-(8- DOF), yi x xi-i = 0(2- DOF)

Parametric form of the solution: The dimension of the optimal solution is parameterized into three axial rotation angles, so as to increase the convergence speed of the solution. s'(0) =[10 0L 0 ]

These angles are used to derive the original solution vector S. According to convention, zer2 and zMP 2 initially belong to the plane containing YDP2 and a, a is a vector

connecting OMCP2 and DIP2, a known rotation amount 00 related to vector a clarifies

PIP2 and nd L2 fully determine {PP2 and MP2} for subsequent rotation of

ZPP 2 and ZMP2.

Using the constraint optimization routine fmioncon in the optimization toolbox and

solving s (0) with the Levenberg-Marquardt algorithm, an objective function secondarily minimizes the parametric axial rotation angle:

F= - 1=" 6, n

where i is an angle index, n = 3 is a total number of angles, and a tolerance between

the solution and the objective function is 0.05 rad, which is less than of a full angle

range0'2. A constraint tolerance is 0.1, that is, 5% of the interval -1,]in which the unit vector is located. The maximum allowed number of iterations and function evaluations is 1000. These tolerance values are manually selected so that most of the solution points converge on the assumed number of degrees of freedom assigned to each joint. In this embodiment, the foregoing method is used to establish a mathematical model during the reach-to-grasp task. It should be noted that, during the establishment of the digital kinematics solution model, all hand three-dimensional marker position data needs to be converted into that in a "global" coordinate system that is attached to the mirror. Specifically, a skeleton of a hand model is defined as a group of joints connected by a rigid segment. In this embodiment, a local conversion matrix of the joint is multiplied by a "motion matrix", that is, a rotation matrix of the specific joint represents motion of the model. In this embodiment, a problem processed by directional kinematics is to calculate a relative azimuth of a hand coordinate system relative to a mirror coordinate system. The hand

coordinate system is defined as Oxyz. Another global coordinate system is defined as OY The two coordinate systems have a same origin 0, and unit vectors corresponding to axes X,YZ of the hand coordinate system are ijk respectively. Unit vectors corresponding to axes X,Y, Z of the mirror coordinate system are respectively.

Therefore, by definition, the unit vectors in the mirror coordinate system may be represented as:

Correspondingly, the unit vectors i, j'k in the hand coordinate system may be

represented as:

IH 1,OO}T, H {O,1,O}T H {OO,1}T

Then, corresponding unit vectors i, j'k in the hand coordinate system are represented

in the mirror coordinate system. By using vector 1 as an example, coordinates of the unit

vector in the mirror coordinate system are written:

A component ixm of the coordinate axis X is analyzed, and the value of the

component is the length of the vector projected onto the axis X of the mirror coordinate system:

= cos(X I)= cos(i)

-7 Coisacsnofn where is a norm (length) of the unit vector 1, and os(Y,) sacosineofan

angle between vectors i and I.Because i and 1 are unit vectors, the foregoing formula may also be written as:

Cos,i) cos(Y, )= where is a scalar (dot) product of Y and I, to calculate a scalar product 1 Y. It does not matter in which coordinate system these vectors are measured for the scalar product, as long as they are represented in the same system because rotation does not change the angle

between vectors. Therefore, I. =IHYH = M. TM M M = cos(IH IH)= cos(I . I )

For simplicity, Y-I, JKk, and superscripts of cos(',T), cos(, j), and

cos (,k)are omitted in the following parts of this embodiment. Similarly, we can get: Mm

Y and z

Therefore, now the vector 1 in the mirror coordinate system can be written as

- Ii. K . In addition, similarly, M =. and

ikJ j k Tcan be obtained.

Now, J in the hand coordinate system have a complete set of representations in the mirror coordinate system. These vectors can form a convenient matrix:

I-II- I- Fcos(II) cos(I,j) cos(I, k) =T,-i J-j J-k = cos(J,i) cos(J,j) cos(J,k) =DCM M K-i K-j K.kj cos(K,i) cos(K,j) cos(K,k)d

The matrix is a direction cosine matrix. Apparently, the matrix is composed of cosines of all possible combinations of angles of unit vectors in the hand coordinate system and the mirror coordinate system. _H -H -H

Similarly, unit vectors I , JK in the mirror coordinate system are represented in the

hand coordinate system in an essentially symmetrical way, and iJK and i'j3' may be simply exchanged for implementation, and a result is: IH -7 i H -j . - }j H

A form of the composited matrix is as follows:

I-I I cos(I,i) cos(J,i) cos(K.I) IJK =[j J K = os(I,j) cos(J,j) cos(Kj) =DCMH

I-k J-k K-k cos(I,j) cos(J,k) cos(K-k) -B _ B B jB

Now assuming that the vector ' ry ' in the hand coordinate system, let us

determine its coordinates , r1f in the mirror coordinate system by using the

known rotation matrix DCMM. Starting from the first coordinate component r M -M -M-M

r r cos(I ,r )

Rotation does not change the scale of the vector, and does not change the angle between two vectors that experience the same rotation, so if we represent some vectors in different rotational coordinates, the norm and angle between the vectors will not change: -M -H H-M -H

. In addition, cos(I ,r )=cos(I ,r); depending on this attribute, we can derive:

-M, -M) -H-H -H, -H -H -H -H T -M M =r OS(I ,r r COS(I,r) r)=irx= 'TH HzH

The above formula is substituted into 'I-k to obtain: M -H -H H H H r =I*r rHI-i+r, j+rI-k

In the same manner, the following formulas may be represented:

rM = rHj. 1rH j.+rH j

M = HT Hk H

Finally, this may be described as a matrix form:

rM rm X-iT7 -- r"

e M]LK- H FLLI 1:r = CMG -H

K.j K rH

Finally, the subject is instructed to configure a hand marker set and complete a reach-to-grasp task. After that, a motion trajectory signal generated by during reach-to-grasp needs to be analyzed, and key parameters are extracted to evaluate the sensorimotor function of the subject. For this reason, a minimum-jerk trajectory (MJT) model is proposed, which is used to evaluate the qualitative characteristics and quantitative details observed in a plane multi-joint motion experiment, and is used to analyze motion coordination. The model is mathematically modeled by defining an objective function, and the dynamic optimization theory is used to determine the unique motion trajectory that produces the optimal performance. Then, the unique motion trajectory is compared with a real-time motion trajectory of the subject, and a smaller difference indicates good motion coordination. Specifically, the motion trajectory of the subject's hand is obtained in real time by using the motion capture system. The "trajectory" refers to planning and control of motion of the hand in a kinematics aspect, specifically, a configuration of the hand in space and a motion speed at which the hand moves from an initial positiontoafinalposition. Researchers believe that the simplicity of motion control is achieved by planning a hand trajectory in external space. To describe this behavior, a spontaneous mathematical model of a hand motion tissue is proposed. In the work described herein, the dynamic optimization theory is used, and a standard function for describing a motion target needs to be defined during dynamic optimization. The function is mathematically represented as a time integral of a performance indicator, and depends on system input, output, and internal variables. A set of differential equations for describing a response of a system to the input is established. When the hand moves from the initial position to the final position within a given time tf , the standard function for minimizing the target is the time integral of the square of the jerk amplitude:

1 f4, d X + dy C=2- dtY dtYJdt

C is the standard function, and x and Y are the hand position coordinates. The

mathematical expressions of x(t) and y(t) will be found so that the standard function in the above equation is minimal. A variational method and an optimal control theory method are used to search for a minimum-jerk trajectory model. The mathematically predicted trajectory model is compared with an experimental motion record, and a difference between the motion trajectory of the subject and the MJT model is analyzed, so as to evaluate the sensorimotor function of the subject. The trajectory model is subject to dynamic constraints imposed by system differential equations and algebraic constraints imposed at an end point or during motion.

Unconstrained point-to-point motion: It is known that x(t) and y(t) calculated according to the foregoing equations are minimum jerk trajectory of a fifth-order polynomial in time, and the standard function determines a form of the motion trajectory. Generally, for

any function x(t), it is differentiable in a time period 06tt, and any performance index I

t,x,xJx,x,...d" - -df 1] is integrated in a time interval to obtain an unconstrained motion function:

C(x(t))= Itx, dt I dt" ]

When x(t) is a solution of the Euler-Poisson equation, an extreme value may be assumed: 0I d 8I d" 0I .+ (-1)" =0 ax dtOax dt" ax"

I= + YI. 2 is known, and the following equation is further obtained:

d3 aY2 d3 J2 3 dt ay dts69 3=0

We can separate these terms according to two position components to obtain: y dxX= 0 ddY= 0 6 dt 6 dt Assuming that the motion starts and ends at a zero speed and a zero acceleration, finally, expressions of the motion trajectory of the hand, that is, the MJT model are as follows:

x(t)=x"+(x"-x")(-10( ) +15( )4-6( )5)

y(t)=y'+(y'- y')(-10( )+15( )4-6( )5) tf tf tf

(xss indicates the start position coordinates, (x'Y) indicates the end position

coordinates, and t" indicates the movement time. The point-to-point motion trajectories of a patient with a sensorimotor dysfunction and a healthy subject are shown in FIG. 4. 19 is original kinematics data, 20 is MJT models of the patient and the healthy subject, 21 is the motion start points, and 22 is the motion end positions. By comparison, the motion coordination of the subject can be analyzed. The foregoing embodiment provides the system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision. First, thumb and index finger kinematics are determined by using the hand marker set and the motion capture system. In this embodiment, the motion capture system may use an existing system. In this embodiment, fewer markers and comprehensive analysis methods are used to implement precise measurement of joint digital kinematics. In this embodiment, the proposed method includes coordinate system calibration and estimation of the solution model of three-dimensional digital kinematics. In addition, the MJT model is used to analyze finger motion coordination. In a specific test, a process shown in FIG. 5 may be used. First, prepare for a test, including system calibration, pasting markers and collecting basic information of a subject. After the subject is familiar with test steps, start a formal test. For each "grasp" test, the subject is required to specify the specific operation flow and precautions before the start of the test. The subject is sitting at a specified position. The ulnar side of the right hand is placed in a specified start area of the subject. A PC sends a "test start" instruction to the subject. The subject contacts the virtual target with the thumb and the index finger, and gently pinches the virtual target presented on the mirror with the two fingers to complete the test at a natural speed. Because the visual feedback from the moving hand is separated by the mirror, the subject finds the start area via the tactile sense, the area has a significantly different texture, and three unrelated fingers (the middle finger, the ring finger, and the little finger) are bent. After multiple times of experimental training, the subject carries out a total of 10 (or other number of times) consecutive reach-to-grasp tests with and without visual supervision. The break time between tests is several seconds. Certainly, if another solution is to set another type of motion capture system to obtain real-time kinematics data of the same grasped object, or simply modify the geometry of the grasped object, or change only the placement position of the mirror operating system, the solution shall be considered as the same invention as the present disclosure. The foregoing descriptions are merely exemplary embodiments of the present disclosure, but are not intended to limit the present disclosure. The present disclosure may include various modifications and changes for a person skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure shall fall within the protection scope of the present disclosure. The specific implementations of the present disclosure are described above with reference to the accompanying drawings, but are not intended to limit the protection scope of the present disclosure. A person skilled in the art should understand that various modifications or deformations may be made without creative efforts based on the technical solutions of the present disclosure, and such modifications or deformations shall fall within the protection scope of the present disclosure.

Claims

CLAIMS What is claimed is:

1. A system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision, comprising: a mirror system, a three-dimensional reflective marker system, a motion capture system, and an analysis system, wherein: the mirror system comprises a platform and a mirror vertically disposed on the platform, multiple markers are disposed in the mirror, a reflective surface and a shaded surface of the mirror form a symmetrical structure, and a side of the reflective surface of the mirror is configured to dispose a physical object; the three-dimensional reflective marker system comprises a marker-cluster disposed on each of a thumb tip and an index finger tip, a marker attached to a joint of each of the thumb and the index finger, a marker-cluster on the hand back, and a single marker on the wrist, so as to represent a motion trajectory of each of the thumb and the index finger; the motion capture system comprises multiple cameras disposed around the platform, which are configured to obtain a real target position, a mirror position, position information of a marker at a certain moment, and a motion trajectory of a hand marker-cluster in a process in which a subject grasps a virtual target; the analysis system is configured to perform conversion between a hand coordinate system and a mirror coordinate system on each point in a motion trajectory of a collected hand marker-cluster, and establish a mathematical model of three-dimensional digital kinematics during grasping of the thumb and the index finger based on the converted motion trajectory; and extract an initial position and an end position in the converted motion trajectory, determine an optimal motion trajectory from the initial position to the end position by using a minimum-jerk trajectory model, and evaluate grasp kinematic coordination by comparing the mathematical model with the optimal motion trajectory.

2. The system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision according to claim 1, wherein the multiple hand marker-clusters comprise a nail marker-cluster used to mark a distal phalanx of the thumb and a nail marker-cluster used to mark a distal phalanx of the index finger, a marker used to mark a joint of the finger and a marker used to mark ajoint of the thumb, a hand back marker-cluster used to mark a second metacarpal bone, and a wrist single marker used to mark a styloid process of a ulna.

3. The system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision according to claim 2, wherein the nail marker-cluster comprises a base, a cluster comprising three non-collinear markers is connected to the base by using a connector, and the base can be fastened to the nail by using a fastener, wherein the marker-clusters are non-collinear, and are used to define a local three-dimensional coordinate system fastened to each segment, and x, y, and z axes point to a right end, an upper end, and a proximal end respectively from a back angle.

4. The system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision according to claim 1, wherein the single marker at each of the joint of the index finger and the joint of the thumb is used to determine a position of the following joint: an interphalangeal joint and a metacarpophalangeal joint of the thumb, and a distal interphalangeal joint, a proximal interphalangeal joint, and a metacarpophalangeal joint of the index finger.

5. The system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision according to claim 1, wherein the motion capture system has fixed capture frequency.

6. The system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision according to claim 1, wherein the analysis system comprises a joint rotation center position recognition module, configured to specify a reference coordinate system from a marker-cluster fixed on a nail, and observe relative motion between markers on adjacent segments, wherein a conversion matrix of the reference coordinate system is defined by a rotation matrix fixed on a selected coordinate origin on the mirror coordinate system, and the conversion matrix is applied to collected marker position data, to perform spherical fitting and calculate a three-dimensional coordinate system of the joint center to obtain a joint center position on the mirror coordinate system.

7. The system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision according to claim 1, wherein the analysis system comprises a mathematical model construction module, configured to separately establish a mathematical model of three-dimensional digital kinematics during grasping of the thumb and the index finger based on degrees of freedom, a segment length, a center position, and a segment coordinate system of each fingerjoint.

8. The system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision according to claim 1, wherein the analysis system comprises a coordinate system conversion module, configured to define a hand coordinate system and a mirror coordinate system, and the two coordinate systems have a same origin to separately describe a unit vector in the hand coordinate system and the mirror coordinate system, convert a unit vector in the hand coordinate system into a vector in the mirror coordinate system, and determine a conversion matrix between the two.

9. The system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision according to claim 1, wherein the analysis system comprises a motion trajectory analysis module, configured to: set motion to start and end at a zero speed and a zero acceleration, construct a hand motion trajectory model based on a start position coordinate, an end position coordinate, and movement time, determine, by using a dynamic optimization theory, an optimal unique motion trajectory to be generated, and compare an acquired subject motion trajectory and the unique motion trajectory to evaluate a sensorimotor function of the subject.

10. The system for testing and analyzing reach-to-grasp kinematic coordination with and without visual supervision according to claim 9, wherein a specific process of comparing an acquired subject motion trajectory and the unique motion trajectory to evaluate a sensorimotor function of the subject comprises: when the hand moves from an initial position to a final position within a given time, minimizing a time integral of a square of a jerk amplitude as the unique motion trajectory with the optimal performance, and comparing a collected motion trajectory with the unique motion trajectory to determine a response speed of the subject's hand grasp action and accuracy of the motion trajectory.