Limb Movement Function Assessment System

Abstract

The present invention provides a limb movement function evaluation system, which can be applied to the technical field of movement evaluation. The limb movement function evaluation system includes: a display; a first wearable sensor worn on the limb of an object; and a processor connected to the display and the first wearable sensor, configured to: control the display to sequentially display position indication images of K regions to sequentially indicate position points in the K regions to the object, and simultaneously receive first position data from the first wearable sensor to obtain first limb position data of the K regions, the K regions are distributed in different orientations of the object in the space where the object is located, and K is an integer greater than 1; determine limb movement information of the K regions according to the first limb position data of the K regions; generate a limb evaluation result of the object according to the limb movement information of the K regions. The present invention can objectively evaluate the movement performance of the object's limb in each region and improve the accuracy of limb movement function evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of motion assessment, and particularly to a limb motion function assessment system. Background Art

[0002] For stroke survivors, due to damage to their brain tissue, upper limb motor dysfunction may occur.

[0003] Motion function assessment is important for the overall rehabilitation process and final rehabilitation effect of patients. In the related art, the accuracy of assessing the motion function of patients is low. Summary of the Invention

[0004] In view of the above problems, the present invention provides a limb motion function assessment system.

[0005] According to a first aspect of the present invention, there is provided a limb motion function assessment system, including: a display; a first wearable sensor worn on a limb of an object; and a processor connected to the display and the first wearable sensor, configured to: control the display to sequentially display position indication images of K regions to sequentially indicate position points in the K regions to the object, and simultaneously receive first position data from the first wearable sensor to obtain first limb position data of the K regions, the K regions are distributed in different orientations of the object in the space where the object is located, and K is an integer greater than 1; determine limb motion information of the K regions according to the first limb position data of the K regions; and generate a limb assessment result of the object according to the limb motion information of the K regions.

[0006] According to an embodiment of the present invention, the above system further includes an image collector; the processor is further connected to the image collector and is further configured to: after determining the first limb position data of the (k - 1)th region, control the display to display the position indication image of the kth region, and simultaneously control the image collector to collect a spatial image of the space, the position indication image of the kth region indicates the position point of the kth region; identify the limb position point of the object in the collected spatial image; and in the case where the position point of the kth region matches the identified limb position point, determine the first position data collected during the display period of the position indication image of the kth region as the first limb position data of the kth region, where k is a positive integer less than or equal to K.

[0007] According to an embodiment of the present invention, the first limb position data of the k-th region includes I pieces of first limb position data; the position indication image of the k-th region includes I position indication images, and each of the I position indication images indicates a different position point within the k-th region, where I is an integer greater than 1; the processor is further configured to: after determining the (i - 1)-th limb position data, control the display to display the i-th position indication image, and at the same time control the image collector to collect a spatial image of the space; identify the limb position point of the object from the collected spatial image; and in the case where the i-th position point of the i-th position indication image matches the identified limb position point, determine the first position data collected during the display period of the i-th position indication image as the i-th first limb position data, where i is a positive integer less than or equal to I.

[0008] According to an embodiment of the present invention, the processor is further configured to: in the spatial image collected at the j-th moment, identify the j-th limb position point; match the j-th limb position point with the i-th position point, so as to, in the case where the j-th limb position point matches the i-th position point, obtain the display duration of the i-th position indication image according to the start display moment of the i-th position indication image and the j-th moment.

[0009] According to an embodiment of the present invention, the system further includes a second wearable sensor worn on the torso of the object; the processor is further configured to: for the k-th region among the K regions, determine the k-th motion trajectory information and the k-th motion displacement information of the object's limb according to the first position data of the k-th region and the second position data collected by the second wearable sensor; calculate the k-th motion speed information of the object's limb according to the k-th motion trajectory information and the display duration of the position indication image of the k-th region; and determine the limb motion information of the k-th region according to the k-th motion trajectory information, the k-th motion displacement information, the display duration, and the k-th motion speed information, where k is a positive integer less than or equal to K.

[0010] According to an embodiment of the present invention, the position indication image of the k-th region includes I position indication images, and the limb motion information of the k-th region includes I pieces of limb motion information, where I is an integer greater than 1; the processor is further configured to: according to the i-th position indication image among the I position indication images, determine the i-th standard motion value from the standard motion values of the k-th region, where i is a positive integer less than or equal to I; determine the i-th motion difference value according to the i-th limb motion information and the i-th standard motion value among the I pieces of limb motion information; determine the k-th region motion difference value according to the statistical value of the I motion difference values; and generate a limb evaluation result according to the K region motion difference values.

[0011] According to an embodiment of the present invention, the processor is further configured to: perform a motion function evaluation on the limb motion information of K regions to obtain a limb motion evaluation value; and generate a limb evaluation result according to the limb motion evaluation value and the K-region motion difference values.

[0012] According to an embodiment of the present invention, the processor is further configured to: determine a motion recommendation region from the K regions according to the limb motion evaluation value and the K-region motion difference values; and control the display to indicate the motion recommendation region to the object.

[0013] According to an embodiment of the present invention, the processor is further configured to: obtain the limb information to be evaluated of the object; and determine the position indication images of the K regions from the position indication images of P kinds of limbs, where each of the P kinds of limbs is in a different direction of the same body, and P is a positive integer greater than 1.

[0014] According to an embodiment of the present invention, the above system further includes a reference sensor placed statically in the space where the object is located; the processor is also connected to the reference sensor, and the processor is further configured to: calibrate the coordinate system of the first wearable sensor based on the coordinate system of the reference sensor, and collect first position data through the first wearable sensor according to the relative position between the reference sensor and the first wearable sensor.

[0015] According to an embodiment of the present invention, the processor controls the display to sequentially display the position points of different regions and simultaneously receives the first limb position data of different regions collected by the first wearable sensor. Since the first wearable sensor is worn on the limb of the object, the first limb position data of the object's limb in each region can be freely and flexibly collected through the first wearable sensor. Thus, according to the first limb position data of different regions, the limb motion information of the object's limb in different regions can be determined. In this way, based on the limb motion information of the object's limb in different regions, the motion performance of the object's limb in each region can be accurately and objectively evaluated, improving the accuracy of the motion evaluation of the object's limb. And, since the present invention can accurately evaluate the motion of the object's limb only using a processor, a display, and a wearable sensor, it has the advantages of low cost, high precision, unrestricted use scenarios, no need for professional physicians to participate, and can perform motion evaluation independently without supervision to meet the needs of frequent evaluation during the rehabilitation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Through the following description of the embodiments of the present invention with reference to the drawings, the above content and other objects, features, and advantages of the present invention will become clearer. In the drawings:

[0017] Figure 1A Schematically shows a schematic diagram of a limb motion function evaluation system according to an embodiment of the present invention.

[0018] Figure 1B Schematically shows a schematic diagram of an object performing an action according to an embodiment of the present invention.

[0019] Figure 2 Schematically shows a schematic diagram of the position points of K regions according to an embodiment of the present invention.

[0020] Figure 3A Schematically shows a schematic diagram of a model training method according to an embodiment of the present invention.

[0021] Figure 3B Schematically shows the sorting result of the feature importance obtained by screening the trained first predetermined model.

[0022] Figure 3C Schematically shows the feature threshold for making the regression model perform optimally according to an embodiment of the present disclosure.

[0023] Figure 3D Schematically shows a schematic diagram of the fitting curve of regression model 2 according to an embodiment of the present disclosure.

[0024] Figure 4 Schematically shows a schematic diagram of a region recommendation method according to an embodiment of the present invention.

[0025] Figure 5 Schematically shows a schematic diagram of the modules of a processor according to an embodiment of the present invention. Detailed implementation manners

[0026] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. In the following detailed description, for the sake of explanation, many specific details are set forth in order to provide a comprehensive understanding of the embodiments of the present invention. However, it is obvious that one or more embodiments can be implemented without these specific details. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily obscuring the concepts of the present invention.

[0027] The terms used herein are merely for describing specific embodiments and are not intended to limit the present invention. The terms "including", "comprising" and the like used herein indicate the presence of the described features, steps, operations and / or components, but do not exclude the presence or addition of one or more other features, steps, operations or components.

[0028] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0029] In the case of using expressions such as "at least one of A, B, and C", generally, it should be interpreted according to the meaning that those skilled in the art usually understand this expression (for example, "a system having at least one of A, B, and C" should include, but not be limited to, a system having only A, only B, only C, having A and B, having A and C, having B and C, and / or having A, B, and C, etc.).

[0030] In the technical solution of the present invention, the user information involved (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) are all information and data that have been authorized by the user or fully authorized by all parties. Moreover, the processing of relevant data, such as collection, storage, use, processing, transmission, provision, disclosure, and application, all comply with relevant laws, regulations, and standards, adopt necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or reject.

[0031] In the scenario of making automated decisions using personal information, the methods, devices, and systems provided by the embodiments of the present invention all provide corresponding operation entrances for users to choose to agree or reject the results of automated decisions; if the user chooses to reject, then it enters the expert decision-making process. Here, the expression "automated decision" refers to the activity of automatically analyzing and evaluating a person's behavior habits, hobbies, or economic, health, credit status, etc. through a computer program and making a decision. Here, the expression "expert decision" refers to the activity of a person who specializes in a certain field, has specialized experience, knowledge, and skills and reaches a certain professional level to make a decision.

[0032] In the process of implementing the inventive concept of the present invention, the inventors found that the scale method for evaluating the motor function of patients has problems such as a long evaluation process, strong subjectivity of the results, and dependence on the observation ability of rehabilitation physicians. In addition, the evaluation accuracy of the scale is low, lacks sensitivity to subtle changes in the patient's motor performance, and it is difficult to accurately evaluate the patient's motor function.

[0033] Kinematics quantitative assessment technology has been widely used in neuroscience and clinical practice because it can explain the mechanism of motor function recovery. However, the optical motion capture system based on kinematics quantitative assessment technology has a high manufacturing cost, and a customized laboratory equipped with fixed cameras is required to evaluate the patient's motion; the robotic system with motion capture function based on kinematics quantitative assessment technology has a low degree of freedom of its mechanical structure, so it is difficult to capture images of the entire range of the patient's upper limb movement.

[0034] In view of this, an embodiment of the present invention provides a limb movement function evaluation system. By using wearable sensors, the movement function of a user's limb can be accurately evaluated, and it has the advantages of low cost, high degree of freedom, and unrestricted usage scenarios.

[0035] Figure 1A Schematically shows a schematic diagram of a limb movement function evaluation system according to an embodiment of the present invention. Figure 1B Schematically shows a schematic diagram of an object performing an action according to an embodiment of the present invention.

[0036] As Figure 1A and 1B shown, the limb movement function evaluation system includes a first wearable sensor 110, a display 120, and a processor 130. The first wearable sensor 110 can be a sensor such as an Inertial Measurement Unit (IMU). The first wearable sensor 110 can be worn on the limb of the object 140. The object 140 can refer to the user who uses this limb movement evaluation system. The limb of the object 140 can be the left arm or the right arm of the object 140, etc. Specifically, a single limb of the object 140 can include the upper arm, the forearm, and the hand. Correspondingly, the first wearable sensor 110 can be worn on the upper arm, the forearm, or the back of the hand of one side limb of the object 140. There can be multiple first wearable sensors 110, and multiple first wearable sensors 110 can be respectively worn at at least one position among the upper arm, the forearm, and the back of the hand of one side limb of the object 140. When the first wearable sensor 110 is worn on the limb of the object 140, the first wearable sensor 110 can collect the first position data of the limb of the object 140. The first position data characterizes the position of the limb of the object 140 in the space where the object 140 is located. For example, the first wearable sensor 110 can determine the first position data based on data such as the measured acceleration and angular velocity of the limb of the object 140. The first wearable sensor 110 is connected to the processor 130 and sends the collected first position data to the processor 130. For example, there can be a communication connection between the first wearable sensor 110 and the processor 130.

[0037] The processor 130 can be a device such as a terminal device. The processor 130 is also connected to the display 120. For example, there can be a communication connection or an electrical connection between the display 120 and the processor 130, and the present invention does not make a limitation in this regard. The display 120 can be integrated on the terminal device or independent of the terminal device, and the present invention does not make a limitation in this regard.

[0038] The processor 130 can control the display 120 to sequentially display the position indication images of K regions. In Figure 1BAmong them, a certain one of the K regions is shown by a dashed box. Correspondingly, the position points within this region are shown by circular icons. The position indication image can be a continuous video frame or an independent image, and the present invention does not limit this. The position indication image can include position points, and the present invention is not limited thereto. The position indication image can also include virtual objects. The virtual objects in the position indication images of the K regions can be located at the same position, while the position points in the position indication images of the K regions can be respectively located in the regions of K directions relative to the virtual object. In this way, the position indication image of each of the K regions indicates the position point of this region. For example, the virtual object can be generated based on the body shape information self-uploaded by the object 140, so that the object 140 can have a stronger sense of immersion when performing motion evaluation. It should be noted that before obtaining any information of the object 140 and processing this information, the permission of the object 140 has been obtained. The same applies to other parts of the present invention and will not be elaborated here.

[0039] The processor 130 controls the display 120 to display the position indication images of the K regions, can sequentially indicate the position points of the K regions to the object 140, and at the same time receive the first position data from the first wearable sensor 110, and use the received first position data as the first limb position data of the limb of the object 140. In this way, the first limb position data of the limb of the object 140 in the K regions can be obtained. Among them, the K regions are distributed in different orientations of the object 140 within the space where the object 140 is located, and K is an integer greater than 1. For example, the K regions can be respectively located in the upper right, lower right, upper left, lower left, etc. of the object 140, and are not limited thereto. The K regions can also include the regions in front of the object 140, and the present invention does not limit this. For example, the processor 130 can control the display 120 to display the position indication image of the "upper right" region. According to the position indication image of the "upper right" region displayed by the display 120, the object 140 can control the limb wearing the first wearable sensor 110 to extend towards the "upper right" in the space where the object 140 is located. During this process, the first position data of the limb of the object 140 in the "upper right" region can be collected as the first limb position data of the object 140 in the "upper right" region. In this way, the position indication images of each region are sequentially displayed to indicate that the object 140 extends the limb towards each corresponding direction in the space where the object 140 is located, so that the first limb position data of the K regions can be obtained.

[0040] The processor 130 may determine the limb movement information of K regions based on the first limb position data of the K regions. The limb movement information may characterize the movement of the limbs of the object 140 when driving the limbs. The limb movement information may include, but is not limited to, information such as the movement trajectory information, movement displacement information, movement speed information, average change rate of limb acceleration (Jerk), normalized jerk (NJ), and spectral arc length of speed (SAL). For example, for the first limb position data of each region, the limb movement information of that region may be calculated.

[0041] The processor 130 may generate a limb evaluation result of the object 140 based on the limb movement information of the K regions. The limb evaluation result characterizes the movement ability of the limbs of the object 140. For example, the processor 130 may evaluate the limb movement function of the object 140 in the K regions based on the limb movement information of the K regions, and thus generate a limb evaluation result of the object 140 by combining the limb movement states of the K regions.

[0042] According to an embodiment of the present invention, the processor 130 controls the display 120 to sequentially display the position points of different regions, and at the same time receives the first limb position data of different regions collected by the first wearable sensor 110. Since the first wearable sensor 110 is worn on the limb of the object 140, the first limb position data of the limb of the object 140 in each region can be freely and flexibly collected via the first wearable sensor 110, and thus the limb movement information of the limb of the object 140 in different regions can be determined according to the first limb position data of different regions. In this way, based on the limb movement information of the limb of the object 140 in different regions, the movement performance of the limb of the object 140 in each region can be accurately and objectively evaluated, improving the accuracy of the movement evaluation of the limb of the object 140. Moreover, since the present invention can accurately evaluate the movement of the limb of the object 140 only by using the processor 130, the display 120, and the wearable sensor, it has the advantages of low cost, high precision, unrestricted use scenarios, no need for professional physicians to participate, and can perform movement evaluation independently without supervision, thus meeting the needs of frequent evaluation during the rehabilitation process.

[0043] According to an embodiment of the present invention, the processor 130 may also obtain the limb information to be evaluated of the object 140. The limb information to be evaluated may be the information of the limb of the object 140 wearing the first wearable sensor 110. For example, the limb information to be evaluated may include information such as the upper left limb (such as the left arm), the upper right limb (such as the right arm), and so on.

[0044] After obtaining the limb information to be evaluated of the object 140, the processor 130 can also determine the position indication images of K regions from the position indication images of P kinds of limbs according to the limb information to be evaluated. Each of the P kinds of limbs is in a different direction of the same body, and P is a positive integer greater than 1. For example, the P kinds of limbs can refer to the limbs located at different positions of the body, such as the upper left limb and the upper right limb mentioned above.

[0045] For example, relative to the position point indicated by the position indication image of the upper left limb, the position point indicated by the position indication image of the upper right limb will be closer to the right. In this way, the object 140 wearing the first wearable sensor 110 can ensure that only the upper right limb moves towards the position point as much as possible while other parts of the body do not move, thereby improving the accuracy of evaluating the motion state of the limb wearing the first wearable sensor 110. Similarly, vice versa, which will not be elaborated here.

[0046] According to an embodiment of the present invention, the above system further includes a reference sensor (such as an IMU) placed statically in the space where the object 140 is located. For example, the reference sensor can be placed on the floor or the table around the object 140. The processor 130 is also connected to the reference sensor and calibrates the coordinate system of the first wearable sensor 110 based on the coordinate system of the reference sensor. The processor 130 and the reference sensor can be communicatively connected or electrically connected. For example, when it is detected that the object 140 has put on the first wearable sensor 110 on the limb, the processor 130 can control the display 120 to display a calibration indication image. The calibration indication image is used to instruct the object 140 to complete a predetermined action. For example, the predetermined action can include actions such as abduction and adduction of the shoulder joint, flexion and extension of the elbow joint, and flexion and extension of the wrist joint, etc. Correspondingly, the calibration indication image can also correspondingly include indication images of abduction and adduction of the shoulder joint, indication images of flexion and extension of the elbow joint, and indication images of flexion and extension of the wrist joint, etc. When the display 120 displays the calibration indication image, the object 140 can perform actions according to the indication of the calibration indication image. During this process, the processor 130 can use the coordinate system of the reference sensor as a reference, and based on the data collected by the reference sensor and the data collected by the first wearable sensor 110, determine the first position data of the first wearable sensor 110 according to the relative position between the first wearable sensor 110 and the reference sensor. In this way, the calibration process can be completed. After that, when the display 120 displays the position indication image, the processor 130 can collect the first position data through the first wearable sensor 110 according to the relative position between the reference sensor and the first wearable sensor 110, and use the collected first position data as the first limb position data. In this way, accurate first limb position data is obtained.

[0047] According to an embodiment of the present invention, the above system may further include an image collector, which may be, for example, a depth camera. The processor 130 is also connected to the image collector. For example, there may be an electrical connection or a communication connection between the processor 130 and the image collector. The processor 130 may control the image collector to collect a spatial image of the space where the object 140 is located. At the same time, the processor 130 may also control the display 120 to sequentially display the indication position images of K regions, and when displaying the position indication image of each region, match the position points in the position indication image with the limb position points of the limbs of the object 140 in the collected spatial image.

[0048] In the case where the position points in the position indication image do not match the limb position points in the spatial image, and the display duration of the position indication image is greater than or equal to a predetermined duration, the first position data received during the display period of the position indication image may be determined as the first limb position data, and the position indication image of the next region may be displayed; in the case where the position points in the position indication image match the limb position points in the spatial image, even if the display duration of the position indication image is less than the predetermined duration, the first position data received during the display period of the position indication image may also be determined as the first limb position data, and the image of the next region may be displayed. Repeat this process until the position indication images of all K regions are completely displayed, and the first limb position data of the K regions is obtained.

[0049] Specifically, after determining the first limb position data of the (k - 1)th region, control the display 120 to display the position indication image of the kth region, and at the same time control the image collector to collect a spatial image of the space where the object 140 is located, where k is a positive integer less than or equal to K. The position indication image of the kth region indicates the position points of the kth region. In the case where the processor 130 controls the image collector to collect the image in this space, since the object 140 is in this space, the image can show the actions of the object 140 in this space and the limb positions of the object 140 when performing actions (in this example, also when the image is collected). In this way, the limb position points of the object 140 can be identified in the collected spatial image, and thus the limb position points of the object 140 can be matched with the position points of the kth region.

[0050] In one embodiment, the position information of the limb position points of the object 140 in the spatial image can be determined according to the positions of the limb position points of the object 140 in the spatial image, and the position information of the position points in the position indication image can be determined according to the positions of the position points in the position indication image. In this way, the position information of the limb position points can be matched with the position information of the position points in the position indication image to obtain the matching result between the limb position points and the position points in the position indication image. Moreover, the present invention is not limited thereto. In another embodiment, after the limb position points are recognized, the first relative position relationship between the main body of the object 140 in the image (such as the torso position of the object 140) and the limb position can be determined. For the position indication image, there may be a second relative position relationship between the virtual object therein and the position points. The first relative position relationship can be matched with the second relative position relationship, and the matching result can be used as the matching result between the position points in the k-th region and the recognized limb position points. When the position points in the k-th region match the recognized limb position points, the processor 130 can determine the first position data collected during the display period of the position indication image in the k-th region as the first limb position data in the k-th region. In this way, by matching the limb position points in the spatial image of the object 140 collected in real time with the position points in the position indication image of the k-th region being displayed at this time, accurate limb position data that can characterize the movement of the limbs of the object 140 in the k-th region is obtained, improving the accuracy of the limb position data, and thus the accuracy of the limb evaluation result can be improved.

[0051] Figure 2 FIG. schematically shows a schematic diagram of the position points of K regions according to an embodiment of the present invention. As Figure 2 shown, I, II, III, IV, and V therein respectively represent different regions, and the position points of each region can be multiple. The position indication images of each region can be multiple, so as to indicate different position points in each position indication image. It should be noted that Figure 2 the numbers in are used to represent the order of the position points around the number. For example, it can refer to the order in which the position points are displayed. Figure 2 The number of regions and the number of position points in are only for illustration, and the present invention is not limited thereto. For example, the position indication image of the k-th region includes I position indication images, and each of the I position indication images indicates different position points within the k-th region, where I is an integer greater than 1. Correspondingly, the first limb position data of the k-th region includes I first limb position data. For example, the k-th region is in the k-th direction relative to the object 140 in space. The i-th position point indicated by the i-th position indication image among the I position indication images is farther from the object 140 in the k-th direction than the (i - 1)-th position point indicated by the (i - 1)-th position indication image, but it is not limited thereto.

[0052] For example, when the processor 130 needs to display the position indication images of the k-th region, it can sequentially display I position indication images and simultaneously control the image collector to collect the spatial image of the space where the object 140 is located. When respectively displaying each position indication image of the k-th region, the position points in the position indication image are matched with the limb position points of the limbs of the object 140 in the collected spatial image.

[0053] When the position points in the position indication image do not match the limb position points in the spatial image, and the display duration of the position indication image is greater than or equal to the predetermined duration, the first position data received during the display period of the position indication image can be determined as the first limb position data, and the next position indication image of the k-th region is displayed; when the position points in the position indication image match the limb position points in the spatial image, even if the display duration of the position indication image is less than the predetermined duration, the first position data received during the display period of the position indication image can also be determined as the first limb position data, and the next position indication image of the k-th region is displayed. Repeat this process until all the position indication images of the K regions are displayed, and the first limb position data of the K regions is obtained.

[0054] Specifically, after determining the (i - 1)-th limb position data, the processor 130 can control the display 120 to display the i-th position indication image and simultaneously control the image collector to collect the spatial image of the space. After collecting the spatial image, the processor 130 can identify the limb position points of the object 140 from the spatial image and match the limb position points of the object 140 with the position points in the i-th position indication image. The matching method can refer to the description above and will not be elaborated here. Thus, when the i-th position point in the i-th position indication image matches the identified limb position points, the first position data collected during the display period of the i-th position indication image is determined as the i-th first limb position data, where i is a positive integer less than or equal to I. Repeat this process to collect the I first limb position data of the I position indication images and use the I first limb position data as the first limb position data of the above-mentioned k-th region. On this basis, by matching the limb position points in the spatially immediate image of the object 140 with the position points in the i-th position indication image being displayed at this time, multiple limb position data that can accurately represent the movement of the limbs of the object 140 in the k-th region are obtained, improving the accuracy of the limb position data, and thus the accuracy of the limb assessment result can be improved.

[0055] Further, in the embodiments of the present invention, the processor 130 may identify the j-th limb position point in the spatial image collected at the j-th moment, and match the j-th limb position point with the i-th position point. In the case where the j-th limb position point matches the i-th position point, the duration between the start display moment of the i-th position indication image and the j-th moment is determined as the display duration of the i-th position indication image, or the display period of the i-th position indication image may be determined according to the start display moment of the i-th position indication image and the j-th moment. On this basis, the display duration of the position indication image in the k-th region may be determined based on the display durations of I position indication images, and the display period of the position indication image in the k-th region is similar, which will not be elaborated here. In this way, the display duration and display period of the position indication image can be accurately determined, thereby improving the accuracy of the first limb position data.

[0056] According to an embodiment of the present invention, the above system further includes a second wearable sensor (such as an IMU), which is worn on the torso of the object 140, for example, it can be worn on the chest. The processor 130 may also determine the k-th motion trajectory information and the k-th motion displacement information of the limb of the object 140 for the k-th region among the K regions according to the first limb position data of the k-th region and the second position data collected by the second wearable sensor. For example, taking the above-mentioned i-th first limb position data as an example, with the second position data instantaneously collected by the second wearable sensor as a reference, when the first wearable sensor 110 moves, based on the relative position relationship between the i-th first limb position data and the second position data, the coordinate information of the position points on the movement trajectory of the first wearable sensor 110 is determined, and the k-th motion trajectory information is determined based on this coordinate information. And, with the second position data as a reference, based on the relative position relationship between the first position data and the second position data, the sum of the three-dimensional distances between the position point farthest from the second position data and the position point closest to the second position data during the movement of the first wearable sensor 110 is determined to obtain the k-th motion displacement information. Specifically, since multiple first wearable sensors 110 can be worn on the forearm, the lower arm, and the back of the hand respectively, the motion trajectory information may include the motion trajectory information of the forearm, the motion trajectory information of the lower arm, and the motion trajectory information of the back of the hand based on the chest position as a reference, and the motion displacement information may include the motion displacement information of the forearm, the motion displacement information of the lower arm, and the motion displacement information of the back of the hand based on the chest position as a reference. However, the present invention is not limited thereto. In some embodiments, the motion trajectory information and motion displacement information of other parts may also be determined based on the forearm, the lower arm, or the arm as a reference, and the method is similar to the above process, which will not be elaborated here.

[0057] Thereafter, the processor 130 calculates the k-th motion speed information of the limb of the object 140 according to the k-th motion trajectory information and the position indicating the display duration of the image in the k-th region. For example, the motion speed per second of the limb can be determined based on the derivative of the motion trajectory information with respect to time. Assuming the display duration is M seconds, the maximum motion speed can be determined from M motion speeds, and the average speed of the limb during the display duration can be determined based on the average value of the M motion speeds. The maximum motion speed and the average speed can be determined as the motion speed information. M is a positive integer greater than 1.

[0058] The processor 130 can determine the limb motion information of the k-th region according to the k-th motion trajectory information, the k-th motion displacement information, the display duration, and the k-th motion speed information. For example, the second position data can also be used as the position reference, and based on the relative position relationship between the first limb position data and the second position data, the angular change range information of the limb wearing the first wearable sensor 110 during the display period of the i-th position indicating image can be determined. For example, the angular change range information can include the angular change range information of the forearm (corresponding to the change angle of the shoulder joint), the angular change range information of the lower arm (corresponding to the change angle of the elbow joint), and the angular change range of the back of the hand (corresponding to the change angle of the wrist joint).

[0059] Specifically, the limb motion information can include the average change rate of the acceleration of the limb (Jerk), the normalized jerk (Normalized Jerk, NJ), and the spectral arc length of the speed (Spectral Arc Length, SAL). The smaller the values of these three are, the smoother the limb motion is. The calculation formulas are as follows:

[0060] (1)

[0061] (2)

[0062] (3)

[0063] Among them can be set as:

[0064] (4)

[0065] where v max is the maximum joint speed; L is the length of the motion trajectory; t is the sampling duration; t tot is the task execution duration (corresponding to the display duration of each position indicating image); j is the third derivative of the joint motion trajectory; v(t) is the motion speed; v(ω) is the Fourier amplitude spectrum of v(t); ω is the target frequency value, that is, the frequency point that satisfies specific conditions of the frequency; is the normalized amplitude spectrum with a frequency equal to r; r is the frequency variable used to determine ω; is the normalized amplitude spectrum relative to v(0); v(0) is the value of the Fourier amplitude spectrum at the frequency ω = 0; is a predetermined threshold value, which can be adaptively selected based on a given threshold value to a predetermined maximum value as the upper bound.

[0066] In this way, based on the kinematic evaluation method and combining the above limb movement information, an objective and accurate evaluation of the upper limb movement function can be achieved, and more detailed clinical kinematic information can be provided.

[0067] According to an embodiment of the present invention, the processor 130 can determine the i-th standard movement value from the standard movement values of the k-th region according to the i-th position indication image in the I position indication images. For example, the standard movement values of the k-th region may include I standard movement values. The i-th standard movement value among the I standard movement values characterizes the movement value of a limb in a normal state when performing an action according to the position point in the i-th position indication image. The standard movement value may include standard movement trajectory information, standard movement displacement information, standard movement speed information, standard average change rate of acceleration, standard normalized jerk, and spectral arc length of standard speed, etc., of a limb in a normal state.

[0068] The processor 130 may also determine the i-th motion difference value according to the i-th limb motion information and the i-th standard motion value among the I limb motion information. The motion difference value may be a motion difference index (MNI). For example, the i-th motion difference value may be determined based on the difference between the i-th normalized jerk and the i-th standard normalized jerk. And the present invention is not limited thereto. In another embodiment of the present invention, the motion difference value of the i-th normalized jerk may be obtained based on the ratio between the i-th normalized jerk and the i-th standard normalized jerk, thereby realizing the standardization of the normalized jerk. Similarly, the motion difference value of the average change rate of the i-th acceleration, the motion difference value of the spectral arc length of the i-th velocity, the motion difference value of the i-th motion trajectory information, the motion difference value of the i-th motion displacement information, and the motion difference value of the i-th motion velocity information may be determined. Then, for the motion difference value of the i-th normalized jerk, the motion difference value of the average change rate of the i-th acceleration, the motion difference value of the spectral arc length of the i-th velocity, the motion difference value of the i-th motion trajectory information, the motion difference value of the i-th motion displacement information, and the motion difference value of the i-th motion velocity information, a dimensionality reduction calculation is performed using the principal components analysis (PCA) algorithm to obtain the i-th motion difference value. Thereafter, the processor 130 may also determine the k-th regional motion difference value according to the statistical value of the I motion difference values. For example, the k-th regional motion difference value may be determined based on the sum average value of the I motion difference values.

[0069] The processor 130 may also generate a limb evaluation result according to the K regional motion difference values. For example, the K regional motion difference values may be compared with a predetermined difference value to generate a limb evaluation result. For example, in the case where the k-th regional motion difference value is less than the predetermined difference value, a limb evaluation result indicating that the limb motion is normal in this region may be generated. Otherwise, a limb evaluation result indicating that the limb motion is abnormal may be generated. Thus, according to the K regional motion difference values that can represent the difference between the motion state of the object 140's limb in the K regions and the normal state, the motion conditions of the object 140's limb in each region and the limb evaluation result can be accurately and comprehensively determined.

[0070] According to an embodiment of the present invention, the processor 130 may also perform a motion function evaluation on the limb motion information of the K regions to obtain a limb motion evaluation value (for example, it may be the predicted upper limb evaluation (Fugl-Meyer Assessment Upper Extremity Scale, FMA-UE) score), and generate a limb evaluation result according to the limb motion evaluation value and the K regional motion difference values.

[0071] For example, the trained first predetermined model can be used to process the limb movement information of K regions to obtain the weight information of each dimension of the limb movement information of the K regions. Among them, the information of each dimension includes the above-mentioned movement trajectory information, movement displacement information, movement speed information, average change rate of acceleration, normalized jerk, and spectral arc length of speed. Then, the trained second predetermined model is used to process the above-mentioned information of each dimension and the weight information to obtain a limb movement evaluation value. The trained first predetermined model is obtained by training an initial model based on sample limb movement information and label evaluation values. For example, the input of the first predetermined model can be the limb movement information collected by the user during the movement (the initial sample size can be about 300), and the prediction label can be set as the user's label FMA-UE score (i.e., the label evaluation value), and the model finally outputs the weight information of each dimension of the limb movement information. Specifically, after different limb movement information is input into the first predetermined model, the first predetermined model can perform fitting analysis based on the limb movement information and the label FMA-UE score, and use the contribution degree of each feature in the limb movement information when predicting the score as the weight information of the feature. The initial model of the first predetermined model is constructed based on the Extreme Gradient Boosting (XGBoost). The trained second predetermined model can be obtained by training an initial regression model based on sample limb movement information, sample weight information corresponding to the sample limb movement information, and label evaluation values. During the training process, the input of the initial regression model is the limb movement information (X value) sorted based on the sample weight information and the label evaluation value (Y value), and the output of the second predetermined model is the predicted label evaluation value (Y' value). Among them, the label evaluation value is used as the training label for training the second predetermined model, and the label evaluation value can be preset and is not limited herein. The initial regression model can be constructed based on machine learning algorithms, such as K-Nearest Neighbor Regression, Random Forest Regression, Support Vector Regression, Ridge Regression and other models. The following will be combined with Figure 3A to elaborate on the process of training the second predetermined model.

[0072] Figure 3A Schematically shows a schematic diagram of the model training method according to an embodiment of the present invention.

[0073] As Figure 3A shown, the model training method of this embodiment may include operations S310 to S340.

[0074] In operation S310, the trained first predetermined model is used to process the sample limb movement information to obtain sample weight information.

[0075] In operation S320, the embedding method is used to screen out the feature information of the dimension that makes the initial regression model perform optimally from the sample limb movement information. The feature information of the dimension that makes the initial regression model perform optimally described herein may include at least one of motion trajectory information, motion displacement information, motion speed information, average change rate of acceleration, normalized jerk, and spectral arc length of speed.

[0076] In operation S330, the initial regression model is trained using the screened feature information and label evaluation values. During the training process, 80% of the samples are randomly selected as the training set, and the remaining 20% of the samples are used as the validation set. The initial regression model is trained using the sample limb movement information of the training set, and the accuracy of the initial regression model is verified using the sample limb movement information of the validation set until the model training is completed.

[0077] In operation S340, the performance of the initial regression model is verified. For example, the five-fold cross-validation method can be used to compare the limb movement evaluation values output by the model for the validation set with the label evaluation values corresponding to the sample limb movement information, so as to verify the performance of the initial regression model. The performance of the initial regression model can be optimized by adjusting the model parameters of the initial regression model or selecting different machine learning algorithms, so as to obtain the trained predetermined model. For example, the label evaluation value can be the standard motion evaluation value corresponding to the sample limb movement information. For example, the standard motion evaluation value can be the FMA-UE score in the normal state. To improve the model performance, new accumulated patient data will be collected later to regularly train and update the model parameters. After obtaining the i-th limb movement information, the processor 130 can also calculate the i-th motion difference value according to the i-th limb movement information and the i-th standard motion value. The calculation method is the same as above and will not be elaborated here. In this way, accurate limb movement values can be obtained by processing the limb movement information using the trained predetermined model, and thus accurate motion difference values can be obtained.

[0078] Figure 3B Schematically shows the sorting result of the feature importance screened by the first predetermined model (i.e., the above-mentioned XGBoost) trained in operation S310. In Figure 3B schematically shows the importance of features 1 to 32 determined by XGBoost. Among them, features 1 to 32 may include 32 features such as motion trajectory information, motion displacement information, motion speed information, average change rate of acceleration, normalized jerk, and spectral arc length of speed. Referring to Figure 3B it can be seen that the importance of features 1, 2, and 3 is relatively high. Thus, the above-screened features may include features 1, 2, and 3. Among them, the number of nodes of the trained XGBoost can be 100 - 200, and the maximum depth can be 4 - 6.

[0079] Figure 3C Schematically shows the feature threshold that makes the regression model perform best determined by the embedding method described in operation S320 according to an embodiment of the present disclosure. In Figure 3C Schematically shows the thresholds that can make regression models 1 to 4 perform optimally. R 2 Represents the coefficient of determination, which is used to measure the ability of the model to explain the data.

[0080] Figure 3D Schematically shows the fitting schematic diagram of regression model 2 according to an embodiment of the present disclosure. Among them, R 2 Represents the coefficient of determination, which is used to measure the ability of the model to explain the data. PCC represents the Pearson correlation coefficient, which is used to measure the linear correlation degree between two variables. Refer to Figure 3D It can be seen that the R 2 between the predicted evaluation value and the true evaluation value of regression model 2 can reach 0.84, and the PCC can reach 0.92, indicating a good fitting degree. Further, after the five-fold cross-validation of operation S340, the average R 2 of regression model 2 is 0.83, the average root mean squared error (RMSE) is 7.83, and the average mean absolute error (MAE) is 4.6, indicating that the model performance is relatively stable. Thus, regression model 2 can be used as the second predetermined model described above. Among them, regression model 2 can be a random forest regression model, and the number of decision trees in this regression model can be set to 250 to 350.

[0081] According to an embodiment of the present invention, the processor 130 may determine a motion recommendation area from the K areas according to the limb motion evaluation value and the K area motion difference values output by the second predetermined model, and may control the display 120 to indicate the motion recommendation area to the object 140. For example, when the limb motion evaluation value is greater than or equal to a predetermined standard motion evaluation value (e.g., FMA-UE = 32 points), it may be determined that the limb motion ability of the object 140 is relatively strong, and the area exceeding the predetermined range among the K area motion difference values may be recommended as the motion recommendation area, so as to recommend that the moving object 140 control the limb to move to the position point of the motion recommendation area for exercise. When the limb motion evaluation value is less than the predetermined standard motion evaluation value (e.g., FMA-UE = 32 points), it may be determined that the limb motion ability of the object 140 is relatively weak. In this case, all K areas need to be determined as the motion recommendation areas, and the object 140 is recommended to perform simple stretching training in the whole area to improve the level of its upper limb motor function. In this way, a personalized rehabilitation plan can be formulated for the patient according to the patient's motion performance in combination with the K area motion difference values, providing guidance for further rehabilitation plans and improving the user experience.

[0082] Figure 4 FIG. schematically shows a schematic diagram of a region recommendation method according to an embodiment of the present invention.

[0083] As Figure 4 shown, the region recommendation method of this embodiment includes operations S410 to S450.

[0084] In operation S410, the principal component analysis algorithm is used to reduce the dimension of the limb motion information, and the motion difference value is calculated.

[0085] In operation S420, the sum average value of the motion difference values of each region is determined as the region motion difference value of this region.

[0086] In operation S430, it is determined whether the limb motion evaluation value is greater than or equal to a predetermined standard motion evaluation value. If yes, operation S440 is executed; if no, operation S450 is executed.

[0087] In operation S440, some of the K regions are recommended. The partial regions here are the regions exceeding the predetermined range among the above K region motion difference values.

[0088] In operation S450, all K regions are recommended.

[0089] In one embodiment, the processor and the display of the present invention may be integrated into a VR device. The motion evaluation process may be implemented based on a virtual reality (VR) game to further improve the user experience. The following combination Figure 5Specific descriptions will be given hereinafter. Figure 5 FIG. schematically shows a schematic diagram of modules of a processor according to an embodiment of the present invention.

[0090] As Figure 5 shown, the processor 500 of this embodiment may include a user information management module 510, an inertial sensing calibration module 520, a VR evaluation game module 530, a motion data analysis module 540, a motion function evaluation module 550, and a rehabilitation plan design module 560. For example, the user information management module 510 is used to create and store basic user information, including name, gender, age, disease course, stroke type, affected side, and upper limb joint parameters; the inertial sensing calibration module 520 is used to perform the above calibration on the first wearable sensor 110; the VR evaluation game module 530 is used to visualize and gamify the evaluation process, enabling the user to complete the evaluation in a game through human-computer interaction; the motion data analysis module 540 is used to analyze and process the game task performance of the user during the evaluation process and the motion data collected in real time by the nine-axis IMU; the motion function evaluation module 550 is used to evaluate the upper limb motion function of the user and predict its FMA-UE score; the rehabilitation plan design module 560 is used to recommend the areas that need intensive training for the user according to the motion performance of the user's limb and the FMA-UE score output by the motion function evaluation module 550.

[0091] Specifically, the user information management module 510 receives the basic information input by the user, including name (first name), gender (select "male" or "female"), age (number), disease course (number, unit: month), stroke type (select "hemorrhagic stroke" or "ischemic stroke"), affected side (select "left side" or "right side"), and three-dimensional anatomical distances of the upper limb (3×3 matrix). Then, the user wears 4 IMUs on the chest, the affected upper arm, the forearm, and the back of the hand respectively, and places 1 IMU (i.e., the reference sensor) on the desktop for spatial positioning. Next, the user sequentially completes three calibration actions of "shoulder abduction / adduction", "elbow flexion / extension", and "wrist flexion / extension" according to the requirements in the inertial sensing calibration module 520, thereby completing the calibration of the 4 IMU sensors worn by the user.

[0092] After completing the calibration of the IMU with the user's body segments, the user enters the VR evaluation game module 530. Meanwhile, the nine-axis IMU collects the user's position data in real time during the game, and the VR evaluation game module 530 records the user's game performance. The game design in the VR evaluation game module 530 mainly targets the patient's arm-lifting and control abilities, and simplifies and transforms the FMA-UE scale to obtain a task paradigm. The task paradigm consists of 30 reach-to-target tasks (i.e., the above-mentioned position points), and these task points are arranged in five different regions in the upper left, lower left, upper right, lower right of the vertical plane and the horizontal plane according to the distance of the upper limb reaching. During the evaluation process, the subject 140 needs to lift the arm and try to reach out to touch these task points. A time limit is set for each task point. If the task point is successfully touched within the specified time, it is regarded as completing a task; otherwise, it is regarded as not completed. When a task is successfully completed or timed out, the system will automatically jump to the next task point until all task points are evaluated.

[0093] After the evaluation game ends, the motion data analysis module 540 will analyze and process the position data collected by the IMU and the task performance of the VR evaluation game. The position data analysis mainly includes calculating the upper limb joint angles and positions using the nine-axis data of the IMU, and further extracting multi-dimensional limb movement information. The multi-dimensional limb movement information includes joint movement trajectories, smoothness, speed, displacement, and range of motion. The task performance analysis mainly includes operations such as sorting out the user's task point completion status and completion time. Thereafter, the motion function evaluation module 550 will calculate the feature weights using the XGBoost algorithm according to the multi-dimensional limb movement information extracted by the data analysis module, perform feature screening using the embedding method, and process the screened features using a pre-trained predetermined model to output the user's FMA-UE score, that is, the limb movement evaluation value. Among them, the character models in the VR evaluation game have two types: male characters and female characters, which can be adjusted according to the user's gender in the user information management module 510. Similarly, the "affected side" information can adjust the game settings. If the patient has left-sided damage, the evaluation game targets the patient's left upper limb, and vice versa for the right side. In this way, it can make the user have a stronger sense of immersion and improve the user experience.

[0094] The rehabilitation program design module 560 takes the movement data of healthy people as the baseline, calculates the movement difference value between the patient and the healthy people's movement data in each region, and combines the FMA-UE score output by the motion function evaluation module 550 to recommend the regions that need intensive training for the user.

[0095] In the present invention, the IMU realizes the real-time collection of human movement data, and on this basis, combines virtual reality technology to realize the visualization and gamification of the evaluation process. Through kinematic analysis and processing of the collected data, an objective and accurate evaluation of the upper limb movement function is completed, and a personalized rehabilitation program can be formulated according to the patient's movement performance.

[0096] According to an embodiment of the present invention, any multiple of the user information management module 510, the inertial sensing calibration module 520, the VR evaluation game module 530, the motion data analysis module 540, the motion function evaluation module 550, and the rehabilitation plan design module 560 may be combined and implemented in one module, or any one of them may be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules may be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present invention, at least one of the user information management module 510, the inertial sensing calibration module 520, the VR evaluation game module 530, the motion data analysis module 540, the motion function evaluation module 550, and the rehabilitation plan design module 560 may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on chip, a system on substrate, a system on package, an application specific integrated circuit (ASIC), or may be implemented by any other reasonable means such as integrating or packaging circuits, etc., in hardware or firmware, or implemented in any one of the three implementation manners of software, hardware, and firmware, or in an appropriate combination of any several of them. Alternatively, at least one of the user information management module 510, the inertial sensing calibration module 520, the VR evaluation game module 530, the motion data analysis module 540, the motion function evaluation module 550, and the rehabilitation plan design module 560 may be at least partially implemented as a computer program module, and when the computer program module is run, it can execute the corresponding functions.

[0097] According to an embodiment of the present invention, the program code for executing the computer program provided by the embodiment of the present invention may be written in any combination of one or more programming languages. Specifically, these computing programs may be implemented using high-level procedures and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages include, but are not limited to, programming languages such as Java, C++, Python, the "C" language, or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (for example, by using an Internet service provider to connect through the Internet).

[0098] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code, and the above-mentioned module, segment of a program, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, as well as combinations of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0099] Those skilled in the art can understand that the features described in the various embodiments of the present invention can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features described in the various embodiments of the present invention can be combined and / or combined in various ways. All such combinations and / or combinations fall within the scope of the present invention.

[0100] The above describes the embodiments of the present invention. However, these embodiments are only for illustrative purposes and not for limiting the scope of the present invention. Although the embodiments are described separately above, this does not mean that the measures in each embodiment cannot be used advantageously in combination. Without departing from the scope of the present invention, those skilled in the art can make various substitutions and modifications, and all such substitutions and modifications should fall within the scope of the present invention.

Claims

1. A limb movement function evaluation system, characterized in that, Comprising: A display; A first wearable sensor worn on the limb of an object; And A processor, connected to the display and the first wearable sensor, configured to: control the display to sequentially display position indication images of K regions to sequentially indicate position points in the K regions to the object, and simultaneously receive first position data from the first wearable sensor to obtain first limb position data of the K regions, the K regions being distributed in different orientations of the object in the space where the object is located, K being an integer greater than 1; determine limb movement information of the K regions according to the first limb position data of the K regions; generate a limb evaluation result of the object according to the limb movement information of the K regions; Wherein, the limb movement function evaluation system further includes an image collector; the processor is further connected to the image collector and is further configured to: after determining the first limb position data of the (k - 1)th region, control the display to display the position indication image of the kth region, and simultaneously control the image collector to collect a spatial image of the space, the position indication image of the kth region indicating the position point of the kth region; identify the limb position point of the object in the collected spatial image; and in the case where the position point of the kth region matches the identified limb position point, determine the first position data collected during the display period of the position indication image of the kth region as the first limb position data of the kth region, k being a positive integer less than or equal to K; The first limb position data of the kth region includes I first limb position data; the position indication image of the kth region includes I position indication images, each of the I position indication images indicating a different position point in the kth region, I being an integer greater than 1; the processor is further configured to: after determining the (i - 1)th limb position data, control the display to display the ith position indication image, and simultaneously control the image collector to collect a spatial image of the space; identify the limb position point of the object from the collected spatial image; and in the case where the ith position point of the ith position indication image matches the identified limb position point, determine the first position data collected during the display period of the ith position indication image as the ith first limb position data, i being a positive integer less than or equal to I; The processor is further configured to: identify the jth limb position point in the spatial image collected at the jth moment; match the jth limb position point with the ith position point, so as to, in the case where the jth limb position point matches the ith position point, obtain the display duration of the ith position indication image according to the start display moment of the ith position indication image and the jth moment; Among them, the limb movement function evaluation system further includes a second wearable sensor worn on the torso of the object; the processor is further configured to: for the k-th region among the K regions, determine the k-th movement trajectory information and the k-th movement displacement information of the object's limb according to the first limb position data of the k-th region and the second position data collected by the second wearable sensor; calculate the k-th movement speed information of the object's limb according to the k-th movement trajectory information and the display duration of the position indication image of the k-th region; and determine the limb movement information of the k-th region according to the k-th movement trajectory information, the k-th movement displacement information, the display duration, and the k-th movement speed information, where k is a positive integer less than or equal to K.

2. The system according to claim 1, characterized in that, The position indication image of the k-th region includes I position indication images, and the limb movement information of the k-th region includes I limb movement information, where I is an integer greater than 1; The processor is further configured to: Determine the i-th standard movement value from the standard movement values of the k-th region according to the i-th position indication image among the I position indication images, where i is a positive integer less than or equal to I; Determine the i-th movement difference value according to the i-th limb movement information among the I limb movement information and the i-th standard movement value; Determine the k-th region movement difference value according to the statistical value of the I movement difference values; And Generate the limb evaluation result according to the K region movement difference values.

3. The system according to claim 2, wherein The processor is further configured to: Perform a movement function evaluation on the limb movement information of the K regions to obtain a limb movement evaluation value; and Generate the limb evaluation result according to the limb movement evaluation value and the K region movement difference values.

4. The system according to claim 3, wherein The processor is further configured to: Determine a movement recommendation region from the K regions according to the limb movement evaluation value and the K region movement difference values; And Control the display to indicate the movement recommendation region to the object.

5. The system according to claim 1, wherein The processor is further configured to: Obtain the limb information to be evaluated of the object; and Determine the position indication images of the K regions from the position indication images of P limbs according to the limb information to be evaluated, where each of the P limbs is in a different direction of the same body, and P is a positive integer greater than 1.

6. The system according to claim 1, wherein It further includes a reference sensor placed statically in the space where the object is located; The processor is further connected to the reference sensor, and the processor is further configured to: Calibrate the coordinate system of the first wearable sensor based on the coordinate system of the reference sensor, and collect the first position data through the first wearable sensor according to the relative position between the reference sensor and the first wearable sensor.

Images