A closed-loop monitoring and intelligent early warning method and system based on a posture sensor

Abstract

The embodiment of the application relates to the technical field of intelligent monitoring, and discloses a closed-loop monitoring and intelligent early warning method and system based on posture sensors, which comprises the following steps: first, acquiring real-time monitoring data of first and second posture sensors to form first and second initial data sets; then, correcting the initial data sets according to the deformation characteristics of the sensor installation parts to obtain first and second target data sets; then, determining the real-time motion angle of the first joint body relative to the second joint body based on the target data sets; then, constructing a time sequence simulation joint motion picture according to the real-time motion angle; and finally, generating an early warning signal according to the time sequence simulation joint motion picture or the real-time motion angle of the joint to remind the user to adjust the motion mode or stop the motion. In this way, the technical scheme of the application can reduce the measurement error caused by the deformation of the soft tissue around the joint, improve the joint motion monitoring accuracy, effectively prevent the motion injury, and improve the safety and rationality of the motion.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent monitoring, in particular to a closed-loop monitoring and intelligent early warning method and system based on a posture sensor. BACKGROUND

[0002] In the field of human motion health, the monitoring and evaluation of joint motion are of great significance for preventing motion injuries, improving motion effects, and rehabilitation medicine, etc. With the continuous improvement of people's health consciousness, the number of people participating in various sports activities is increasing, but the incidence of motion injuries is also rising. For example, in common knee joint motion, due to improper motion posture, excessive motion intensity, or potential problems of the joint itself, etc., it is easy to cause serious consequences such as knee sprain and ligament injury, which not only affects the individual's daily life and motion ability, but also may cause long-term health hazards.

[0003] Currently, the monitoring of joint motion mainly relies on traditional observation methods and simple measurement tools. The observation method is often affected by the subjective factors of the observer, and it is difficult to accurately and comprehensively capture the subtle changes and complex characteristics of joint motion. Simple measurement tools, such as angle meters, can only provide single-dimensional motion data and cannot reflect the multi-dimensional motion state of the joint in the entire motion process in real time and dynamically. In addition, the existing monitoring methods mostly lack real-time feedback and early warning mechanisms, and when the joint motion is abnormal, the user cannot be reminded to adjust the motion mode or stop the motion in time, thereby increasing the risk of motion injury. SUMMARY

[0004] The main purpose of the present application is to provide a closed-loop monitoring and intelligent early warning method and system based on a posture sensor, which aims to solve the technical problems of inaccurate joint motion data monitoring and lack of real-time feedback mechanism in the prior art.

[0005] To achieve the above purpose, in a first aspect, the present application provides a closed-loop monitoring and intelligent early warning method based on a posture sensor, which is applied to the monitoring and early warning of joint motion, and the method comprises:

[0006] Respectively acquiring real-time monitoring data of a first posture sensor and a second posture sensor to obtain a first initial data set and a second initial data set, wherein the first initial data set includes at least one of pitch motion data, yaw motion data and roll motion data of a first joint body, and the second initial data set includes at least one of pitch motion data, yaw motion data and roll motion data of a second joint body;

[0007] According to the deformation characteristics of the sensor installation part, data correction is performed on the first initial data set and the second initial data set to obtain a first target data set and a second target data set;

[0008] determining a real-time motion angle of the first joint body relative to the second joint body according to the first target data set and the second target data set;

[0009] constructing a time-series simulation joint motion map according to the real-time motion angle of the first joint body relative to the second joint body;

[0010] generating a warning signal according to the time-series simulation joint motion map or the real-time motion angle of the joint, the warning signal being used to remind a user to adjust a motion mode or stop the motion.

[0011] In a possible implementation, the motion data includes one or more of acceleration, displacement, angle or motion trajectory.

[0012] In a possible implementation, the joint is a knee joint, the first initial data set includes a pitch motion angle and a yaw motion angle of a thigh, and the second initial data set includes a pitch motion angle and a yaw motion angle of a shank; the data correction of the first initial data set and the second initial data set according to the deformation characteristics of the installation positions is specifically:

[0013] correcting the pitch motion angle and the yaw motion angle of the thigh according to a first deformation coefficient of a first attitude sensor at a thigh installation position to obtain the first target data set, the first deformation coefficient representing a deformation capability of soft tissue at the thigh installation position;

[0014] correcting the pitch motion angle and the yaw motion angle of the shank according to a second deformation coefficient of a second attitude sensor at a shank installation position to obtain the second target data set, the second deformation coefficient representing a deformation capability of soft tissue at the shank installation position.

[0015] In a possible implementation, the correction of the pitch motion angle and the yaw motion angle of the thigh according to the first deformation coefficient of the first attitude sensor at the thigh installation position to obtain the first target data set includes:

[0016] constructing a thigh soft tissue deformation correction model according to the first deformation coefficient, the correction model taking the first deformation coefficient as a key parameter and being used to describe a deformation law of soft tissue at the thigh installation position under different motion angles based on human soft tissue mechanics characteristics;

[0017] inputting initial measurement data into the thigh soft tissue deformation correction model, and correcting and calculating the pitch motion angle measurement value and the yaw motion angle measurement value in the initial measurement data by using the correction model to obtain corrected pitch motion angle and yaw motion angle;

[0018] The corrected pitch movement angle and the yaw movement angle are integrated to form a first target data set containing the corrected movement angle information.

[0019] In a possible implementation, determining the real-time movement angle of the first joint body relative to the second joint body according to the first target data set and the second target data set comprises:

[0020] The real-time pitch movement angle of the first joint body relative to the second joint body is obtained according to the real-time pitch movement angle of the first joint body, the real-time pitch movement angle of the second joint body, and a pitch transformation relationship, the pitch transformation relationship being a pitch movement coordinate transformation between the first joint body coordinates and the second joint body coordinates.

[0021] The real-time yaw movement angle of the first joint body relative to the second joint body is obtained according to the real-time yaw movement angle of the first joint body, the real-time yaw movement angle of the second joint body, and a yaw transformation relationship, the yaw transformation relationship being a yaw movement coordinate transformation between the first joint body coordinates and the second joint body coordinates.

[0022] In a possible implementation, constructing the time sequence simulation joint motion map according to the real-time movement angle of the first joint body relative to the second joint body comprises:

[0023] The real-time movement angle data of the first joint body relative to the second joint body is obtained in time sequence.

[0024] The real-time movement angle data at each time point is mapped to a corresponding position of a joint model according to a preset joint model, the joint model being a three-dimensional model simulating a connection relationship between the first joint body and the second joint body.

[0025] The joint model states at each time point after mapping are combined in time sequence to form a time sequence simulation joint motion map capable of dynamically displaying a movement process of the first joint body relative to the second joint body.

[0026] In a possible implementation, the early warning signal comprises one or more of a voice signal, a vibration signal, or an APP push signal.

[0027] In a possible implementation, the early warning signal is generated according to the time sequence simulation joint motion map or the real-time movement angle of the joint, comprising:

[0028] The time sequence simulation joint motion map is input into a risk identification model to obtain a risk coefficient representation value.

[0029] Different levels of early warning signals are generated according to the risk coefficient representation value.

[0030] When the risk coefficient representation value is greater than or equal to the first threshold value, the user is reminded to stop the movement in a voice mode; when the risk coefficient representation value is greater than or equal to the second threshold value and less than the first threshold value, the user is reminded to adjust the movement mode in a vibration mode.

[0031] In a possible implementation, the inputting the time sequence simulation joint motion graph into a risk identification model to obtain a risk coefficient representation value comprises:

[0032] performing feature extraction on the time sequence simulation joint motion graph to obtain a joint motion feature vector, the joint motion feature vector comprising at least two of a joint motion angle feature, a joint motion speed feature, a joint motion acceleration feature, and a joint motion trajectory feature;

[0033] inputting the joint motion feature vector into a pre-trained risk identification model, the risk identification model being constructed based on a deep learning algorithm;

[0034] the risk identification model performing calculation according to the input joint motion feature vector to output a risk coefficient representation value, the risk coefficient representation value reflecting a possibility of a risk existing in a current joint motion state.

[0035] In a second aspect, the embodiments of the present application further provide a closed-loop monitoring and intelligent early warning system, comprising:

[0036] a first attitude sensor, the first attitude sensor being arranged on a first joint body;

[0037] a second attitude sensor, the second attitude sensor being arranged on a second joint body, the second joint body being rotationally connected with the first joint body to form a joint; and

[0038] a memory and a processor, the memory being used to store program code, and the processor being used to call the program code to execute the method according to the first aspect.

[0039] Different from the prior art, the closed-loop monitoring and intelligent early warning method based on the attitude sensor provided in the embodiments of the application first acquires real-time monitoring data of the first attitude sensor and the second attitude sensor to obtain a first initial data set and a second initial data set; then performs data correction on the first initial data set and the second initial data set according to deformation characteristics of the sensor installation parts to obtain a first target data set and a second target data set; then determines a real-time motion angle of the first joint body relative to the second joint body according to the first target data set and the second target data set; then constructs a time sequence simulation joint motion picture according to the real-time motion angle of the first joint body relative to the second joint body; and finally generates an early warning signal according to the time sequence simulation joint motion picture or the real-time motion angle of the joint to remind the user to adjust the motion mode or stop the motion. In this way, the technical solution of the application can effectively reduce the measurement error caused by the deformation of the soft tissue around the joint, improve the accuracy of joint motion monitoring, and can present the joint motion process in real time, dynamically and intuitively, timely perceive the abnormal motion state and issue an early warning, effectively prevent motion injury, and improve the safety and rationality of the motion. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only show some embodiments of the application, and for those skilled in the art, other drawings can also be obtained from the structures shown in the drawings without creative labor.

[0041] Figure 1 A schematic diagram of the arrangement of the attitude sensor in the joint in some embodiments of the application;

[0042] Figure 2 A flowchart of the closed-loop monitoring and intelligent early warning method based on the attitude sensor in some embodiments of the application;

[0043] Figure 3 A flowchart of step S200 of the closed-loop monitoring and intelligent early warning method based on the attitude sensor in some embodiments of the application;

[0044] Figure 4 A flowchart of step S300 of the closed-loop monitoring and intelligent early warning method based on the attitude sensor in some embodiments of the application;

[0045] Figure 5 A hardware structure schematic diagram of the closed-loop monitoring and intelligent early warning system in some embodiments of the application.

[0046] The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0048] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present application are only used to explain the relative positional relationship, movement condition, etc. between components in a certain posture (as shown in the drawings), and if the certain posture changes, the directional indications also change accordingly.

[0049] In addition, the description of "first", "second" and the like in the present application is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first", "second" can explicitly or implicitly include at least one of the features. In addition, "and / or" throughout the text includes three solutions, for example, A and / or B includes A technical solution, B technical solution, and A and B simultaneously meet the technical solution; in addition, the technical solutions of each embodiment can be combined with each other, but it must be based on the realization of ordinary skill in the art, when the combination of technical solutions appears contradictory or unachievable, it should be considered that the combination of technical solutions does not exist, nor within the protection scope of the present application.

[0050] In the field of human motion health, the monitoring and evaluation of joint motion are of great significance for preventing sports injuries, improving sports effects, and rehabilitation medicine, etc. With the continuous improvement of people's health consciousness, the number of people participating in various sports activities is increasing, but the incidence of sports injuries is also rising. For example, in common knee joint motion, due to improper exercise posture, excessive exercise intensity or potential problems in the joint itself, etc., it is easy to cause serious consequences such as knee sprain and ligament injury, which not only affects the individual's daily life and exercise ability, but also may cause long-term health hazards.

[0051] Currently, the monitoring of joint movement mainly relies on traditional observation methods and simple measurement tools. Observation methods are often affected by the subjective factors of observers, making it difficult to accurately and comprehensively capture the subtle changes and complex characteristics of joint movement. Simple measurement tools, such as angle gauges, can only provide single-dimensional movement data and cannot dynamically reflect the multi-dimensional movement state of the joint during the entire movement process in real time. In addition, existing monitoring methods mostly lack real-time feedback and early warning mechanisms, and when joint movement is abnormal, they cannot timely remind users to adjust the movement mode or stop movement, thereby increasing the risk of movement injury.

[0052] The following describes the closed-loop monitoring and intelligent early warning system for implementing the closed-loop monitoring and intelligent early warning method based on posture sensors. As shown in Figure 1 In the embodiments of the present application, the closed-loop monitoring and intelligent early warning system at least includes a first posture sensor 100 and a second posture sensor 200, wherein the first posture sensor 100 is arranged on a first joint body L100; the second posture sensor 200 is arranged on a second joint body L200, and the second joint body L200 is rotationally connected with the first joint body L100 to form a joint, which can be a knee joint, a shoulder joint, a hip joint, an elbow joint, an ankle joint, etc.

[0053] Optionally, the first posture sensor 100 and the second posture sensor 200 in the present application can be inertial measurement units (IMU, Inertial Measurement Unit), which are devices for measuring the three-axis attitude angle (or angular velocity) and acceleration of an object. Displacement detection can also be realized by integrating displacement detection function on the basis of detecting angle, angular velocity and acceleration.

[0054] It should be noted that although the logical sequence is shown in the flowchart, in some cases, the steps shown or described can be performed in an order different from that shown here. Please refer to the attached Figures 1-4 The method includes the following steps S100-S500:

[0055] Step S100, respectively acquiring real-time monitoring data of the first posture sensor and the second posture sensor to obtain a first initial data set and a second initial data set, the first initial data set including at least one of pitch movement data, yaw movement data and roll movement data of the first joint body, and the second initial data set including at least one of pitch movement data, yaw movement data and roll movement data of the second joint body;

[0056] Specifically, the first initial data set can include pitch motion data, yaw motion data and roll motion data of the first joint body. The pitch motion data can be one or more of an angle, an acceleration, a displacement amount or a motion trajectory of the pitch motion; the yaw motion data can also be one or more of an angle, an acceleration, a displacement amount or a motion trajectory of the yaw motion; and the roll motion data can also be one or more of an angle, an acceleration, a displacement amount or a motion trajectory of the roll motion.

[0057] The pitch motion refers to the up-and-down motion of the joint body around the horizontal axis, such as the motion of the neck when a person raises or lowers his head; the yaw motion refers to the left-and-right rotation of the joint body around the vertical axis, like the rotation of the head when a person turns his head left or right; and the roll motion refers to the rotation of the joint body around the longitudinal axis, similar to the motion of the torso when a person twists his body. These data can be of various types, such as an angle value that can directly reflect the degree of rotation of the joint body in different motion directions; acceleration data that can reflect the speed change of the joint body motion; a displacement amount that can represent the position movement of the joint body in space; and a motion trajectory that can intuitively show the path of the joint body motion.

[0058] For example, the pitch motion data, the yaw motion data and the roll motion data of the first joint body sensed by the first attitude sensor can be obtained to obtain the first initial data set, the pitch motion data, the yaw motion data and the roll motion data of the second joint body sensed by the second attitude sensor can be obtained to obtain the second initial data set, and the first initial data set and the second initial data set can be taken as the data basis for joint motion analysis.

[0059] In step S200, the first initial data set and the second initial data set are subjected to data correction according to the deformation characteristics of the sensor mounting positions to obtain the first target data set and the second target data set.

[0060] In the joint motion monitoring process based on the attitude sensor, the first initial data set and the second initial data set obtained in step S100 contain the motion information of the first joint body and the second joint body, but the data has deviations. This deviation is mainly caused by the installation position of the attitude sensor and the characteristics of the soft tissues around the human joints.

[0061] The soft tissues around the human joints, such as muscles and skin, have elasticity and deformation ability. When the joint body moves, the soft tissues will deform accordingly. Taking the knee joint as an example, when the knee joint is bent, the soft tissues of the thigh and the lower leg will be squeezed and stretched. Since the attitude sensor is installed at a specific position of the joint body, such as the outer side of the thigh to reduce the influence of other environmental factors, and the second attitude sensor is generally installed on the outer side of the lower leg, the deformation of the soft tissues will cause the motion state of the attitude sensor itself to be inconsistent with the actual motion state of the joint body. This makes the motion data measured by the attitude sensor unable to accurately reflect the real motion of the joint body.

[0062] In order to obtain data truly reflecting the movement of the joint body, the first initial data set and the second initial data set are subjected to data correction in the embodiments of the present application. The data correction needs to be performed according to the deformation characteristics of the sensor mounting positions. The deformation characteristics of soft tissues at different joint positions are different. For example, the soft tissues around the knee joint are relatively thick and strong in flexibility, and the deformation degree during movement can be relatively large; while the soft tissues around the wrist joint are relatively thin, and the deformation degree can be relatively small. Through research and analysis of the mechanical properties and deformation rules of the soft tissues at specific joint positions, the deformation characteristics are determined, and specific parameters (such as deformation coefficients) are used to describe them.

[0063] For each item of movement data in the first initial data set, such as the pitch movement data and the yaw movement data, adjustment is performed in combination with the corresponding deformation characteristics. If the deformation coefficient of the first attitude sensor mounting position in a certain movement direction is known, the movement data in this direction in the initial data set can be corrected and calculated according to the coefficient, so as to reduce or even eliminate the error caused by the deformation of soft tissues. The same correction method is also used for the second initial data set, and finally the first target data set and the second target data set that can more accurately reflect the actual movement state of the first joint body and the second joint body are obtained.

[0064] The target data set obtained through data correction can lay a reliable data foundation for subsequent steps of accurately calculating the real-time movement angle of the first joint body relative to the second joint body, constructing a time sequence simulation joint motion diagram, and generating a warning signal.

[0065] In an embodiment, the joint is a knee joint, the first initial data set includes the pitch movement angle and the yaw movement angle of the thigh, and the second initial data set includes the pitch movement angle and the yaw movement angle of the lower leg; the step S200 of performing data correction on the first initial data set and the second initial data set according to the deformation characteristics of the sensor mounting positions to obtain the first target data set and the second target data set comprises:

[0066] Step S210, correcting the pitch movement angle and the yaw movement angle of the thigh according to a first deformation coefficient of the first attitude sensor at the thigh mounting position to obtain the first target data set, the first deformation coefficient representing the deformation ability of the soft tissues at the thigh mounting position;

[0067] Step S220, correcting the pitch movement angle and the yaw movement angle of the lower leg according to a second deformation coefficient of the second attitude sensor at the lower leg mounting position to obtain the second target data set, the second deformation coefficient representing the deformation ability of the soft tissues at the lower leg mounting position.

[0068] The first deformation coefficient is used to represent the deformation ability of the soft tissue at the thigh mounting position. Different individuals and different thigh positions have different deformation abilities of the soft tissue. For example, people who often do sports may have more developed muscles at the thigh position, and the soft tissue is relatively good in toughness, and the deformation coefficient may be small. For people who lack exercise, the soft tissue at the thigh position may be relatively loose, and the deformation coefficient may be large. Through research and analysis of a large number of samples, the deformation coefficient of the soft tissue at the thigh mounting position of a specific individual can be determined.

[0069] The second deformation coefficient is used to represent the deformation ability of the soft tissue at the calf mounting position. Similar to the thigh, the deformation ability of the soft tissue at the calf mounting position is also affected by individual factors, exercise habits, etc. For example, people who often run may have more compact muscles at the calf position, and the deformation coefficient of the soft tissue may be small. For people who sit for a long time, the soft tissue at the calf position may be relatively soft, and the deformation coefficient may be large. Through research and analysis, the deformation coefficient of the soft tissue at the calf mounting position of a specific individual can be determined.

[0070] It can be understood that for the first attitude sensor mounted at the same position of the thigh, the first deformation coefficient corresponding to the pitch motion and the first deformation coefficient corresponding to the yaw motion can be the same or different. When the thigh mounting position of the first attitude sensor has relatively uniform mechanical properties of the surrounding soft tissue in all directions, the first deformation coefficients of the pitch motion and the yaw motion can be the same. For example, in the middle outer region of the thigh of some individuals, the muscle distribution is relatively uniform and has no obvious directional difference. In this case, whether the pitch motion (thigh swing forward and backward) or the yaw motion (thigh rotation left and right) is performed, the deformation degree and mode of the soft tissue may be similar. Therefore, the first deformation coefficient used to describe the deformation ability of the soft tissue under the two motions can be set to the same value. Similarly, for the second attitude sensor mounted at the same position of the calf, the second deformation coefficient corresponding to the pitch motion and the second deformation coefficient corresponding to the yaw motion can be the same or different. Specific determination can be made according to individual differences.

[0071] In an embodiment, to improve the accuracy of data correction, the step S210 of correcting the pitch motion angle and the yaw motion angle of the thigh according to the first deformation coefficient of the first attitude sensor at the thigh mounting position to obtain a first target data set comprises:

[0072] A thigh soft tissue deformation correction model is constructed according to the first deformation coefficient. The correction model is based on the mechanical properties of human soft tissue, and takes the first deformation coefficient as a key parameter to describe the deformation law of the soft tissue at the thigh mounting position under different motion angles.

[0073] The initial measurement data is input into the thigh soft tissue deformation correction model, and the correction model is used to correct the pitch motion angle measurement value and the yaw motion angle measurement value in the initial measurement data to obtain the corrected pitch motion angle and the yaw motion angle;

[0074] The corrected pitch motion angle and the yaw motion angle are integrated to form a first target data set containing corrected motion angle information.

[0075] Specifically, human soft tissue has unique mechanical properties such as elasticity, viscoelasticity, etc., which determine that it will deform when subjected to force. In this embodiment, a thigh soft tissue deformation correction model is constructed based on the mechanical properties of human soft tissue, and a first deformation coefficient k1 is introduced as a key parameter in the model. The first deformation coefficient k1 is a quantitative index of the deformation ability of the soft tissue at the thigh mounting position, which comprehensively considers various factors such as the structure of the soft tissue (such as the orientation of muscle fibers, the distribution of fascia, etc.), material properties (such as elastic modulus, Poisson's ratio, etc.), and individual differences (such as age, gender, exercise habits, etc.). Through a large number of experimental research and data analysis, the first deformation coefficient k1 of different individuals and different thigh positions can be determined.

[0076] The core function of the correction model is to describe the deformation law of the soft tissue at the thigh mounting position under different motion angles. When the knee joint performs pitch motion or yaw motion, the thigh soft tissue will be subjected to corresponding force and moment, thereby deforming. The correction model can predict the deformation of the soft tissue according to the input motion angle (such as the pitch motion angle θ 1m measured initially) and the first deformation coefficient k1, providing a basis for subsequent correction of the measurement data.

[0077] Assuming that the thigh pitch motion angle measured by the first attitude sensor is θ 1m , the first deformation coefficient is k1, and the corrected thigh pitch motion angle θ 1t can be calculated through the pre-constructed correction model θ 1t = θ 1m − k1 × f(θ 1m ). Wherein, f(θ 1m ) is a function related to θ 1m , which is used to more accurately describe the influence of soft tissue deformation on the measured angle.

[0078] For the yaw motion angle of the thigh, a similar method is also used for correction. Assuming that the first attitude sensor measures the thigh yaw motion angle θ 2m, according to the first deformation coefficient k1 (in the yaw motion correction, the mechanism of the first deformation coefficient is similar, but the specific value may be different due to the different motion directions, here k1 is still used for simplification) and the corresponding function f(θ 2m ), the corrected yaw motion angle θ 2t .

[0079] The corrected pitch motion angle θ 1t and the yaw motion angle θ 2t are integrated to form a first target data set containing the corrected motion angle information. This data set reflects the motion state of the thigh in the actual motion process in a more accurate manner.

[0080] It can be understood that when the elasticity of the thigh sensor mounting part is greater, the displacement of the sensor elastic displacement is greater, and the difference between the angle value actually sensed by the sensor and the angle value actually moved by the thigh is greater. Therefore, the angle value actually moved by the thigh can be obtained by subtracting the correction value from the angle value actually sensed by the sensor.

[0081] In other embodiments, the pitch motion angle and the yaw motion angle of the lower leg are corrected according to the second deformation coefficient of the second attitude sensor at the lower leg mounting part to obtain a second target data set, and the correction principle is similar to the above, which will not be repeated here.

[0082] Step S300, determining the real-time motion angle of the first joint body relative to the second joint body according to the first target data set and the second target data set;

[0083] In an embodiment, step S300: determining the real-time motion angle of the first joint body relative to the second joint body according to the first target data set and the second target data set, comprising:

[0084] Step S310, obtaining the real-time pitch motion angle of the first joint body relative to the second joint body according to the real-time pitch motion angle of the first joint body, the real-time pitch motion angle of the second joint body and the pitch transformation relationship, and the pitch transformation relationship is the pitch motion coordinate transformation between the first joint body coordinates and the second joint body coordinates;

[0085] Step S320, obtaining the real-time yaw motion angle of the first joint body relative to the second joint body according to the real-time yaw motion angle of the first joint body, the real-time yaw motion angle of the second joint body and the yaw transformation relationship, and the yaw transformation relationship is the yaw motion coordinate transformation between the first joint body coordinates and the second joint body coordinates.

[0086] It can be understood that the first target data set contains the corrected real-time pitch motion angle of the first joint body (e.g., the thigh), and the second target data set contains the corrected real-time pitch motion angle of the second joint body (e.g., the lower leg). These data are accurate data corrected by step S200, and can more truly reflect the actual motion of the joint body.

[0087] The pitch transformation relationship describes the coordinate transformation of the pitch motion between the first joint body coordinate and the second joint body coordinate. In three-dimensional space, the motion of a joint body can be represented by a coordinate system, and different joint bodies have their own coordinate systems. The pitch motion refers to the up-and-down motion of the joint body around the horizontal axis, and the pitch transformation relationship considers the relative position and angle change of the two joint bodies in the pitch motion direction. For example, in the knee joint motion, the pitch motion of the thigh and the lower leg is interrelated, and by establishing the coordinate transformation relationship therebetween, the pitch motion angles of the thigh and the lower leg can be converted to relative angles.

[0088] Based on this, the real-time pitch motion angle of the first joint body relative to the second joint body can be obtained through a specific calculation formula according to the real-time pitch motion angle of the first joint body, the real-time pitch motion angle of the second joint body, and the pitch transformation relationship.

[0089] Similarly, the first target data set and the second target data set respectively provide the corrected real-time yaw motion angle of the first joint body and the second joint body. The yaw motion refers to the left-and-right rotation of the joint body around the vertical axis, and these data are the basis for calculating the relative yaw motion angle.

[0090] The yaw transformation relationship describes the coordinate transformation of the yaw motion between the first joint body coordinate and the second joint body coordinate. Similar to the pitch transformation relationship, it considers the relative position and angle change of the two joint bodies in the yaw motion direction. In the knee joint motion, the yaw motion of the thigh and the lower leg is also interrelated, and by establishing the yaw transformation relationship, their respective yaw motion angles can be converted to relative angles.

[0091] Based on this, the real-time yaw motion angle of the first joint body relative to the second joint body can be obtained through a corresponding calculation formula according to the real-time yaw motion angle of the first joint body, the real-time yaw motion angle of the second joint body, and the yaw transformation relationship.

[0092] It should be noted that the transformation between the coordinate systems of the two joint bodies can be described by Euler angles, and the Euler angle description method and the coordinate system transformation principle are prior art and not an improvement part of the present application, and the specific transformation process is not described in detail here.

[0093] Step S400, constructing a time sequence simulation joint motion diagram according to the real-time motion angle of the first joint body relative to the second joint body;

[0094] In an embodiment, the step S400 of constructing a time sequence simulation joint motion map according to the real-time motion angle of the first joint body relative to the second joint body comprises:

[0095] Obtaining real-time motion angle data of the first joint body relative to the second joint body in time sequence;

[0096] Mapping the real-time motion angle data at each time point to the corresponding position of a joint model according to a preset joint model, the joint model being a three-dimensional model simulating the connection relationship between the first joint body and the second joint body;

[0097] Combining the joint model states at each time point after mapping in time sequence to form a time sequence simulation joint motion map capable of dynamically displaying the motion process of the first joint body relative to the second joint body.

[0098] The real-time motion angle data is the real-time motion angle (which can be the pitch motion angle or the yaw motion angle or the roll motion angle) of the first joint body relative to the second joint body calculated in step S300. These data are obtained in time sequence and record the motion state of the joint at different times.

[0099] The preset joint model is a three-dimensional model simulating the connection relationship between the first joint body and the second joint body. This model accurately describes the geometric structure, motion range and connection mode of the joint, providing a basis for the mapping of real-time motion angle data.

[0100] Mapping the real-time motion angle data at each time point to the corresponding position of a joint model. Specifically, according to the real-time motion angle data, the position and posture of the first joint body relative to the second joint body in the joint model are adjusted to be consistent with the actual motion. For example, if the real-time motion angle data shows that the joint is at a specific pitch and yaw angle at a certain time, then the first joint body is adjusted to the corresponding angle position in the joint model.

[0101] Combining the joint model states at each time point after mapping in time sequence. The joint model state at each time point represents the motion posture of the joint at that time, and by arranging these states in sequence, a continuous motion sequence can be formed.

[0102] Using animation technology, the combined joint model state sequence is converted into a dynamic time sequence simulation joint motion map. This dynamic map can intuitively show the dynamic changes of the first joint body relative to the second joint body in the motion process, including the rotation, flexion and other actions of the joint.

[0103] The time sequence simulation joint motion graph provides an intuitive visual analysis means, enabling doctors and researchers to more clearly observe the movement process of the joint and discover potential movement abnormalities or pathologies. By observing the motion graph, doctors can more accurately determine the type and degree of joint disease, providing a basis for developing a treatment plan. In rehabilitation treatment, the time sequence simulation joint motion graph can be used to evaluate the rehabilitation effect of patients, observe the recovery of joint movement function, and timely adjust the rehabilitation plan.

[0104] Step S500: generating a warning signal according to the time sequence simulation joint motion graph or the real-time movement angle of the joint, the warning signal being used to remind the user to adjust the movement mode or stop the movement.

[0105] The present application can generate a warning signal through the real-time movement angle of the joint. For example, when the angle deviates from the threshold value by a large margin, it indicates that the joint movement may be in an abnormal state, at which time a warning signal is triggered. For example, in the movement of the knee joint, if the real-time pitch or yaw movement angle of the thigh relative to the lower leg exceeds the normal range, it may mean that the joint is subjected to excessive stress or there is a risk of movement injury, and the system will timely issue a warning signal.

[0106] It can be understood that the time sequence simulation joint motion graph can completely present the dynamic changes of the joint during the entire movement process. Compared with simply relying on the real-time movement angle of the joint, the motion graph contains more rich information, such as the position, posture and movement trajectory of the joint at different times, etc. For example, in the flexion and extension movement of the knee joint, the motion graph can clearly show the entire process from the starting position to the maximum flexion angle and back to the starting position, while the real-time movement angle can only provide the angle value at a certain time. Through observation and analysis of the entire movement process, the system can more accurately identify abnormal patterns in joint movement, such as unsmooth movement and abnormal posture, thereby improving the accuracy of the warning. Based on this, the present application can also generate a warning signal according to the time sequence simulation joint motion graph.

[0107] In an embodiment, the step S500 of generating a warning signal according to the time sequence simulation joint motion graph or the real-time movement angle of the joint comprises: inputting the time sequence simulation joint motion graph into a risk identification model to obtain a risk coefficient representation value; generating different levels of warning signals according to the risk coefficient representation value; wherein when the risk coefficient representation value is greater than or equal to a first threshold value, the user is reminded to stop the movement through a voice mode; and when the risk coefficient representation value is greater than or equal to a second threshold value and less than the first threshold value, the user is reminded to adjust the movement mode through a vibration mode.

[0108] Specifically, the time-series simulated joint motion video is first input into a pre-trained risk identification model. This model is trained based on a large amount of normal and abnormal joint motion video data, and can automatically extract feature information in the video through deep learning and other technologies, and judge the risk level of joint motion according to these features. The model outputs a risk coefficient representation value, which reflects the possibility of abnormal joint motion. According to the size of the risk coefficient representation value, the system generates different levels of warning signals.

[0109] When the risk coefficient representation value is greater than or equal to the first threshold value, it means that the joint motion has a high risk and may cause serious injury. At this time, the system will remind the user to stop moving through voice. Voice reminders have the characteristics of directness and clarity, which can ensure that users receive warning information in time and take appropriate action. For example, the system can issue a voice prompt: "Please note that your joint motion has a high risk, please stop moving immediately."

[0110] When the risk coefficient representation value is greater than or equal to the second threshold value and less than the first threshold value, it means that the joint motion has a certain risk, but has not reached the degree that needs to be stopped immediately. The system will remind the user to adjust the movement mode through vibration to reduce the risk. Vibration reminders can attract the user's attention without disturbing the user's normal movement. At the same time, the system can also combine a simple prompt tone to enhance the reminder effect. For example, the device vibrates and is accompanied by a prompt tone: "Please adjust the movement mode to avoid excessive movement."

[0111] Under other necessary conditions, the system can also send warning signals to users or supervisors through APP push.

[0112] In an embodiment, the inputting the time-series simulated joint motion video into a risk identification model to obtain a risk coefficient representation value comprises: performing feature extraction on the time-series simulated joint motion video to obtain a joint motion feature vector, the joint motion feature vector comprising at least two of joint motion angle features, joint motion speed features, joint motion acceleration features, and joint motion trajectory features; inputting the joint motion feature vector into a pre-trained risk identification model, the risk identification model being constructed based on a deep learning algorithm; the risk identification model calculating based on the input joint motion feature vector and outputting a risk coefficient representation value, the risk coefficient representation value reflecting the possibility of risk in the current joint motion state.

[0113] Specifically, computer vision and image processing technologies can be used to analyze and process the time-series simulated joint motion video. For example, through edge detection, feature point tracking and other methods, the position information of the joint is extracted, and then the angle, speed, acceleration and other features are calculated.

[0114] A deep learning algorithm is used to build a risk identification model, such as a convolutional neural network (CNN), a recurrent neural network (RNN), and its variants (such as LSTM, GRU), etc. These deep learning models have strong feature learning and classification capabilities, and can automatically learn the complex relationship between joint motion features and risks from a large amount of training data. A large amount of labeled normal and abnormal joint motion data is used to train the model. During the training process, the model continuously adjusts its parameters so that the risk coefficient representation value output after inputting the joint motion feature vector can accurately reflect the possibility of the joint motion state being risky. The extracted joint motion feature vector is input into the pre-trained risk identification model. The model calculates according to the input feature vector and outputs a risk coefficient representation value. This value is a continuous numerical value, and the range can be set according to the specific application scenario, for example, between 0 and 1, and the larger the value, the higher the possibility of the joint motion state being risky.

[0115] For example, in the scenario of monitoring the knee joint motion state. First, the time series simulation knee joint motion diagram is feature extracted to obtain a feature vector including the knee joint flexion angle, motion speed, acceleration, and motion trajectory, etc. Then, these feature vectors are input into the pre-trained LSTM-based risk identification model. After calculation, the model outputs a risk coefficient representation value, for example, 0.7 (assuming the risk coefficient range is 0-1). If the first threshold is set to 0.8 and the second threshold is set to 0.5, since the risk coefficient representation value is greater than the second threshold but less than the first threshold, the system will issue a vibration reminder at this time, prompting the user to adjust the motion mode.

[0116] The application can also store the monitored angle data, simulation analysis results, and early warning records, etc. information, and establish a personal joint motion health database. Using data analysis technology, personalized exercise suggestions and rehabilitation programs are provided for users to help them better protect their joints.

[0117] Based on this, this application provides a closed-loop monitoring and intelligent early warning method based on attitude sensors. First, real-time monitoring data from a first attitude sensor and a second attitude sensor are acquired to obtain a first initial dataset and a second initial dataset. Then, the first initial dataset and the second initial dataset are corrected according to the deformation characteristics of the sensor mounting location to obtain a first target dataset and a second target dataset. Next, the real-time motion angle of the first joint relative to the second joint is determined based on the first target dataset and the second target dataset. Then, a time-series simulated joint animation is constructed based on the real-time motion angle of the first joint relative to the second joint. Finally, an early warning signal is generated based on the time-series simulated joint animation or the real-time motion angle of the joint to remind the user to adjust their movement or stop exercising. Thus, the technical solution of this application can effectively reduce measurement errors caused by deformation of the soft tissues around the joint, improve the accuracy of joint motion monitoring, and present the joint motion process in real-time, dynamically, and intuitively, promptly detect abnormal motion states and issue early warnings, effectively prevent sports injuries, and improve the safety and rationality of exercise.

[0118] like Figure 5 As shown, Figure 5 The diagram below shows the hardware structure of a closed-loop monitoring and intelligent early warning system in some embodiments of this application. The closed-loop monitoring and intelligent early warning system provided in the embodiments of this application further includes a memory 1000 and a processor 2000. The memory 1000 is used to store computer-readable instructions, and the processor 2000 is used to call the computer-readable instructions to execute the closed-loop monitoring and intelligent early warning method based on attitude sensors as described above.

[0119] The processor 2000 is configured to provide computing and control capabilities to control the closed-loop monitoring and intelligent early warning system to perform corresponding tasks, for example, to control the closed-loop monitoring and intelligent early warning system to perform the posture sensor-based closed-loop monitoring and intelligent early warning method in any of the above method embodiments. The method includes: obtaining first initial data set and second initial data set by respectively acquiring real-time monitoring data of the first posture sensor and the second posture sensor, wherein the first initial data set includes at least one of pitch motion data, yaw motion data and roll motion data of the first joint body, and the second initial data set includes at least one of pitch motion data, yaw motion data and roll motion data of the second joint body; performing data correction on the first initial data set and the second initial data set according to the deformation characteristics of the sensor mounting part to obtain first target data set and second target data set; determining the real-time motion angle of the first joint body relative to the second joint body according to the first target data set and the second target data set; constructing a time sequence simulation joint motion map according to the real-time motion angle of the first joint body relative to the second joint body; and generating an early warning signal according to the time sequence simulation joint motion map or the real-time motion angle of the joint, wherein the early warning signal is used to remind the user to adjust the movement mode or stop moving.

[0120] The processor 2000 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a hardware chip or any combination thereof; and can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof. The above-mentioned PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof.

[0121] The memory 1000, as a non-transitory computer readable storage medium, can be used to store non-transitory software programs, non-transitory computer executable programs and modules, such as program instructions / modules corresponding to the closed-loop monitoring and intelligent early warning method based on the attitude sensor in the embodiments of the present application. The processor 2000 can implement the closed-loop monitoring and intelligent early warning method based on the attitude sensor in any of the method embodiments described above by running the non-transitory software programs, instructions and modules stored in the memory 1000.

[0122] Specifically, the memory 1000 can include volatile memory (VM), such as random access memory (RAM); the memory 1000 can also include non-volatile memory (NVM), such as read-only memory (ROM), flash memory, hard disk drive (HDD) or solid-state drive (SSD) or other non-transitory solid-state storage devices; the memory 1000 can also include a combination of the above types of memory.

[0123] In summary, the closed-loop monitoring and intelligent early warning system of the present application adopts the technical solutions of any one of the above-mentioned closed-loop monitoring and intelligent early warning method embodiments based on the attitude sensor, and therefore at least has the beneficial effects brought by the technical solutions of the above-mentioned embodiments, which will not be repeated here.

[0124] The embodiments of the present application also provide a computer readable storage medium, such as a memory including program code, which can be executed by a processor to complete the closed-loop monitoring and intelligent early warning method based on the attitude sensor in the above-mentioned embodiments. For example, the computer readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CDROM), a magnetic tape, a floppy disk and an optical data storage device, etc.

[0125] The embodiments of the present application also provide a computer program product, which includes one or more program codes stored in a computer readable storage medium. The processor of the early warning system reads the program codes from the computer readable storage medium, and the processor executes the program codes to complete the steps of the closed-loop monitoring and intelligent early warning method based on the attitude sensor provided in the above-mentioned embodiments.

[0126] Those skilled in the art can understand that all or part of the steps of the above-mentioned embodiments can be completed by hardware, or by program code related hardware, and the program can be stored in a computer readable storage medium. The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0127] It should be noted that the above-described device embodiments are merely illustrative, and the units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e. they can be located in one place or distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0128] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be implemented by means of software plus a general hardware platform, and of course, can also be implemented by hardware. Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing related hardware, and the program can be stored in a computer readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiment methods. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0129] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structural transformation made under the inventive concept of the present application, or direct / indirect application in other related technical fields is included in the patent protection scope of the present application.

Claims

1. A closed-loop monitoring and intelligent early warning method based on attitude sensors, characterized in that, The method, applied to the monitoring and early warning of joint movement, includes: Real-time monitoring data from the first attitude sensor and the second attitude sensor are acquired to obtain a first initial dataset and a second initial dataset. The first initial dataset includes at least one of pitch motion data, yaw motion data and roll motion data of the first joint body. The second initial dataset includes at least one of pitch motion data, yaw motion data and roll motion data of the second joint body. The first initial dataset and the second initial dataset are corrected based on the deformation characteristics of the sensor mounting location to obtain the first target dataset and the second target dataset; The real-time motion angle of the first joint body relative to the second joint body is determined based on the first target dataset and the second target dataset. A time-series simulated joint animation is constructed based on the real-time motion angle of the first joint body relative to the second joint body. A warning signal is generated based on the time-series simulated joint animation or the real-time movement angle of the joint. The warning signal is used to remind the user to adjust the movement mode or stop the movement. Determining the real-time motion angle of the first joint body relative to the second joint body based on the first target dataset and the second target dataset includes: The real-time pitch angle of the first joint relative to the second joint is obtained based on the real-time pitch angle of the first joint, the real-time pitch angle of the second joint, and the pitch transformation relationship. The pitch transformation relationship is the pitch coordinate transformation between the coordinates of the first joint and the coordinates of the second joint. The real-time yaw angle of the first joint body relative to the second joint body is obtained based on the real-time yaw angle of the first joint body, the real-time yaw angle of the second joint body, and the yaw transformation relationship. The yaw transformation relationship is the yaw motion coordinate transformation between the coordinates of the first joint body and the coordinates of the second joint body. The step of constructing a time-series simulated joint animation based on the real-time motion angle of the first joint body relative to the second joint body includes: The real-time motion angle data of the first joint body relative to the second joint body are obtained sequentially according to time order; According to the preset joint model, the real-time motion angle data at each time point is mapped to the corresponding position of the joint model. The joint model is a three-dimensional model that simulates the connection relationship between the first joint body and the second joint body. The mapped joint model states at each time point are combined in chronological order to form a time-series simulated joint animation that can dynamically display the movement process of the first joint body relative to the second joint body.

2. The closed-loop monitoring and intelligent early warning method based on attitude sensors as described in claim 1, characterized in that, The motion data includes one or more of the following: acceleration, displacement, angle, or motion trajectory.

3. The closed-loop monitoring and intelligent early warning method based on attitude sensors as described in claim 2, characterized in that, The joint is the knee joint. The first initial dataset includes the pitch angle and yaw angle of the thigh, and the second initial dataset includes the pitch angle and yaw angle of the lower leg. The step of correcting the first initial dataset and the second initial dataset based on the deformation characteristics of the sensor mounting location to obtain the first target dataset and the second target dataset includes: The first target dataset is obtained by correcting the pitch and yaw angles of the thigh based on the first deformation coefficient of the first attitude sensor at the thigh mounting site. The first deformation coefficient characterizes the deformation capability of the soft tissue at the thigh mounting site. The second target dataset is obtained by correcting the pitch and yaw angles of the lower leg based on the second deformation coefficient of the second attitude sensor at the lower leg mounting location. The second deformation coefficient characterizes the deformation capability of the soft tissue at the lower leg mounting location.

4. The closed-loop monitoring and intelligent early warning method based on attitude sensors as described in claim 3, characterized in that, The first target dataset is obtained by correcting the pitch and yaw angles of the thigh based on the first deformation coefficient of the first attitude sensor at the thigh mounting location, including: A thigh soft tissue deformation correction model is constructed based on the first deformation coefficient. The correction model is based on the mechanical properties of human soft tissue and uses the first deformation coefficient as a key parameter to describe the deformation law of the soft tissue at the thigh installation site under different motion angles. The initial measurement data is input into the thigh soft tissue deformation correction model. The correction model is used to correct the pitch angle and yaw angle measurements in the initial measurement data to obtain the corrected pitch angle and yaw angle. The corrected pitch and yaw angles are integrated to form a first target dataset containing the corrected motion angle information.

5. The closed-loop monitoring and intelligent early warning method based on attitude sensor as described in claim 1, characterized in that, The warning signal includes one or more of the following: voice signal, vibration signal, or APP push signal.

6. The closed-loop monitoring and intelligent early warning method based on attitude sensors as described in claim 5, characterized in that, Based on the time-series simulated joint animation or the real-time motion angle of the joint, an early warning signal is generated, including: The time-series simulated joint animation is input into the risk identification model to obtain the risk coefficient characterization value; Different levels of early warning signals are generated based on the risk coefficient characterization value; Specifically, when the risk coefficient value is greater than or equal to the first threshold, the user is reminded to stop exercising via voice; when the risk coefficient value is greater than or equal to the second threshold but less than the first threshold, the user is reminded to adjust their exercise mode via vibration.

7. The closed-loop monitoring and intelligent early warning method based on attitude sensors as described in claim 6, characterized in that, The step of inputting the time-series simulated joint animation into the risk identification model to obtain the risk coefficient characterization value includes: Feature extraction is performed on the time-series simulated joint animation to obtain a joint motion feature vector, which includes at least two of the following: joint motion angle feature, joint motion velocity feature, joint motion acceleration feature, and joint motion trajectory feature. The joint motion feature vector is input into a pre-trained risk identification model, which is constructed based on a deep learning algorithm. The risk identification model calculates based on the input joint motion feature vector and outputs a risk coefficient representation value, which reflects the likelihood of a risk in the current joint motion state.

8. A closed-loop monitoring and intelligent early warning system, characterized in that, include: A first attitude sensor is disposed on a first joint body; A second attitude sensor is disposed on a second joint body, and the second joint body is rotatably connected to the first joint body to form a joint. as well as A memory and a processor, wherein the memory is used to store program code; The processor is used to call the program code to perform the method as described in any one of claims 1 to 7.

Images