Health risk prediction method and equipment based on behavior data

Abstract

The invention is suitable for the technical field of risk prediction, and particularly relates to a health risk prediction method and equipment based on behavior data, and the method comprises the steps: obtaining foot information in each walking period in real time; acquiring ridge movement information in each walking period in real time; analyzing according to the plurality of pieces of ridge movement information and the plurality of pieces of foot information to obtain a potential energy data chain; analyzing according to the historical tumble event to obtain a tumble threshold value; analyzing according to the falling threshold and the potential energy data chain to obtain a predicted falling period; and the control device controls a prediction reminding device to give out a fall risk early warning based on the predicted fall period. According to the health risk prediction method based on behavior data provided by the invention, before the infant may fall down, parents can be timely and effectively reminded to provide supporting force in advance so as to prevent the infant from falling down.

Description

Technical Field

[0001] The present application belongs to the technical field of risk prediction, and in particular, relates to a method and device for predicting health risks based on behavioral data. Background Art Health risk prediction methods based on behavioral data refer to technologies that assess an individual's future health status or risk of disease through their behavioral patterns and lifestyle data. For example, predictions of obesity or other chronic diseases can be made through an individual's eating habits and exercise frequency, predictions of mental health or cardiovascular disease can be made through an individual's stress level and emotional state, and predictions of falls in infants and toddlers.

[0002] In the related art, especially in the prediction of falls of infants and young children in the process of learning to walk, because the bones of infants who need to learn to walk are still developing, falls may cause fractures in relatively fragile bones such as wrists, elbows and clavicles. Parents are usually required to pay attention to the movement posture of infants and young children at all times or use modern technology such as wearable devices to monitor the movement pattern of infants and young children and send out an alarm when the device detects abnormal movements, reminding parents to protect infants and young children in time to reduce the occurrence of falls of infants and young children. Because parents fail to pay attention to the movement posture of infants and young children in time, they may not be able to provide support in time, which may cause infants and young children to fall. When the wearable device monitors the movement pattern of infants and young children and sends out an alarm when the device detects abnormal movements, the device that reminds them of falls may only send out an alarm reminder when abnormal movements are detected. When parents hear the alarm reminder, the infants and young children may have already started to fall or are already in the process of falling, resulting in parents being unable to provide support in time, which ultimately causes the infants and young children to fall. Summary of the invention

[0003] The embodiments of the present application provide a health risk prediction method and device based on behavioral data, which can improve the problem of infants and young children falling due to parents' inability to provide timely support to infants and young children during the process of infants and young children learning to walk.

[0004] In a first aspect, the present application embodiment provides a health risk prediction method based on behavioral data, comprising: Acquire foot information in each walking cycle in real time; wherein the foot information includes plantar pressure information and gait information, the plantar pressure information is used to reflect the distribution of plantar pressure of the user, the gait information includes step length and walking cycle, the step length is used to reflect the distance between the heel of one foot when the user touches the ground and the heel of the same foot when the same foot touches the ground next time, and the walking cycle is used to indicate the time from one foot of the user leaving the ground to the next time the other foot completely touches the ground after leaving the ground; Acquire the spinal movement information in each walking cycle in real time; wherein the spinal movement information is used to reflect the movement trajectory of the spinal monitoring point; A potential energy data chain is obtained by analyzing the plurality of spine movement information and the plurality of foot information; wherein the potential energy data chain is used to reflect a data chain composed of potential energy data corresponding to each walking cycle of the user, and the potential energy data is used to reflect the energy required to maintain the balance of the body; Analyze historical fall events to obtain a fall threshold; wherein the fall threshold is used to reflect the potential energy data when the user falls; The predicted fall cycle is obtained by analyzing the fall threshold and the potential energy data chain; wherein the predicted fall cycle is used to reflect the time from the start of the latest walking cycle of the user to the time when the potential energy data reflected by the potential energy data chain is equal to the fall threshold; The control device controls the prediction and reminder device to issue a fall risk warning based on the predicted fall cycle.

[0005] The above technical solutions in the embodiments of the present application have at least the following technical effects: The health risk prediction method based on behavioral data provided in the embodiment of the present application first obtains in real time the plantar pressure information for reflecting the plantar pressure distribution of the infant and the foot information for reflecting the step length and gait information of the walking cycle generated by the infant in the process of learning to walk, and at the same time obtains in real time the spinal movement information for reflecting the movement trajectory of the spinal monitoring point in each walking cycle in the process of learning to walk, and then analyzes based on multiple spinal movement information and multiple foot information to obtain a potential energy data chain for reflecting the potential energy data corresponding to each walking cycle of the user for reflecting the energy required to maintain the balance of the body, and then analyzes based on historical fall events to obtain a fall threshold for reflecting the potential energy data when the user falls, and then analyzes based on the fall threshold and the potential energy data chain to obtain the time from the latest walking cycle of the user to the time when the potential energy data reflected by the potential energy data chain is equal to the fall threshold, and finally the control device controls the prediction reminder device to issue a fall risk warning based on the predicted fall cycle. This method can effectively obtain a potential energy data chain by analyzing the real-time acquired spine movement information and the real-time acquired foot information, and obtain a fall threshold by analyzing historical fall events, and then obtain a predicted fall cycle by analyzing the fall threshold and the potential energy data chain. It can predict the situation in which the potential energy data of each infant's subsequent walking cycle falls below the fall threshold based on the walking differences of infants and young children, and issue an alarm in advance, so that when the infant has not yet started to fall, the parents are promptly reminded to provide support in advance to prevent the infant from falling.

[0006] In a possible implementation manner of the first aspect, the analyzing the spine movement information and the foot information to obtain a potential energy data link includes: Performing dimension segmentation according to the spine movement information to obtain horizontal offset information; wherein the horizontal offset information is used to reflect the offset trajectory of the spine monitoring point in the horizontal direction; Analyze the foot information to obtain a trajectory of a force point and a symmetry index; wherein the trajectory of the force point is used to reflect the trajectory of a concentrated point of pressure on the sole of the foot during a walking cycle; and the symmetry index is used to reflect the symmetry of the bilateral lower limbs during a walking cycle; A first coupling coefficient and a second coupling coefficient are obtained by analyzing the horizontal offset information and the trajectory of the force point; wherein the first coupling coefficient is used to reflect the similarity between the offset trajectory of the spine monitoring point in the horizontal offset information in the walking cycle and the trajectory of the force point of the left foot, and the second coupling coefficient is used to reflect the similarity between the offset trajectory in the right direction of the horizontal offset information in the walking cycle and the trajectory of the force point of the right foot; A synchronization coefficient is obtained by performing sum processing on the first coupling coefficient and the second coupling coefficient; wherein the synchronization coefficient is used to reflect the coordination ability of the foot spine motion chain in one walking cycle; A potential energy data link is obtained by analyzing the synchronization coefficients in a plurality of walking cycles and the symmetry indexes in a plurality of walking cycles.

[0007] In a possible implementation manner of the first aspect, analyzing according to the foot information to obtain a trajectory of a force point includes: Acquire a sole image; wherein the sole image is used to reflect the shape of the sole of the user's foot; Assigning the plantar pressure information of the foot information to the corresponding position in the plantar graph according to the position and performing analysis and processing to obtain a spatiotemporal pressure thermodynamic map; wherein the spatiotemporal pressure thermodynamic map is used to reflect the pressure distribution at different positions of the user's sole when the user's sole contacts the ground; According to the time-space pressure thermodynamic map, the pressure is analyzed and processed in the direction of increasing pressure to obtain multiple force points and timestamps corresponding to the multiple force points; wherein the force point is used to reflect the concentrated point of pressure on the sole of the foot during a walking cycle; and the timestamp is used to reflect the specific time point of the force point; The plurality of force points are sorted in ascending order based on the timestamps to obtain a force point trajectory.

[0008] In a possible implementation manner of the first aspect, analyzing according to the foot information to obtain a symmetry index includes: Acquire a left air time and a right air time in the walking cycle of the foot information; wherein the left air time is used to reflect the time when the sole of the left foot is free of pressure, and the right air time is used to reflect the time when the sole of the right foot is free of pressure; Obtaining a left step length and a right step length from the foot information; wherein the left step length is used to reflect the distance between the heel of the left foot when it lands and the heel of the left foot when it lands again, and the right step length is used to reflect the distance between the heel of the right foot when it lands and the heel of the right foot when it lands again; Comparing the left dwell time with the left step length, to obtain a left comparison value; wherein the left comparison value is used to reflect the ratio between the left dwell time and the left step length; Comparing the right hovering time with the right step length, to obtain a right comparison value; wherein the right comparison value is used to reflect the ratio between the right hovering time and the right step length; A symmetry index is obtained by processing the left contrast value and the right contrast value.

[0009] In a possible implementation manner of the first aspect, the obtaining the first coupling coefficient and the second coupling coefficient by analyzing the horizontal offset information and the focus point trajectory includes: Processing is performed according to the horizontal offset information to obtain a left offset trajectory and a right offset trajectory; wherein the left offset trajectory is used to reflect the offset trajectory of the spine monitoring point in the horizontal offset information in the left direction, and the right offset trajectory is used to reflect the offset trajectory in the right direction in the horizontal offset information; Processing the right foot trajectory of the right offset trajectory and the force point trajectory to obtain a first coupling coefficient; The second coupling coefficient is obtained by processing the left foot trajectory of the left offset trajectory and the force point trajectory.

[0010] In a possible implementation manner of the first aspect, the analyzing according to the plurality of synchronization coefficients and the plurality of symmetry indexes to obtain a potential energy data link includes: Processing is performed according to the synchronization coefficient and the symmetry index to obtain potential energy data; wherein the potential energy data is used to reflect the overall stability of the human body's dynamic balance system; The potential energy data corresponding to the plurality of walking cycles are sorted according to the walking cycle sequence to obtain a potential energy data chain.

[0011] In a possible implementation manner of the first aspect, analyzing historical fall events to obtain a fall threshold includes: Acquire multiple fall potential energy data of historical fall events; wherein the fall potential energy data is used to reflect the potential energy data when the user falls; An analysis is performed based on the plurality of fall potential energy data to obtain average potential energy data, and the average potential energy data is confirmed as a fall threshold.

[0012] In a possible implementation manner of the first aspect, the analyzing according to the fall threshold and the potential energy data link to obtain a predicted fall cycle includes: Analyze the potential energy data chain to obtain the final potential energy; wherein the final potential energy is used to reflect the potential energy data of the user's latest completed walking cycle; Performing screening processing according to the potential energy data chain to obtain a plurality of reduced data; wherein the reduced data is used to indicate that the i-th potential energy data in the potential energy data chain is less than the i-1-th potential energy data, and the i+1-th potential energy data is less than the i-th data point of the i-1-th potential energy data; Processing is performed according to the plurality of the reduction data to obtain a fall risk curve; wherein the fall risk curve is used to reflect a curve formed by the plurality of the reduction data; Processing the fall risk curve and the plurality of reduction data to obtain a fall trend; wherein the fall trend is used to reflect the rising trend of the potential energy data in one walking cycle; The predicted fall cycle is obtained by processing the final potential energy, the fall threshold and the fall trend.

[0013] In a possible implementation manner of the first aspect, the processing according to the fall risk curve and the plurality of reduction data to obtain a fall trend includes: Processing the fall risk curve and the plurality of reduction data to obtain a plurality of change rates; wherein the change rates are used to indicate the slope of a tangent line in the fall risk curve corresponding to the reduction data; Processing is performed based on the multiple change rates to obtain an average change rate, and the average change rate is confirmed as a falling trend; wherein the average change rate is used to reflect the average value between the multiple change rates.

[0014] In a second aspect, the embodiment of the present application provides a health risk prediction system based on behavioral data, including: A first acquisition module is used to acquire foot information in real time during each walking cycle; wherein the foot information includes plantar pressure information and gait information, the plantar pressure information is used to reflect the distribution of plantar pressure of the user, the gait information includes step length and walking cycle, the step length is used to reflect the distance between the heel of one foot when the user touches the ground and the heel of the same foot when the same foot touches the ground next time, and the walking cycle is used to indicate the time from one foot of the user leaving the ground to the next time the other foot completely touches the ground after leaving the ground; A second acquisition module is used to acquire the spinal movement information in each walking cycle in real time; wherein the spinal movement information is used to reflect the movement trajectory of the spinal monitoring point; A first analysis module is used to analyze the plurality of spine movement information and the plurality of foot information to obtain a potential energy data chain; wherein the potential energy data chain is used to reflect a data chain composed of potential energy data corresponding to each walking cycle of the user, and the potential energy data is used to reflect the energy required to maintain the balance of the body; A second acquisition module is used to analyze historical fall events to obtain a fall threshold; wherein the fall threshold is used to reflect the potential energy data when the user falls; A third acquisition module is used to analyze the fall threshold and the potential energy data chain to obtain a predicted fall cycle; wherein the predicted fall cycle is used to reflect the time from the start of the latest walking cycle of the user to the time when the potential energy data reflected by the potential energy data chain is equal to the fall threshold; The control module is used to control the device to control the prediction and reminder device to issue a fall risk warning based on the predicted fall cycle.

[0015] In the third aspect, an embodiment of the present application provides a health risk prediction device based on behavioral data, including a prediction reminder device and a control device, wherein the prediction reminder device is electrically connected to the control device, and the control device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements a method as described in any one of the first aspects above.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of the first aspects above is implemented.

[0017] In a fifth aspect, an embodiment of the present application provides a computer program, which, when executed on a health risk prediction device based on behavioral data, enables the health risk prediction device based on behavioral data to execute the health risk prediction method based on behavioral data described in any one of the above-mentioned first aspects.

[0018] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 It is a flowchart of a health risk prediction method based on behavioral data provided by an embodiment of the present application; Figure 2 This is a schematic diagram of the implementation process of a health risk prediction method based on behavioral data provided in an embodiment of the present application; Figure 3 is a structural diagram of a health risk prediction system based on behavioral data provided by an embodiment of the present application; Figure 4 It is a structural schematic diagram of a control device of a health risk prediction device based on behavioral data provided in one embodiment of the present application. DETAILED DESCRIPTION

[0021] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0022] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0023] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0024] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0025] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0026] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0027] In the related art, especially in the prediction of falls of infants and young children in the process of learning to walk, because the bones of infants who need to learn to walk are still developing, falls may cause fractures in relatively fragile bones such as wrists, elbows and clavicles. Parents are usually required to pay attention to the movement posture of infants and young children at all times or use modern technology such as wearable devices to monitor the movement pattern of infants and young children and send out an alarm when the device detects abnormal movements, reminding parents to protect infants and young children in time to reduce the occurrence of falls of infants and young children. Because parents fail to pay attention to the movement posture of infants and young children in time, they may not be able to provide support in time, which may cause infants and young children to fall. When the wearable device monitors the movement pattern of infants and young children and sends out an alarm when the device detects abnormal movements, the device that reminds them of falls may only send out an alarm reminder when abnormal movements are detected. When parents hear the alarm reminder, the infants and young children may have already started to fall or are already in the process of falling, resulting in parents being unable to provide support in time, which ultimately causes the infants and young children to fall.

[0028] To solve the above problems, the embodiments of the present application provide a health risk prediction method and device based on behavioral data. In the method, first, in real time, the foot pressure information for reflecting the foot pressure distribution of the infant and the foot information for reflecting the step length and the gait information of the walking cycle generated by the infant in the process of learning to walk is obtained in a walking cycle for indicating the time from one foot of the infant to the next time the other foot leaves the ground. At the same time, the spine movement information for reflecting the movement trajectory of the spine monitoring point in each walking cycle in the process of learning to walk is obtained in real time. Then, according to the plurality of spine movement information and the plurality of foot information, a potential energy data chain is obtained to reflect the potential energy data corresponding to each walking cycle of the user for reflecting the energy required to maintain the balance of the body. Then, according to the historical fall events, a fall threshold for reflecting the potential energy data when the user falls is obtained. Then, according to the fall threshold and the potential energy data chain, an analysis is performed to obtain the time for reflecting the time from the latest walking cycle of the user to the time when the potential energy data reflected by the potential energy data chain is equal to the fall threshold. Finally, the control device controls the prediction reminder device to issue a fall risk warning based on the predicted fall cycle. This method can effectively obtain a potential energy data chain by analyzing the real-time acquired spine movement information and the real-time acquired foot information, and obtain a fall threshold by analyzing historical fall events, and then obtain a predicted fall cycle by analyzing the fall threshold and the potential energy data chain. It can predict the situation in which the potential energy data of each infant's subsequent walking cycle falls below the fall threshold based on the walking differences of infants and young children, and issue an alarm in advance, so that when the infant has not yet started to fall, the parents are promptly reminded to provide support in advance to prevent the infant from falling.

[0029] The health risk prediction method based on behavioral data provided in the embodiment of the present application can be applied to a health risk prediction device based on behavioral data. In this case, the health risk prediction device based on behavioral data is the executor of the health risk prediction method based on behavioral data provided in the embodiment of the present application. The embodiment of the present application does not impose any restrictions on the specific type of the health risk prediction device based on behavioral data.

[0030] The health risk prediction device based on behavioral data includes a prediction reminder device and a control device. The prediction reminder device is electrically connected to the control device. The prediction reminder device includes a detection device and an early warning device. The detection device includes a first detection mechanism and a second detection mechanism. The first detection device is used to detect the movement trajectory of the spinal monitoring point. For example, the first detection mechanism can be a wearable device provided with a displacement sensor, such as a detection strap or a detection belt. The second detection mechanism is used to detect the pressure on the sole of the foot of the user and the walking cycle of the user. For example, the second detection mechanism can be an intelligent insole or a combination device integrating a pressure sensor and a timer. Among them, the pressure sensor is used to detect the pressure on the sole of the foot of the user. The timer is used to detect the walking cycle of the user. The early warning device is used to remind the user of falling. For example, the early warning device can be a sound alarm or a flashing alarm light. The control device is used to supervise and control the fall prediction process.

[0031] For example, the control device can be a single-chip microcomputer, a microcontroller, an application-specific integrated circuit, a mobile phone, a tablet computer, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a smart large screen, a smart TV, a handheld device with wireless communication function, a desktop computer, a handheld device with wireless communication function, a computer, a laptop computer, a handheld computing device, etc.

[0032] In order to better understand the health risk prediction method based on behavioral data provided in the embodiment of the present application, the specific implementation process of the health risk prediction method based on behavioral data provided in the embodiment of the present application is exemplarily introduced below.

[0033] Figure 1 and Figure 2 A schematic flow chart of a health risk prediction method based on behavioral data provided in an embodiment of the present application is shown. Figure 1 and Figure 2 , health risk prediction methods based on behavioral data include: S100, obtaining foot information in real time during each walking cycle; wherein the foot information includes plantar pressure information and gait information, the plantar pressure information is used to reflect the distribution of plantar pressure of the user, the gait information includes step length and walking cycle, the step length is used to reflect the distance between the heel of one foot when the user touches the ground and the heel of the same foot when the same foot touches the ground next time, and the walking cycle is used to indicate the time from one foot of the user leaving the ground to the next time the other foot completely touches the ground after leaving the ground.

[0034] Exemplarily, wearable devices such as smart insoles or plantar pressure sensors can be installed on the soles of the user's feet or soles to obtain foot information. Foot information can also be manually input. The mechanism for obtaining plantar pressure information through the device is to detect the pressure exerted by the sole of the foot on the insole when the user walks through the built-in flexible pressure sensor of the wearable device, so that the electrical characteristics of the pressure sensor change. For example, in a resistive sensor, the change in pressure exerted by the sole of the user on the insole will cause a change in the resistance value, and then the change in resistance value is measured by the circuit and converted into a digital signal, and finally the plantar pressure information is obtained. The mechanism for obtaining gait information through the device is that the step length can be estimated through the sensor integrated in the shoe or leg by the inertial measurement unit (IMU). The accelerometer and gyroscope in the inertial measurement unit (IMU) provide information about the direction and speed of the user when walking, and then the displacement information is obtained by double-integrating the acceleration data, that is, the step length. The walking cycle can be determined by detecting two consecutive identical gait events.

[0035] S200, acquiring spine movement information in real time during each walking cycle; wherein the spine movement information is used to reflect the movement trajectory of the spine monitoring point.

[0036] It can be understood that the spine monitoring point refers to the monitoring point close to the spine after the user wears the wearable device such as a detection strap or a detection belt. The detection point refers to the position point of the displacement sensor set in the detection strap or the detection belt.

[0037] S300, analyzing multiple spine movement information and multiple foot information to obtain a potential energy data chain; wherein the potential energy data chain is used to reflect a data chain composed of potential energy data corresponding to each walking cycle of the user; the potential energy data is used to reflect the energy required to maintain a balanced body balance.

[0038] It is understandable that when a user walks, the body needs to provide corresponding potential energy for the body to remain stable. When the user's walking stability is not high, more potential energy is required for the body to remain stable than when the user is stable. When the user's walking stability is not low, less potential energy is required to keep the body stable than when the user is unstable.

[0039] Exemplarily, the horizontal offset trajectory of the spinal monitoring point can be extracted by processing the spinal movement information, and then the foot information can be analyzed to obtain the trajectory of the concentrated point reflecting the pressure on the sole of the user in a walking cycle and the symmetry of the lower limbs of the user in a walking cycle. Then, the horizontal offset trajectory of the spinal monitoring point and the trajectory of the concentrated point of pressure on the sole of the user in a walking cycle are analyzed to obtain the similarity between the trajectory of the concentrated point reflecting the pressure in the left and right feet respectively and the corresponding horizontal offset trajectory of the spinal monitoring point to the left and right directions. Then, processing is performed based on the two obtained similarities to obtain the coordination ability of the motion chain between the foot and spine in a walking cycle. Finally, the coordination ability of the foot-spine motion chain of the user in multiple walking cycles and the symmetry of the lower limbs of the two sides are analyzed to obtain a potential energy data chain.

[0040] The analysis model can also be used to analyze the multiple spine movement information and the multiple foot information to obtain a potential energy data chain, that is, the multiple spine movement information and the multiple foot information are input into the analysis model, and the analysis model outputs the corresponding potential energy data chain. The training process of the analysis model can be performed by using the data processed from the multiple spine movement information, the multiple foot information and the potential energy data chain as the training data set of the analysis model, and then inputting the training data set of the analysis model into the analysis model for training and learning, and finally obtaining the analysis model.

[0041] In a possible implementation, in step S300, the potential energy data chain is obtained by analyzing the spine movement information and the foot information, including: S310, performing dimension segmentation according to the spine movement information to obtain horizontal offset information; wherein the horizontal offset information is used to reflect the offset trajectory of the spine monitoring point in the horizontal direction.

[0042] It is understandable that during walking, the user first lifts the left or right foot to the ground, and then lifts the right or left foot to the ground to form a complete walk. In this process, the spine monitoring point is offset due to the lifting of the left and right feet and the change of the posture of the left and right feet.

[0043] Exemplarily, the ridge movement information in the three-dimensional space domain can be directly projected into the two-dimensional space domain in the direction of the user's walking bird's-eye view to eliminate the difference in the height dimension of the ridge movement information in the three-dimensional space domain, and finally obtain the horizontal offset information.

[0044] S320, analyzing the foot information to obtain a trajectory of a force point and a symmetry index; wherein the trajectory of the force point is used to reflect the trajectory of a concentrated point of pressure on the sole of the foot during a walking cycle; and the symmetry index is used to reflect the symmetry of the bilateral lower limbs during a walking cycle.

[0045] Exemplarily, the pressure distribution of different positions on the sole of the user's foot during the time when the sole of the user's foot contacts the ground can be obtained by processing a graph reflecting the shape of the sole of the user's foot and the sole pressure information, and then the pressure distribution of different positions on the sole of the user's foot during the time when the sole of the user contacts the ground can be analyzed to obtain a concentration point that can reflect the pressure on the sole of the user's foot during a walking cycle in a direction that increases in time series, and then the corresponding concentration points at multiple times are processed to obtain a force point trajectory. It is also possible to obtain multiple frame pressure data by processing the sole pressure information in frames, and then calculate the center point of the sole pressure (i.e., the force point) according to the multiple frame pressure data, and finally connect the continuous pressure center points to form a trajectory that changes with time, i.e., the force point trajectory. The center point of calculating the sole pressure can be calculated by the weighted average method according to the pressure value of each sensor and the position of the pressure value.

[0046] The time when the sole of the left foot is not under pressure and the time when the sole of the right foot is not under pressure can be directly obtained from the walking cycle in the foot information, and then the distance between the heel when the left foot lands and the heel when the next left foot lands and the distance between the heel when the right foot lands and the heel when the next right foot lands can be directly obtained from the step length information in the foot information, and then the time when the sole of the left foot is not under pressure and the distance between the heel when the left foot lands and the heel when the next left foot lands are processed to obtain a first ratio, and then the time when the sole of the right foot is not under pressure and the distance between the heel when the right foot lands and the heel when the next right foot lands are processed to obtain a second ratio, and finally the first ratio and the second ratio are analyzed to obtain a symmetry index. It is also possible to directly obtain the time when the left sole is under pressure and the time when the right sole is under pressure in the concentrated walking cycle in the foot information, and then directly obtain the distance between the heel when the left foot lands and the heel when the next left foot lands, and the distance between the heel when the right foot lands and the heel when the next right foot lands from the step length information in the foot information, and then process the time when the left sole is under pressure and the distance between the heel when the left foot lands and the heel when the next left foot lands to obtain a third ratio, and then process the time when the right sole is under pressure and the distance between the heel when the right foot lands and the heel when the next right foot lands to obtain a fourth ratio, and finally analyze the third ratio and the fourth ratio to obtain a symmetry index.

[0047] In a possible implementation, in step S320, analyzing the foot information to obtain a trajectory of the force points includes: S321, obtaining a sole graphic; wherein the sole graphic is used to reflect the shape of the sole of the user's foot.

[0048] It can be understood that the sole graphic refers to the graphic of the area on the sole of the user's foot that is under pressure when the user stands steadily on the ground.

[0049] Exemplarily, the plantar graphics can be manually input. It is also possible to obtain multiple contact points that generate pressure on the soles of the feet of the user when walking through wearable devices such as smart insoles, and then construct a graph of the multiple contact points to finally form a plantar graphics. The step of graph construction can be to first sort the multiple contact points by using a convex hull algorithm, and then connect the multiple contact points in sequence according to the sorting result to obtain a closed graph after the multiple contact points are connected, and then use interpolation or fitting methods (such as Bezier curves, B-spline curves, etc.) to convert the straight edges formed by the two-to-two connected points in the closed graph into smoother curves, and finally form a plantar graphics.

[0050] S322, assigning the plantar pressure information of the foot information to the corresponding position in the plantar graphic according to the position and analyzing and processing, to obtain a spatiotemporal pressure thermodynamic map; wherein the spatiotemporal pressure thermodynamic map is used to reflect the pressure distribution at different positions of the user's sole when the user's sole contacts the ground.

[0051] It can be understood that the spatiotemporal pressure heat map is a visualization tool that combines data in the time and space dimensions with plantar pressure information. It can display the pressure distribution in different areas of the sole of the foot at different time points through different colors.

[0052] Exemplarily, the foot pressure data collected at different times can be mapped to the plantar graphics through an image processing library (such as OpenCV or PIL) in a programming language (such as Python) to obtain static heat maps at different times, and then the static heat maps at different times can be combined into an animation through commonly used drawing libraries in Python (such as Matplotlib and Seaborn) to show the changes in foot pressure in the plantar graphics in the time domain, and finally a spatiotemporal pressure heat map is obtained.

[0053] S323, analyzing and processing the spatiotemporal pressure thermodynamic map in the direction of increasing pressure to obtain multiple force points and timestamps corresponding to the multiple force points; wherein the force point is used to reflect the concentrated point of pressure on the sole of the foot during a walking cycle; and the timestamp is used to reflect the specific time point of the force point.

[0054] It can be understood that the force point refers to the point on the sole of the user's foot that is subjected to the greatest pressure during each walking cycle of the user.

[0055] For example, the spatiotemporal pressure heat map can be analyzed according to the direction of increasing pressure on the sole of the foot to obtain the corresponding maximum pressure points at multiple time points, and then the maximum pressure points can be confirmed as the application points, and the time points corresponding to the multiple maximum pressure points can be confirmed as the timestamps corresponding to the multiple application points.

[0056] S324, sorting the multiple force points in ascending order based on the timestamps to obtain force point trajectories.

[0057] It can be understood that by arranging the force points with smaller timestamps in the front and the force points with larger timestamps in the back, a force point trajectory arranged in ascending time order is obtained. Because the force point refers to the point on the sole of the user's foot that bears the greatest pressure in each walking cycle, that is, the force point is a point in the sole graph, the position of the force point in the sole graph can be represented by a two-dimensional coordinate.

[0058] Exemplarily, if there is a set of force point data, wherein the set of force point data are (2, 3), (7, 8), (5, -3), (3, -4) and (6, 2), and the timestamp corresponding to the force point (2, 3) is 0.1s, the timestamp corresponding to the force point (3, -4) is 0.2s, the timestamp corresponding to the force point (5, -3) is 0.3s, the timestamp corresponding to the force point (6, 2) is 0.4s, and the timestamp corresponding to the force point (7, 8) is 0.5s, then the force point trajectory is sorted in the order of force point (2, 3) to force point (3, -4) to force point (5, -3) to force point (6, 2) to force point (7, 8), thereby obtaining the force point trajectory, and so on.

[0059] With this configuration, the migration law of the pressure center of the sole of the foot when the user walks can be reconstructed using the sole pressure information and timestamp, which can provide an evaluation index for the user's gait stability.

[0060] In a possible implementation, in step S320, analyzing the foot information to obtain a symmetry index includes: S325, obtaining the left air time and the right air time in the walking cycle of the foot information; wherein the left air time is used to reflect the time when the sole of the left foot is free of pressure, and the right air time is used to reflect the time when the sole of the right foot is free of pressure.

[0061] It can be understood that the walking cycle of a unilateral lower limb is divided into a support period and a hovering period. The support period refers to the period when the unilateral lower limb has not completely left the ground. The hovering period refers to the period when the unilateral lower limb has completely left the ground. When the unilateral lower limb has not completely left the ground, because the unilateral lower limb will also serve as a supporting part of the body to contact and support the ground, thereby maintaining body balance, the unilateral lower limb will still be under pressure when it has not completely left the ground.

[0062] For example, the time during the walking cycle during which the wearable device such as a smart insole does not detect pressure on the sole of the left foot can be used as the left air time. Similarly, the time during the walking cycle during which the wearable device such as a smart insole does not detect pressure on the sole of the right foot can be used as the right air time.

[0063] S326, obtaining the left stride length and the right stride length from the foot information; wherein the left stride length is used to reflect the distance between the heel of the left foot when it touches the ground and the heel of the left foot when it touches the ground again, and the right stride length is used to reflect the distance between the heel of the right foot when it touches the ground and the heel of the right foot when it touches the ground again.

[0064] For example, the sensor can be used to detect the horizontal movement distance of the left foot in each walking cycle, and then the horizontal movement distance of the left foot in each walking cycle can be confirmed as the left step length. Similarly, the horizontal movement distance of the right foot in each walking cycle can be confirmed as the right step length.

[0065] S327, performing a comparison process based on the left hovering time and the left step length to obtain a left comparison value; wherein the left comparison value is used to reflect the ratio between the left hovering time and the left step length; It can be understood that the left contrast value = left hovering time ÷ left step length.

[0066] For example, if the left dwell time is 0.64 s and the left step length is 9.5 cm, the left contrast value is 6.737, and so on.

[0067] S328, performing a comparison process according to the right hovering time and the right step length to obtain a right comparison value; wherein the right comparison value is used to reflect the ratio between the right hovering time and the right step length; It can be understood that right contrast value = right hovering time ÷ right step length.

[0068] For example, if the right dwell time is 0.62 s and the right step length is 10.2 cm, the right contrast value is 6.078, and so on.

[0069] S329, performing processing according to the left contrast value and the right contrast value to obtain a symmetry index.

[0070] It can be understood that the symmetry index = left contrast value ÷ right contrast value. A symmetry index close to 1 indicates that the left and right footsteps are relatively symmetrical, while a symmetry index far from 1 indicates asymmetry.

[0071] For example, if the left contrast value is 6.737 and the right contrast value is 6.078, the symmetry index is 1.108 (6.737÷6.078), and so on.

[0072] With this setting, the symmetry index is used to quantify the gait characteristics, and the gait symmetry of the user can be intuitively evaluated. An asymmetrical gait may be an early sign of an impending fall. By monitoring the symmetry index in real time, changes in the symmetry index before the user falls can be detected.

[0073] S330, analyzing the horizontal offset information and the trajectory of the force point to obtain a first coupling coefficient and a second coupling coefficient; wherein the first coupling coefficient is used to reflect the similarity between the offset trajectory of the spine monitoring point in the horizontal offset information in a walking cycle in the left direction and the trajectory of the force point of the left foot, and the second coupling coefficient is used to reflect the similarity between the offset trajectory in the right direction in the horizontal offset information in a walking cycle and the trajectory of the force point of the right foot.

[0074] It can be understood that during walking, the weight of the body is concentrated on the right side of the body as the left foot is lifted, which gradually increases the pressure on the right foot. And because the body needs to maintain balance, the spinal monitoring point also shifts to the right side of the body as the left foot is lifted. In this process, due to the connection relationship of the motion chain of the foot spine, the trajectory of the right foot's fulcrum is synchronized with the deviation trajectory of the spinal monitoring point in the right direction. Similarly, the trajectory of the left foot's fulcrum is also synchronized with the deviation trajectory of the spinal monitoring point in the left direction.

[0075] Exemplarily, the horizontal offset information can be processed to obtain the point offset trajectory of the horizontal offset information in the left direction and the offset trajectory of the horizontal offset information in the right direction, respectively. Then, the first coupling coefficient is obtained by processing the trajectory of the right foot's force point and the offset trajectory of the horizontal offset information in the right direction. Then, the second coupling coefficient is obtained by processing the trajectory of the left foot's force point and the offset trajectory of the horizontal offset information in the left direction.

[0076] It is also possible to process the horizontal offset information to obtain multiple first curvatures that can reflect the curvature of each trajectory point in the horizontal offset information, and then process it according to the force point trajectory to obtain multiple second curvatures that can reflect each trajectory point in the force point trajectory of the left foot and a third curvature that can reflect each trajectory point in the force point trajectory of the right foot, and then perform time domain comparative analysis based on the multiple first curvatures and the multiple second curvatures, compare the first curvature and the second curvature corresponding to the same time point to obtain multiple first ratios between the corresponding first curvatures and the second curvatures, and then process it according to the multiple first ratios to obtain a first probability density, and confirm the first probability density as the first coupling coefficient. Similarly, the second coupling coefficient is obtained by processing the multiple first curvatures and the multiple third curvatures.

[0077] In a possible implementation, in step S330, the first coupling coefficient and the second coupling coefficient are obtained by analyzing the horizontal offset information and the trajectory of the focus point, including: S331, processing is performed according to the horizontal offset information to obtain a left offset trajectory and a right offset trajectory; wherein the left offset trajectory is used to reflect the offset trajectory of the spine monitoring point in the horizontal offset information in the left direction, and the right offset trajectory is used to reflect the offset trajectory in the right direction in the horizontal offset information.

[0078] For example, the user's forward direction may be identified as a baseline, and then the horizontal offset information may be separated by the baseline, and the horizontal offset information on the left side of the baseline may be identified as a left offset track, and the horizontal offset information on the right side of the baseline may be identified as a right offset track.

[0079] S332, processing the right foot trajectory according to the right offset trajectory and the force point trajectory to obtain a first coupling coefficient.

[0080] Exemplarily, time domain segmentation can be performed through right offset information to obtain multiple first offset points, time domain segmentation can be performed on the right foot trajectory of the impact point trajectory to obtain multiple second offset points, and the first displacement angle and the first displacement amount between each two adjacent first offset points can be obtained by comparing the adjacent points of the multiple first offset points, and then the second displacement angle and the displacement amount between each two adjacent second offset points can be obtained by comparing the leading points of the multiple second offset points, and then multiple displacement angle ratios can be obtained by comparing the multiple first displacement angles with the second displacement angles, and then multiple displacement ratios can be obtained by comparing the multiple first displacement amounts with the second displacement amounts, and then the first average value between the multiple displacement angle ratios and the second average value between the multiple displacement amounts are calculated respectively, and finally the third average value between the first average value between the multiple displacement angle ratios and the second average value between the multiple displacement amounts is calculated and the third average value finally obtained is confirmed as the first coupling coefficient. Wherein, the displacement angle refers to the angle formed between two adjacent offset points. The displacement amount refers to the distance between two adjacent offset points.

[0081] S333, processing the left foot trajectory according to the left offset trajectory and the force point trajectory to obtain a second coupling coefficient.

[0082] Exemplarily, the second coupling coefficient may be obtained in a manner similar to the first coupling coefficient obtained in step S332 , which will not be described in detail herein.

[0083] With such a setting, the first coupling coefficient and the second coupling coefficient are obtained by analyzing the horizontal offset information and the trajectory of the point of application, which can provide basic data for the subsequent analysis of the potential energy data when the user falls.

[0084] S340, performing sum processing on the first coupling coefficient and the second coupling coefficient to obtain a synchronization coefficient; wherein the synchronization coefficient is used to reflect the coordination ability of the foot spine motion chain in a walking cycle.

[0085] It can be understood that the synchronization coefficient=first coupling coefficient+second coupling coefficient.

[0086] For example, if the first coupling coefficient is 0.42 and the second coupling coefficient is 0.38, the synchronization coefficient is 0.8 (0.42+0.38), and so on.

[0087] S350, analyzing the synchronization coefficients within the multiple walking cycles and the symmetry indexes within the multiple walking cycles to obtain a potential energy data link.

[0088] For example, the potential energy data may be obtained by processing the synchronization coefficient and the symmetry index, and then the potential energy data in a plurality of walking cycles may be arranged in increasing order of the walking cycles to finally obtain a potential energy data chain.

[0089] It is also possible to construct data for multiple synchronization coefficients and multiple symmetry indexes through a data construction model to obtain a potential energy data chain, that is, input multiple synchronization coefficients and multiple symmetry indexes into the data construction model, and the data construction model then outputs the corresponding potential energy data chain. The training process of the data construction model can be achieved by using the data processed from multiple synchronization coefficients, multiple symmetry indexes and potential energy data chains as a training data set for the data construction model, and then inputting the training data set of the data construction model into the data construction model for training and learning, and finally obtaining the data construction model.

[0090] With this setting, by monitoring the horizontal deviation trajectory of the spine (horizontal deviation information), the individual's dynamic balance ability during walking can be evaluated. The first coupling coefficient and the second coupling coefficient respectively measure the similarity between the left and right deviations of the spine and the corresponding foot force point trajectories, which can help to gain a deeper understanding of the working mechanism of the foot-spine motion chain. By collecting data from multiple walking cycles and calculating the potential energy data chain, the changing trend over a period of time can be tracked, thereby making behavioral predictions on users' falls.

[0091] In a possible implementation, in step S350, a potential energy data chain is obtained by analyzing synchronization coefficients within a plurality of walking cycles and symmetry indexes within a plurality of walking cycles, including: S351, processing is performed according to the synchronization coefficient and the symmetry index to obtain potential energy data; wherein the potential energy data is used to reflect the overall stability of the human body's dynamic balance system.

[0092] It can be understood that different dynamic balance of the human body corresponds to a potential energy data. The lower the dynamic balance of the human body, the higher the potential energy data. The synchronization coefficient can quantify the coordination of foot and spine movement. The symmetry index can reflect the left and right differences when the user walks.

[0093] For example, the ridge potential energy can be obtained by processing the synchronization coefficient, and the symmetry index can be processed to obtain the symmetrical potential energy, and then the potential energy data can be finally obtained by calculating the sum of the ridge potential energy and the symmetrical potential energy.

[0094] The synchronization coefficient and the symmetry index can also be analyzed through the potential energy model to obtain potential energy data, that is, the synchronization coefficient and the symmetry index are input into the potential energy model, and the potential energy model outputs the corresponding potential energy data. The training process of the potential energy model can be performed by processing the synchronization coefficient and the symmetry index with the corresponding potential energy data as the training data set of the potential energy model, and then inputting the training data set of the potential energy model into the potential energy model for training and learning, and finally obtaining the potential energy model.

[0095] In a possible implementation, in step S351, the potential energy data is obtained by processing the synchronization coefficient and the symmetry index, including: S3511, processing is performed according to the synchronization coefficient to obtain the potential energy of the foot spine.

[0096] It can be understood that the calculation formula of the potential energy of the foot spine can be: Among them, E 1 is the potential energy of the foot spine, k is the age coefficient, and D is the synchronization coefficient.

[0097] S3512, processing is performed according to the symmetry index to obtain the symmetry potential energy.

[0098] It can be understood that the calculation formula of symmetric potential energy can be: , where E 2 is the symmetry potential energy and N is the symmetry index.

[0099] S3513, performing sum calculation based on the synchronization coefficient and the symmetry index to obtain potential energy data.

[0100] It can be understood that potential energy data = foot spine potential energy + symmetric potential energy.

[0101] S352, sorting the potential energy data corresponding to the multiple walking cycles according to the walking cycle sequence to obtain a potential energy data chain.

[0102] It can be understood that the potential energy data calculated in each walking cycle is sorted according to the time series, and the potential energy data of each walking cycle can be organized into a potential energy data chain by sorting according to the time series. Because the walking cycle sequence is sorted by the user according to the increasing number of walking cycles, it can be equivalent to the time series.

[0103] With this setting, potential energy data is obtained through joint analysis of synchronization coefficient and symmetry index, and multiple potential energy data are arranged into potential energy data chain according to walking cycle sequence, which can be used to analyze each walking and historical walking of the user, so as to more comprehensively analyze the changes in the stability of the user's body when walking. It can provide basic prediction data for the user's fall event.

[0104] S400, analyzing historical fall events to obtain a fall threshold; wherein the fall threshold is used to reflect potential energy data when the user falls.

[0105] It can be understood that a historical fall event refers to a fall that occurred to the user in the past.

[0106] Exemplarily, multiple potential energy data of historical fall events of users can be obtained, and then the multiple potential energy data can be averaged to finally obtain the average potential energy and confirm the average potential energy as the fall threshold. The fall threshold can also be directly obtained through the potential energy database. The potential energy database refers to a database containing fall potential energy data corresponding to different user ages. These data can be obtained through laboratory experiments, on-site measurements and monitoring, and past experience. After obtaining, the collected data will be sorted, classified and archived, useful information and rules will be extracted, and the relevant data will be saved in the database to form a potential energy database.

[0107] In a possible implementation, in step S400, analyzing historical fall events to obtain a fall threshold includes: S410, obtaining a plurality of fall potential energy data of historical fall events; wherein the fall potential energy data is used to reflect the potential energy data when the user falls.

[0108] It can be understood that historical fall events refer to falls that occurred to the user before fall risk prediction was performed.

[0109] Exemplarily, multiple fall potential energy data may be manually inputted, or multiple fall thresholds may be directly acquired through a potential energy database.

[0110] S420, analyzing the plurality of fall potential energy data to obtain average potential energy data, and confirming the average potential energy data as a fall threshold.

[0111] It can be understood that the potential energy average data refers to the data obtained by averaging multiple fall potential energy data.

[0112] For example, if there are three fall events in the historical fall events, and the fall potential energy data corresponding to the first fall event is 52, the fall potential energy data corresponding to the second fall event is 51, and the fall potential energy data corresponding to the third fall event is 53, then the average potential energy data is 52 [(52+51+53) ÷ 3], and so on.

[0113] Such a setting can provide effective parameters for fall prediction by analyzing the potential energy data of historical fall events to establish the fall threshold.

[0114] S500, analyzing the fall threshold and the potential energy data chain to obtain a predicted fall cycle; wherein the predicted fall cycle is used to reflect the time from the start of the user's latest walking cycle to the time when the potential energy data reflected by the potential energy data chain is equal to the fall threshold.

[0115] Exemplarily, the potential energy data chain can be analyzed to obtain potential energy data reflecting the user's most recent completion of a walking cycle, and then the potential energy data chain can be processed to obtain multiple i-th data points indicating that the potential energy data chain meets the screening conditions, and then the multiple i-th data points are processed to obtain a curve reflecting the multiple i-th data points, and then the curve and the multiple i-th data points are processed to obtain a curve reflecting the degree of the rising trend of the potential energy data with the walking cycle, and finally the potential energy data of the user's most recent completion of a walking cycle, the fall threshold and the degree of the rising trend of the potential energy value with the walking cycle are analyzed to finally obtain the predicted fall cycle.

[0116] The potential energy data chain can also be analyzed to obtain the decline cycle in the potential energy data chain and the potential energy data of the user's latest walking cycle, and then the potential energy data of the user's latest walking cycle, the decline cycle and the fall threshold are analyzed and processed to finally obtain the predicted fall cycle.

[0117] In a possible implementation, in step S500, the predicted fall cycle is obtained by analyzing the fall threshold and the potential energy data chain, including: S510, analyzing the potential energy data chain to obtain the final potential energy; wherein the final potential energy is used to reflect the potential energy data of the user's latest completed walking cycle.

[0118] It can be understood that the last potential energy refers to the last potential energy data in the potential energy data chain.

[0119] For example, if a potential energy data chain is 21-25-23-22-31, the last potential energy is 31, and so on.

[0120] S520, screening and processing are performed according to the potential energy data chain to obtain multiple reduced data; wherein the reduced data is used to indicate that the i-th potential energy data in the potential energy data chain is less than the i-1-th potential energy data, and the i+1-th potential energy data is less than the i-1-th potential energy data.

[0121] It can be understood that the screening condition for the potential energy data chain is that the i-th potential energy data is less than the i-1-th potential energy data, and the i+1-th potential energy data is less than the i-1-th potential energy data. The multiple reduced data after screening are the i-th potential energy data in the potential energy data chain.

[0122] For example, if a potential energy data chain is 21-25-23-22-21-17-22-15-12-13-14, the reduced data includes 23, 22, 21, 15, 12, and so on.

[0123] S530, performing processing according to the plurality of reduction data to obtain a fall risk curve; wherein the fall risk curve is used to reflect a curve formed by the plurality of reduction data.

[0124] Exemplarily, a fall risk curve can be obtained by constructing and drawing multiple reduction data using software tools such as mathematical software (such as MATLAB, Mathematica), programming languages ​​(such as Python's Matplotlib, NumPy, SciPy libraries) or drawing tools.

[0125] S540, obtaining a fall trend according to the fall risk curve and the plurality of reduction data; wherein the fall trend is used to reflect the rising trend of the potential energy data in a walking cycle.

[0126] Exemplarily, the fall risk curve and multiple reduction data can be analyzed and processed to obtain multiple tangent slopes indicating the fall risk curve corresponding to the reduction data, and then the multiple tangent slopes are processed to obtain an average value between the multiple tangent slopes, and finally the average value is confirmed as the fall trend.

[0127] It is also possible to process multiple reduction data to obtain a reduction data range, and then process the fall risk curve according to the reduction data range to obtain multiple curve change rates of the fall risk curve within the reduction data range, and then perform mean processing on the multiple curve change rates to obtain an average change rate, and finally confirm the average change rate as a fall trend. The reduction data range refers to the potential energy data range composed of the maximum data and the minimum data among the multiple reduction data.

[0128] In a possible implementation, in step S540, the fall risk curve and the plurality of reduction data are processed to obtain a fall trend, including: S541, processing the fall risk curve and the plurality of reduction data to obtain a plurality of change rates; wherein the change rate is used to indicate a tangent slope corresponding to the reduction data in the fall risk curve.

[0129] Exemplarily, a plurality of points corresponding to the reduction data may be found in the fall risk curve, and the tangent slope corresponding to the points in the fall risk curve at the reduction data may be obtained, and finally the tangent slope may be confirmed as the rate of change.

[0130] S542, performing processing according to the multiple change rates to obtain an average change rate, and confirming the average change rate as a falling trend; wherein the average change rate is used to reflect the average value between the multiple change rates.

[0131] Exemplarily, if there are five change rates, the first change rate is 3, the first change rate is 5, the first change rate is 7, the first change rate is 2, and the first change rate is 8, then the falling trend is 5 [(3+5+7+2+8) ÷ 5], and so on.

[0132] With this setting, by calculating the tangent slope of the fall risk curve corresponding to the reduction data, the degree of change of the fall risk with the walking cycle can be quantified, thereby enabling accurate prediction of fall events.

[0133] S550, processing is performed according to the final potential energy, the fall threshold and the fall trend to obtain a predicted fall cycle.

[0134] Understandably, ,in, , H is the fall threshold, M is the final potential energy, K is the fall trend, Returns the smallest integer greater than or equal to X.

[0135] For example, if the fall threshold is 52, the final potential is 21, and the fall trend is 5, then the predicted fall cycle is 7. ,like In the example, if X is 6.2, the predicted fall cycle is 7. If X is 6, the predicted fall cycle is 6, and so on.

[0136] With this setting, by continuously analyzing the potential energy data chain and obtaining the terminal potential energy in real time, the potential energy state of the user's current walking cycle can be instantly evaluated to provide data support for real-time monitoring. Then, by screening out the decreasing data that can indicate the decreasing point of the potential energy data, and then constructing a curve of the decreasing data to obtain a fall risk curve, the changing trend of the fall risk can be intuitively displayed, and data can be provided to facilitate the prediction of future fall risks.

[0137] S600: The control device controls the prediction and reminder device to issue a fall risk warning based on the predicted fall cycle.

[0138] For example, a reminder timing can be set in advance, and then the reminder timing and the predicted fall cycle are analyzed to obtain the timing when the control device controls the prediction reminder device to issue a fall risk warning, so as to prevent the user from falling in advance when the fall cycle is predicted. It is also possible to obtain the time required for the user to complete a walking cycle, and then analyze the time and the predicted fall cycle to obtain the remaining time when the user may fall, and then analyze the remaining time to finally obtain the reminder timing of the prediction reminder device.

[0139] With such a setting, the potential energy data chain is obtained by analyzing the real-time acquired spine movement information and the real-time acquired foot information, and then the fall threshold is obtained by analyzing the historical fall events. Then, the predicted fall cycle is obtained by analyzing the fall threshold and the potential energy data chain. It can specifically predict the situation in which the potential energy data in the subsequent walking cycle of each infant falls below the fall threshold, and issue an alarm in advance, so that when the infant has not started to fall, the parents are promptly reminded to provide support in advance to prevent the infant from falling.

[0140] In a possible implementation, in step S600, the control device controls the prediction reminder device to issue a fall risk warning based on the predicted fall cycle, including: S610, obtaining a preset reminder period; wherein the preset reminder period is used to reflect a preset reminder timing.

[0141] It can be understood that the preset reminder period can be used as an opportunity for early warning to warn the user of a fall.

[0142] Exemplarily, the preset reminder cycle can be manually input. The preset reminder cycle can also be directly obtained through the reminder cycle database. The reminder cycle database refers to a database containing reminder cycles corresponding to different user ages. These data can be obtained through laboratory experiments, on-site measurements and monitoring, and past experience. After obtaining, the collected data is sorted, classified and archived, useful information and rules are extracted, and the relevant data is saved in the database to form a reminder cycle database.

[0143] S620: When the predicted fall cycle is equal to the preset reminder cycle, the control device controls the prediction reminder device to issue a fall risk warning.

[0144] It can be understood that the predicted fall cycle is to predict the fall event of the user's subsequent walking cycle through the prediction reminder device. Because when the user walks, the walking cycle gradually increases, the terminal potential energy will also be updated accordingly, and the potential energy data chain will gradually add new potential energy data to increase the length of the data chain, and finally the predicted fall cycle will also change with the gradual increase of the walking cycle. When the predicted fall cycle is equal to the preset reminder cycle, it means that the user is more likely to fall after the predicted fall cycle, and the control device controls the prediction reminder device to issue a fall risk warning.

[0145] Exemplarily, if the preset reminder cycle is 3 walking cycles, when the predicted fall cycle analyzed by the prediction reminder device is 3 walking cycles, the prediction reminder device issues a fall risk warning; when the predicted fall cycle analyzed by the prediction reminder device is 2 walking cycles, the prediction reminder device issues a fall risk warning.

[0146] S630: If the predicted fall cycle is greater than the preset reminder cycle, the control device controls the prediction reminder device not to issue a fall risk warning.

[0147] For example, if the preset reminder cycle is 3 walking cycles, when the predicted fall cycle obtained by the prediction reminder device through analysis is 5 walking cycles, the prediction reminder device does not issue a fall risk warning.

[0148] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0149] Corresponding to the health risk prediction method based on behavioral data in the above embodiment, the embodiment of the present application also provides a health risk prediction system based on behavioral data, and the various modules of the health risk prediction system based on behavioral data can implement the various steps of the health risk prediction method based on behavioral data. Figure 3 A structural block diagram of a health risk prediction system based on behavioral data provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0150] Reference Figure 3 , the health risk prediction system based on behavioral data includes: The first acquisition module is used to acquire foot information in real time during each walking cycle; wherein the foot information includes plantar pressure information and gait information, the plantar pressure information is used to reflect the distribution of plantar pressure of the user, the gait information includes step length and walking cycle, the step length is used to reflect the distance between the heel of one foot when the user touches the ground and the heel of the same foot when the same foot touches the ground next time, and the walking cycle is used to indicate the time from one foot of the user leaving the ground to the next time the other foot completely touches the ground after leaving the ground.

[0151] The second acquisition module is used to acquire the spine movement information in real time during each walking cycle; wherein the spine movement information is used to reflect the movement trajectory of the spine monitoring point.

[0152] The first analysis module is used to analyze multiple spine movement information and multiple foot information to obtain a potential energy data chain; wherein the potential energy data chain is used to reflect a data chain composed of potential energy data corresponding to each walking cycle of the user; the potential energy data is used to reflect the energy required to maintain a balanced body balance.

[0153] The second analysis module is used to analyze historical fall events to obtain a fall threshold; wherein the fall threshold is used to reflect the potential energy data when the user falls.

[0154] The third analysis module is used to analyze according to the fall threshold and the potential energy data chain to obtain a predicted fall cycle; wherein the predicted fall cycle is the time elapsed from the start of the latest walking cycle of the user to the time when the potential energy data reflected by the potential energy data chain is equal to the fall threshold.

[0155] The early warning module is used to control the device to control the prediction reminder device to issue a fall risk early warning based on the predicted fall cycle.

[0156] It should be noted that the information interaction, execution process, etc. between the above-mentioned systems / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0157] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0158] An embodiment of the present application also provides a health risk prediction device based on behavioral data. The health risk prediction device based on behavioral data includes a prediction reminder device and a control device. The prediction reminder device is electrically connected to the control device. Figure 4 This is a schematic diagram of the structure of a control device 4 provided in an embodiment of the present application. Figure 4 As shown, the control device 4 of this embodiment includes: at least one processor 40 ( Figure 4 Only one is shown), at least one memory 41 ( Figure 4 Only one is shown) and a computer program 42 stored in the at least one memory 41 and executable on the at least one processor 40. When the processor 40 executes the computer program 42, the control device 4 implements the steps in any of the above-mentioned health risk prediction method embodiments based on behavioral data, or implements the functions of the modules / units in the above-mentioned system embodiments.

[0159] Exemplarily, the computer program 42 may be divided into one or more modules / units, which are stored in the memory 41 and executed by the processor 40 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 42 in the control device 4.

[0160] The control device 4 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The control device 4 may include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will appreciate that Figure 4It is only an example of the control device 4 and does not constitute a limitation on the control device 4. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, buses, etc.

[0161] The processor 40 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0162] In some embodiments, the memory 41 may be an internal storage unit of the control device 4, such as a hard disk or memory of the control device 4. In other embodiments, the memory 41 may also be an external storage device of the control device 4, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the control device 4. Further, the memory 41 may also include both an internal storage unit and an external storage device of the control device 4. The memory 41 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program. The memory 41 may also be used to temporarily store data that has been output or is to be output.

[0163] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0164] An embodiment of the present application provides a computer program product. When the computer program product runs on a health risk prediction device based on behavior data, the health risk prediction device based on behavior data implements the steps in any of the above-mentioned method embodiments.

[0165] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. According to this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to a health risk prediction device based on behavioral data, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electrical carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk.

[0166] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0167] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0168] In the embodiments provided in the present application, it should be understood that the disclosed health risk prediction system and device based on behavioral data can be implemented in other ways. For example, the above-described embodiment of the health risk prediction system based on behavioral data is merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0169] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0170] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A health risk prediction method based on behavioral data, characterized in that: include: Acquire foot information in each walking cycle in real time; wherein the foot information includes plantar pressure information and gait information, the plantar pressure information is used to reflect the distribution of plantar pressure of the user, the gait information includes step length and walking cycle, the step length is used to reflect the distance between the heel of one foot when the user touches the ground and the heel of the same foot when the same foot touches the ground next time, and the walking cycle is used to indicate the time from one foot of the user leaving the ground to the next time the other foot completely touches the ground after leaving the ground; Acquire the spinal movement information in each walking cycle in real time; wherein the spinal movement information is used to reflect the movement trajectory of the spinal monitoring point; A potential energy data chain is obtained by analyzing the plurality of spine movement information and the plurality of foot information; wherein the potential energy data chain is used to reflect a data chain composed of potential energy data corresponding to each walking cycle of the user, and the potential energy data is used to reflect the energy required to maintain the balance of the body; Analyze historical fall events to obtain a fall threshold; wherein the fall threshold is used to reflect the potential energy data when the user falls; The predicted fall cycle is obtained by analyzing the fall threshold and the potential energy data chain; wherein the predicted fall cycle is used to reflect the time from the start of the latest walking cycle of the user to the time when the potential energy data reflected by the potential energy data chain is equal to the fall threshold; The control device controls the prediction and reminder device to issue a fall risk warning based on the predicted fall cycle.

2. The health risk prediction method based on behavioral data according to claim 1, characterized in that: The step of analyzing the spine movement information and the foot information to obtain a potential energy data chain includes: Performing dimension segmentation according to the spine movement information to obtain horizontal offset information; wherein the horizontal offset information is used to reflect the offset trajectory of the spine monitoring point in the horizontal direction; Analyze the foot information to obtain a trajectory of a force point and a symmetry index; wherein the trajectory of the force point is used to reflect the trajectory of a concentrated point of pressure on the sole of the foot during a walking cycle; and the symmetry index is used to reflect the symmetry of the bilateral lower limbs during a walking cycle; A first coupling coefficient and a second coupling coefficient are obtained by analyzing the horizontal offset information and the trajectory of the force point; wherein the first coupling coefficient is used to reflect the similarity between the offset trajectory of the spine monitoring point in the horizontal offset information in the walking cycle and the trajectory of the force point of the left foot, and the second coupling coefficient is used to reflect the similarity between the offset trajectory in the right direction of the horizontal offset information in the walking cycle and the trajectory of the force point of the right foot; A synchronization coefficient is obtained by performing sum processing on the first coupling coefficient and the second coupling coefficient; wherein the synchronization coefficient is used to reflect the coordination ability of the foot spine motion chain in one walking cycle; A potential energy data link is obtained by analyzing the synchronization coefficients in a plurality of walking cycles and the symmetry indexes in a plurality of walking cycles.

3. The health risk prediction method based on behavioral data according to claim 2, characterized in that: The analyzing the foot information to obtain the trajectory of the force point includes: Acquire a sole image; wherein the sole image is used to reflect the shape of the sole of the user's foot; Assigning the plantar pressure information of the foot information to the corresponding position in the plantar graph according to the position and performing analysis and processing to obtain a spatiotemporal pressure thermodynamic map; wherein the spatiotemporal pressure thermodynamic map is used to reflect the pressure distribution at different positions of the user's sole when the user's sole contacts the ground; According to the time-space pressure thermodynamic map, the pressure is analyzed and processed in the direction of increasing pressure to obtain multiple force points and timestamps corresponding to the multiple force points; wherein the force point is used to reflect the concentrated point of pressure on the sole of the foot during a walking cycle; and the timestamp is used to reflect the specific time point of the force point; The plurality of force points are sorted in ascending order based on the timestamps to obtain a force point trajectory.

4. The health risk prediction method based on behavioral data according to claim 2, characterized in that: The analyzing according to the foot information to obtain a symmetry index includes: Acquire a left air time and a right air time in the walking cycle of the foot information; wherein the left air time is used to reflect the time when the sole of the left foot is free of pressure, and the right air time is used to reflect the time when the sole of the right foot is free of pressure; Obtaining a left step length and a right step length from the foot information; wherein the left step length is used to reflect the distance between the heel of the left foot when it lands and the heel of the left foot when it lands again, and the right step length is used to reflect the distance between the heel of the right foot when it lands and the heel of the right foot when it lands again; Comparing the left dwell time with the left step length, to obtain a left comparison value; wherein the left comparison value is used to reflect the ratio between the left dwell time and the left step length; Comparing the right hovering time with the right step length, to obtain a right comparison value; wherein the right comparison value is used to reflect the ratio between the right hovering time and the right step length; A symmetry index is obtained by processing the left contrast value and the right contrast value.

5. The method for predicting health risks based on behavioral data according to claim 2, characterized in that: The step of analyzing the horizontal offset information and the trajectory of the focus point to obtain a first coupling coefficient and a second coupling coefficient includes: Processing is performed according to the horizontal offset information to obtain a left offset trajectory and a right offset trajectory; wherein the left offset trajectory is used to reflect the offset trajectory of the spine monitoring point in the horizontal offset information in the left direction, and the right offset trajectory is used to reflect the offset trajectory in the right direction in the horizontal offset information; Processing the right foot trajectory of the right offset trajectory and the force point trajectory to obtain a first coupling coefficient; The second coupling coefficient is obtained by processing the left foot trajectory of the left offset trajectory and the force point trajectory.

6. The method for predicting health risks based on behavioral data according to claim 2, characterized in that: The step of analyzing the plurality of synchronization coefficients and the plurality of symmetry indexes to obtain a potential energy data chain includes: Processing is performed according to the synchronization coefficient and the symmetry index to obtain potential energy data; wherein the potential energy data is used to reflect the overall stability of the human body's dynamic balance system; The potential energy data corresponding to the plurality of walking cycles are sorted according to the walking cycle sequence to obtain a potential energy data chain.

7. The method for predicting health risks based on behavioral data according to claim 1, characterized in that: The analysis based on historical fall events to obtain the fall threshold includes: Acquire multiple fall potential energy data of historical fall events; wherein the fall potential energy data is used to reflect the potential energy data when the user falls; An analysis is performed based on the plurality of fall potential energy data to obtain average potential energy data, and the average potential energy data is confirmed as a fall threshold.

8. The method for predicting health risks based on behavioral data according to claim 1, characterized in that: The step of analyzing the fall threshold and the potential energy data link to obtain a predicted fall cycle includes: Analyze the potential energy data chain to obtain the final potential energy; wherein the final potential energy is used to reflect the potential energy data of the user's latest completed walking cycle; Performing screening processing according to the potential energy data chain to obtain a plurality of reduced data; wherein the reduced data is used to indicate that the i-th potential energy data in the potential energy data chain is less than the i-1-th potential energy data, and the i+1-th potential energy data is less than the i-th data point of the i-1-th potential energy data; Processing is performed according to the plurality of the reduction data to obtain a fall risk curve; wherein the fall risk curve is used to reflect a curve formed by the plurality of the reduction data; Processing the fall risk curve and the plurality of reduction data to obtain a fall trend; wherein the fall trend is used to reflect the rising trend of the potential energy data in one walking cycle; The predicted fall cycle is obtained by processing the final potential energy, the fall threshold and the fall trend.

9. The method for predicting health risks based on behavioral data according to claim 8, characterized in that: The step of processing the fall risk curve and the plurality of reduction data to obtain a fall trend includes: Processing the fall risk curve and the plurality of reduction data to obtain a plurality of change rates; wherein the change rates are used to indicate the slope of a tangent line in the fall risk curve corresponding to the reduction data; Processing is performed based on the multiple change rates to obtain an average change rate, and the average change rate is confirmed as a falling trend; wherein the average change rate is used to reflect the average value between the multiple change rates.

10. A health risk prediction device based on behavioral data, characterized in that: It includes a prediction reminder device and a control device, the prediction reminder device is electrically connected to the control device, the control device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method according to any one of claims 1 to 9 when executing the computer program.

Images