An old people fall prevention system based on gait assessment

Abstract

The present application relates to the technical field of pressure detection, and provides an old person fall prevention system based on gait evaluation, comprising a data processor, which is used for executing the following processing strategy: detecting the plantar pressure of an old person during walking, and obtaining the gait instability degree of each single foot support period of the old person according to the change of the plantar pressure; obtaining the imbalance and loss of control degree of the old person according to the gait instability degree, and the plantar pressure difference and time interval difference between adjacent single foot support periods; determining the correlation between the gait instability degree and the change of walking speed of the old person during walking, and obtaining the sluggishness degree of the old person in combination with the change trend of the gait instability degree; and fusing the imbalance and loss of control degree and the sluggishness degree to obtain the fall risk value of the old person, thereby improving the accuracy of the fall risk value acquisition of the old person, and further improving the accuracy and reliability of the fall risk early warning of the old person.

Description

Technical Field

[0001] This invention relates to the field of pressure detection technology, and more specifically to a fall prevention system for the elderly based on gait assessment. Background Technology

[0002] Currently, when issuing early warnings about the risk of falls among the elderly, a fall risk value is typically obtained based on the difference in the elderly person's step length over a period of time, and then an early warning is issued based on this value. However, the data used in obtaining the fall risk value is relatively limited, affecting the accuracy and reliability of the value, and thus impacting the accuracy of the early warning system. Summary of the Invention

[0003] To address the problem of low accuracy in obtaining fall risk values ​​for the elderly in existing technologies, the present invention aims to provide a fall prevention system for the elderly based on gait assessment. The specific technical solution adopted is as follows: This invention provides a fall prevention system for the elderly based on gait assessment, including a data processor, which is used to execute the following processing strategies: The plantar pressure of the elderly was detected during walking, and the degree of gait instability of the elderly in each single-leg support phase was obtained based on the changes in plantar pressure. The degree of gait instability, as well as the differences in plantar pressure and time intervals between adjacent single-leg support phases, is used to determine the degree of imbalance and loss of control in the elderly. Determine the degree of gait instability and changes in walking speed in elderly individuals during walking, and combine this with the trend of gait instability to determine the degree of slowed reaction in the elderly individuals; By combining the degree of imbalance and loss of control with the degree of slow reaction, a fall risk value for the elderly is obtained.

[0004] In an exemplary embodiment, the process of obtaining the degree of gait instability includes: Determine the pressure difference between each area of ​​the sole at every two moments during the single-leg support phase to obtain the degree of unevenness in the sole pressure distribution in each area. The degree of gait instability is obtained by considering the difference in maximum pressure between two regions during the single-leg support phase, combined with the degree of unevenness in plantar pressure distribution in each region.

[0005] In an exemplary embodiment, the step of determining the gait instability based on the difference between the maximum pressures of two regions during the single-leg support phase, combined with the degree of unevenness in plantar pressure distribution in each region, includes: Calculate the average difference between the maximum pressures of all two regions during the single-leg support phase to obtain the overall difference in maximum pressures during the single-leg support phase; Calculate the average value of the unevenness of plantar pressure distribution in each region to obtain the overall unevenness of plantar pressure distribution during the single-foot support phase; The degree of gait instability is obtained by integrating the overall difference in maximum pressure and the overall unevenness of plantar pressure distribution; the degree of gait instability is positively correlated with both the overall difference in maximum pressure and the overall unevenness of plantar pressure distribution.

[0006] In one exemplary embodiment, the process of obtaining the degree of imbalance and loss of control includes: By integrating the differences in plantar pressure between adjacent single-leg support phases, the overall difference in plantar pressure is obtained; Based on the overall differences in plantar pressure and the overall degree of gait instability, the degree of center of gravity deviation during the elderly person's walking process is obtained; The overall difference in time intervals is obtained by considering the difference between each two adjacent time intervals during the walking process; the time interval is the time interval between two adjacent single-leg support periods. The degree of imbalance and loss of control is obtained based on the overall difference between the degree of center of gravity deviation and the time interval; the degree of imbalance and loss of control is positively correlated with both the degree of center of gravity deviation and the overall difference in time interval.

[0007] In one exemplary embodiment, the overall gait instability level is the average of the gait instability levels.

[0008] In an exemplary embodiment, determining the degree of gait instability and changes in walking speed of the elderly person during walking includes: A curve fitting was performed on the gait instability sequence to obtain the first fitting curve; the gait instability sequence was obtained by arranging the various gait instabilities of the elderly in chronological order during the walking process. A second fitted curve is obtained by curve fitting the walking acceleration sequence; the walking acceleration sequence is obtained by arranging the elderly person's walking accelerations in chronological order during the walking process. Obtain the correlation coefficient between the first fitted curve and the second fitted curve, where the correlation is the correlation coefficient.

[0009] In an exemplary embodiment, the process of obtaining the trend of change in gait instability includes: Obtain the first-order difference sequence of the gait instability sequence; The average value of the data in the first-order difference sequence is obtained to obtain the trend of the change in gait instability; the trend of change is positively correlated with the average value of the data.

[0010] In one exemplary embodiment, the process of obtaining the degree of sluggishness includes: The degree of sluggishness is obtained based on the correlation coefficient and the trend of change; the degree of sluggishness is positively correlated with both the correlation coefficient and the trend of change.

[0011] In one exemplary embodiment, the Euclidean norm of the degree of imbalance and loss of control and the degree of sluggishness of reaction are calculated as the fall risk value of the elderly person.

[0012] In one exemplary embodiment, the processing strategy further includes: The elderly person's fall risk value is compared with a preset elderly person's fall risk threshold. If the elderly person's fall risk value is greater than or equal to the preset elderly person's fall risk threshold, a fall warning signal is output.

[0013] This invention offers the following advantages: First, it analyzes the changes in plantar pressure during a single-leg support phase to determine the degree of gait instability in the elderly person during each single-leg support phase. Then, it combines the differences in plantar pressure and time intervals between adjacent single-leg support phases to determine the degree of imbalance and loss of control. Simultaneously, based on the relationship between gait instability and walking speed changes during walking, and considering the trend of gait instability, it determines the degree of reaction sluggishness. Finally, it integrates these two aspects—the degree of imbalance and loss of control and the degree of reaction sluggishness—to obtain the fall risk value for the elderly person. Compared to existing methods for obtaining fall risk values ​​for the elderly, this invention involves a wider variety of data, all closely related to fall risk. Furthermore, it analyzes multiple aspects of a single-leg support phase and each single-leg support phase during walking, improving the accuracy and reliability of fall risk value acquisition, thereby enhancing the accuracy and reliability of fall risk warnings for the elderly. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the hardware composition of a fall prevention system for the elderly based on gait assessment provided in one embodiment of the present invention; Figure 2 This is a flowchart of the method executed by the data processor in the fall prevention system for the elderly based on gait assessment provided in one embodiment of the present invention; Figure 3 This is a pressure waveform diagram provided in one embodiment of the present invention; Figure 4 This is a flowchart of the process for obtaining gait instability according to an embodiment of the present invention; Figure 5 This is a flowchart illustrating the specific implementation of step S12 provided in one embodiment of the present invention; Figure 6 This is a flowchart of the process for obtaining the degree of imbalance and loss of control provided in one embodiment of the present invention; Figure 7This is a flowchart illustrating the process of obtaining relevant information according to an embodiment of the present invention. Detailed Implementation

[0015] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the present invention are described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. All data and information collected in this application have been obtained with full consent.

[0017] This embodiment provides a fall prevention system for the elderly based on gait assessment. It mainly assesses the fall risk of the elderly by analyzing their gait parameters and combines relevant sensors, artificial intelligence and other technologies to achieve real-time monitoring and early warning.

[0018] When an elderly person walks, if one foot is lifted, the body's center of gravity is primarily supported by the other foot. Pressure sensors are installed on the soles of the elderly person's feet. High-precision foot pressure sensors (such as pressure sensor insoles or pressure sensors embedded in ordinary insoles) are selected. In this embodiment, multiple pressure detection points are set at different locations on the pressure sensor insole, or several pressure sensors are embedded in different locations on the insole, thereby dividing the elderly person's sole into multiple regions (for example, dividing the heel into several regions and the forefoot into several regions). By acquiring the pressure in each region, accurate pressure data at different locations on the elderly person's sole can be detected.

[0019] It should be understood that pressure sensors can be powered by small, high-density micro-batteries, such as button batteries. Furthermore, pressure sensors integrate a wireless communication module for wireless transmission of pressure data, eliminating the need for wiring. Additionally, the sampling frequency of the pressure sensor can be set according to actual needs.

[0020] The elderly person is also equipped with sensors to detect changes in walking speed, such as accelerometers. Similarly, the accelerometers are powered by small, high-density micro-batteries, such as button batteries. Moreover, the accelerometers integrate wireless communication modules for wireless transmission of acceleration data.

[0021] This embodiment presets a monitoring period. Based on relevant data within this monitoring period, it determines the fall risk value for the elderly and provides a fall risk warning for the elderly during this monitoring period. The length of this monitoring period can be set according to actual needs.

[0022] Using pressure sensors installed on the soles of the elderly person's shoes, the time period from when one foot first contacts the ground and begins to collect pressure data until the pressure on that foot completely disappears is recorded as one foot support period. Each foot support period represents one step taken. In this embodiment, the starting time of this monitoring period can be the starting time of a certain foot support period during the elderly person's walking process (i.e., the moment when pressure data begins to be detected), and the ending time of a subsequent foot support period (i.e., the moment when the pressure data disappears) can be used as the ending time of this monitoring period, so that the monitoring period includes several foot support periods. It should be understood that two adjacent foot support periods correspond to different feet. The two feet walk alternately, thus obtaining multiple foot support periods in time sequence within this monitoring period.

[0023] This embodiment provides a fall prevention system for the elderly based on gait assessment, including a data processor. This data processor can be a conventional data processing chip, such as a central processing unit, and is equipped on the elderly person's body. The data processor is also powered by a button battery and integrates a wireless communication module for wireless communication with various sensors, such as… Figure 1 As shown, this data processor acquires data from various sensors and processes the data. As an example, the data processor can be configured in the elderly person's smartphone. In one exemplary embodiment, the wireless communication module between the various sensors and the data processor can be a short-range wireless communication module, such as a Bluetooth module, to achieve short-range wireless communication. Additionally, the data processor is also equipped with a long-range wireless communication module, such as a 5G module, for long-distance wireless transmission with a remote backend, enabling monitoring personnel to obtain relevant data about the elderly person in a timely manner. In one exemplary embodiment, the remote backend can be the smartphone of the elderly person's children, and the monitoring personnel can be the elderly person's children.

[0024] This data processor is used to perform tasks such as Figure 2 The processing strategy shown is as follows: Step S1: Detect the plantar pressure of the elderly during walking, and obtain the degree of gait instability of the elderly in each single-leg support phase based on the changes in plantar pressure. Step S2: Based on the degree of gait instability, as well as the differences in plantar pressure and time intervals between adjacent single-leg support phases, determine the degree of imbalance and loss of control in the elderly person; Step S3: Determine the degree of gait instability and changes in walking speed of the elderly during the walking process, and combine the trend of changes in gait instability to obtain the degree of slow reaction of the elderly; Step S4: Combine the degree of imbalance and loss of control with the degree of slow reaction to obtain the fall risk value for the elderly.

[0025] The following detailed explanation of each step, in conjunction with the accompanying drawings, is provided.

[0026] Step S1: Detect the plantar pressure of the elderly during walking, and obtain the degree of gait instability of the elderly in each single-leg support phase based on the changes in plantar pressure.

[0027] Monitoring plantar pressure during walking in older adults can identify abnormal gait patterns, such as uneven pressure distribution or excessive pressure on certain areas. These abnormal gait patterns are often early warning signs of fall risk, and the analysis of foot pressure distribution can be used to develop personalized intervention plans. Figure 3 The image shows a pressure waveform of a certain area on the sole of the foot during a single-leg support phase. The horizontal axis represents time, and the vertical axis represents the plantar pressure value.

[0028] For any single-leg support phase, the degree of gait instability during that phase is determined by the change in plantar pressure. In an exemplary embodiment, such as... Figure 4 As shown, the following is a specific process for obtaining the degree of gait instability: Step S11: Determine the pressure difference between each area of ​​the sole at every two moments during the single-leg support phase to obtain the degree of unevenness in the sole pressure distribution in each area.

[0029] For any given region, determine the plantar pressure sequence during the single-foot support period in that region. The plantar pressure sequence consists of pressure values ​​at multiple moments arranged in chronological order.

[0030] The pressure difference between any two moments in the plantar pressure sequence during the single-leg support phase of a given region is determined. Specifically, the pressure difference is the absolute value of the pressure difference. This process is repeated across all possible moments to obtain the absolute value of the pressure difference between any two possible moments. Then, the average of these absolute values ​​is calculated as the degree of plantar pressure unevenness during the single-leg support phase of that region. A larger absolute value indicates a more uneven plantar pressure distribution during the single-leg support phase, and thus a higher degree of unevenness. If an elderly person exhibits abnormal gait (such as limping, excessive inversion, or eversion), it may cause excessive or insufficient pressure in certain areas, resulting in uneven pressure distribution. Uneven pressure distribution can lead to foot discomfort and fatigue. Good plantar pressure distribution helps maintain a proper center of gravity, reducing deviation and improving stability. This method is used to determine the degree of plantar pressure unevenness during the single-leg support phase in each region.

[0031] Step S12: Based on the difference between the maximum pressure between two regions during the single-leg support period, and combined with the unevenness of the plantar pressure distribution in each region, the degree of gait instability is obtained.

[0032] During normal walking, although there may be slight differences in pressure across different areas of the foot, these differences are usually minor, and the pressure distribution is relatively even. This means that in the gait of a healthy elderly person, there will be no extreme differences in pressure between different areas. Therefore, by combining the differences in pressure between different areas with the degree of unevenness in the plantar pressure distribution in each area, the degree of gait instability during the single-leg support phase can be obtained. In an exemplary embodiment, such as... Figure 5 As shown, the following is a specific implementation process: Step S121: Calculate the average difference between the maximum pressures of all two regions during the single-leg support period to obtain the overall difference in maximum pressures during the single-leg support period.

[0033] Since each region includes pressure values ​​at multiple moments during the single-leg support period, the maximum pressure (i.e., the maximum peak pressure) is determined from the pressure values ​​at multiple moments during the single-leg support period of that region, and this is taken as the maximum pressure of that region during the single-leg support period, thus obtaining the maximum pressure of each region during the single-leg support period.

[0034] The difference between the maximum pressure of any two regions during the single-leg support phase is obtained, specifically the absolute value of the difference in maximum pressure. This is done by iterating through all regions and obtaining the absolute value of the difference in maximum pressure between all possible pairs of regions. Then, the average of these absolute values ​​is calculated as the overall difference in maximum pressure during the single-leg support phase. This overall difference in maximum pressure during the single-leg support phase characterizes the difference in maximum pressure between different regions during this phase. In a normal gait, the maximum pressure of different regions during this single-leg support phase should not differ too much. A larger overall difference in maximum pressure indicates uneven weight-bearing in the elderly person, leading to poor gait stability and an increased risk of falls.

[0035] Step S122: Calculate the average value of the unevenness of plantar pressure distribution in each region to obtain the overall unevenness of plantar pressure distribution during the single-foot support phase.

[0036] The average value of the unevenness of plantar pressure distribution in each region during the single-leg support phase is calculated as the overall unevenness of plantar pressure distribution during that phase. A higher degree of unevenness in the overall plantar pressure distribution during the single-leg support phase indicates that the elderly person is experiencing uneven weight-bearing while walking, which leads to poorer gait stability and increases the risk of falls.

[0037] Step S123: Combine the overall difference in maximum pressure and the uneven distribution of overall plantar pressure to obtain the degree of gait instability.

[0038] The gait instability degree of a single-leg support phase is obtained by integrating the overall difference in maximum pressure and the overall unevenness of plantar pressure distribution during that phase. The gait instability degree is positively correlated with both the overall difference in maximum pressure and the overall unevenness of plantar pressure distribution. In an exemplary embodiment, the gait instability degree of a single-leg support phase is obtained by calculating the product of the overall difference in maximum pressure and the overall unevenness of plantar pressure distribution during that phase. This yields the gait instability degree for each single-leg support phase.

[0039] Step S2: Based on the degree of gait instability, as well as the differences in plantar pressure and time intervals between adjacent single-leg support phases, the degree of imbalance and loss of control in the elderly is determined.

[0040] Step S1 obtains the gait instability level of each single-leg support phase within the monitoring time period. Then, based on the gait instability level of each single-leg support phase, as well as the differences in plantar pressure and time intervals between adjacent single-leg support phases, the degree of imbalance and loss of control in the elderly is obtained. In an exemplary embodiment, such as... Figure 6 As shown, the following is a specific process for obtaining the degree of imbalance and loss of control: Step S21: Integrate the plantar pressure differences between adjacent single-leg support phases to obtain the overall plantar pressure difference.

[0041] For any given single-leg support phase, calculate the average pressure value across all areas and at all times within that phase. This average is taken as the plantar pressure for that single-leg support phase, thus obtaining the plantar pressure for each single-leg support phase. Calculate the difference in plantar pressure between any two adjacent single-leg support phases, specifically the absolute value of the difference. Since two adjacent single-leg support phases correspond to the two feet, this means obtaining the difference in plantar pressure between the two feet during the alternating contact with the ground during walking.

[0042] The plantar pressure difference between each two adjacent single-leg support phases is integrated. Specifically, the average of the absolute values ​​of the plantar pressure difference between each two adjacent single-leg support phases is calculated as the overall difference in plantar pressure during the elderly's walking process.

[0043] Step S22: Based on the overall difference in plantar pressure and the overall degree of gait instability, obtain the degree of center of gravity deviation of the elderly during walking.

[0044] The overall gait instability during walking is obtained from the gait instability during each single-leg support phase. In this embodiment, the average gait instability during all single-leg support phases within the monitoring period is calculated as the overall gait instability during walking.

[0045] The degree of center of gravity deviation during walking in the elderly is obtained based on the overall difference in plantar pressure and the overall degree of gait instability. In an exemplary embodiment, the product of the overall difference in plantar pressure and the overall degree of gait instability is calculated and then normalized. The normalized result is the degree of center of gravity deviation during walking in the elderly. The normalization method here can be a hyperbolic tangent function.

[0046] A higher degree of center of gravity deviation indicates poor gait stability and uneven weight distribution between the elderly person's feet during the monitoring period, suggesting a greater likelihood of center of gravity shift. The greater the degree of center of gravity deviation, the higher the risk of falls. Monitoring the degree of center of gravity deviation during walking allows for the timely identification of potential gait risks and helps in the early warning of potential fall hazards.

[0047] Step S23: Obtain the overall difference in time intervals based on the difference between every two adjacent time intervals during the walking process.

[0048] Judging the degree of imbalance and loss of control in an elderly person's walking process solely by the degree of deviation of the center of gravity has certain limitations. It is also necessary to combine the abnormality of the time interval between the elderly person's steps to comprehensively determine the degree of imbalance and loss of control.

[0049] In an exemplary embodiment, the time interval between any two adjacent single-leg support phases is determined. For any two adjacent single-leg support phases, the start time of the preceding single-leg support phase and the start time of the following single-leg support phase are determined. The time interval between the start times of these two adjacent single-leg support phases is then obtained as the time interval between these two adjacent single-leg support phases. This yields the time interval between any two adjacent single-leg support phases. These time intervals are then arranged chronologically to obtain the time interval sequence during the walking process.

[0050] The process involves obtaining the absolute value of the difference between each pair of adjacent time intervals in the time interval sequence during walking. These absolute differences are then averaged to represent the overall difference in time intervals during walking. It should be understood that during normal walking, the stride is relatively regular (i.e., the stride frequency is relatively regular), and the time taken for adjacent strides is relatively similar. Therefore, the difference between adjacent time intervals is small, resulting in a small overall difference in time intervals. Conversely, if an elderly person experiences imbalance or loss of control while walking, their stride becomes irregular, the difference between adjacent time intervals is larger, and thus the overall difference in time intervals is larger.

[0051] Step S24: Based on the degree of deviation of the center of gravity and the overall difference in time intervals, obtain the degree of imbalance and loss of control.

[0052] The degree of imbalance and loss of control during walking is determined by the degree of center of gravity deviation and the overall difference in time intervals during the walking process. The degree of imbalance and loss of control is positively correlated with both the degree of center of gravity deviation and the overall difference in time intervals.

[0053] In one exemplary embodiment, the overall difference in time intervals during walking is normalized, and then the product of the degree of center of gravity deviation and the normalized overall difference in time intervals is calculated. This product represents the degree of imbalance and loss of control during walking. The normalization method here can be a hyperbolic tangent function. The stronger the degree of imbalance and loss of control, the greater the difference in cadence during the elderly person's walking process when the center of gravity deviates, indicating that the elderly person's imbalance and loss of control during walking is more obvious within the monitoring period.

[0054] Step S3: Determine the degree of gait instability and changes in walking speed of the elderly during the walking process, and combine the trend of changes in gait instability to obtain the degree of slow reaction of the elderly.

[0055] Step S2 analyzes the plantar pressure data of the elderly to assess their balance control ability during walking. However, the risk of falls in the elderly is not only related to balance control ability but also to their reaction time to external disturbances during walking, such as sudden pushes or external interference, which also affects the risk of falls. When faced with sudden events, a faster reaction time usually means that the elderly can better adjust their center of gravity and regain balance; while a slower reaction time indicates that the strength and endurance of the elderly's lower limb muscles are declining, meaning that there is a higher risk of falls during walking.

[0056] An accelerometer is used to detect changes in the elderly person's walking speed during their walk. The more drastic the change in walking speed, the higher the collected acceleration value. It should be understood that the sampling frequency of the accelerometer is set according to actual needs and can be used with the pressure sensor for data acquisition, with both types of data collected synchronously. This allows the elderly person to obtain their walking acceleration at various moments during their walk. Since a single-leg support phase contains multiple moments, there are multiple walking accelerations within a single-leg support phase. Therefore, the average walking acceleration within a single-leg support phase is calculated as the walking acceleration for that single-leg support phase, thus obtaining the walking acceleration for each single-leg support phase.

[0057] First, determine the correlation between the degree of gait instability and changes in walking speed in the elderly person during walking, that is, determine the correlation between the degree of gait instability and walking acceleration. In an exemplary embodiment, such as... Figure 7 As shown, the following is a specific process for obtaining relevant information: Step S31: Perform curve fitting on the gait instability sequence to obtain the first fitted curve.

[0058] Based on the sequence of single-leg support phases during the elderly person's walking process, the gait instability degree of each single-leg support phase is ordered chronologically to obtain a gait instability degree sequence. Then, curve fitting is performed on the gait instability degree sequence to obtain the first fitted curve.

[0059] Step S32: Perform curve fitting on the walking acceleration sequence to obtain the second fitted curve.

[0060] Based on the sequence of single-leg support phases during the elderly person's walking process, the walking acceleration during each single-leg support phase is sorted chronologically to obtain a walking acceleration sequence. Then, curve fitting is performed on the walking acceleration sequence to obtain a second fitted curve.

[0061] Step S33: Obtain the correlation coefficient between the first fitted curve and the second fitted curve.

[0062] The correlation coefficient between the first and second fitted curves is obtained to characterize their correlation. In an exemplary embodiment, the Pearson correlation coefficient between the first and second fitted curves is calculated, and then normalized. Since the Pearson correlation coefficient ranges from -1 to 1, the normalization method is: (Pearson correlation coefficient + 1) / 2. The larger the normalized Pearson correlation coefficient, the more similar the gait instability and acceleration changes of the elderly person during walking, indicating a slower response to external disturbances and a higher risk of falls.

[0063] To obtain the trend of gait instability during walking, in one exemplary embodiment, a first-order difference sequence of gait instability is obtained. Each data point in the first-order difference sequence is the difference between the previous and subsequent gait instability values. Then, the average value of the data in the first-order difference sequence is calculated to obtain the trend of gait instability. The trend is positively correlated with the average value. Specifically, a positive average value indicates that gait instability is generally increasing, and the larger the average value, the stronger the increasing trend. A zero average value indicates that gait instability is generally stable. A negative average value indicates that gait instability is generally decreasing, and the larger the absolute value of the average value, the stronger the decreasing trend. In one exemplary embodiment, the average value is normalized using the sigmoid function. The normalized average value is used as the trend of gait instability. Therefore, the greater the trend of gait instability, the more it indicates that the elderly person's balance ability is gradually declining during walking, and the more it indicates that the elderly person is less responsive to external disturbances during walking, and the higher the risk of falling.

[0064] The higher the normalized Pearson correlation coefficient, the more correlated the degree of gait instability and the change in walking acceleration are. If the degree of gait instability also shows an increasing trend, this co-change can more clearly indicate that the older person has a longer reaction time when faced with a sudden event. For example, when a rapid change in walking acceleration is not accompanied by a corresponding adjustment in the degree of gait instability, it may mean that the older person has failed to make an appropriate response in time, showing that their reaction to a sudden event is slow.

[0065] Based on the normalized Pearson correlation coefficient obtained above and the trend of gait instability, the degree of reaction sluggishness is calculated. The degree of reaction sluggishness is positively correlated with both the normalized Pearson correlation coefficient and the trend of gait instability. In an exemplary embodiment, the product of the normalized Pearson correlation coefficient and the normalized average data is calculated; this product represents the degree of reaction sluggishness during the elderly person's walking process. The degree of reaction sluggishness can objectively assess the elderly person's ability to react to sudden events; a higher degree of reaction sluggishness indicates poorer balance performance.

[0066] Step S4: Combine the degree of imbalance and loss of control with the degree of slow reaction to obtain the fall risk value for the elderly.

[0067] The fall risk of the elderly is characterized by both the degree of imbalance and loss of control and the degree of reaction delay. Therefore, the fall risk value of the elderly is obtained by integrating the degree of imbalance and loss of control and the degree of reaction delay. In an exemplary embodiment, the Euclidean norm of the degree of imbalance and loss of control and the degree of reaction delay are calculated, and the result is used as the fall risk value of the elderly during the monitoring period. The purpose of using the Euclidean norm for calculation is to achieve a balance between the distribution of plantar pressure and the delay in reaction to sudden events, thereby illustrating the fall risk value of the elderly during walking. The higher the fall risk value of the elderly, the higher the fall risk.

[0068] In an exemplary embodiment, after obtaining the fall risk value of the elderly during the monitoring period, the fall risk value can be used to issue an early warning for the elderly's fall risk, thereby achieving fall prevention for the elderly. This embodiment presets a fall risk threshold for the elderly, with a value ranging from 0 to 1. The specific value is set according to the judgment needs. If a safer warning is required, the preset fall risk threshold can be set smaller; in this embodiment, 0.5 is used as an example. The fall risk value of the elderly during the monitoring period is compared with the preset fall risk threshold. If the fall risk value is greater than or equal to the preset fall risk threshold, the data processor generates and outputs a fall warning signal, which is sent to the elderly's children's smartphone terminal device. The children can immediately receive the fall warning signal and conveniently take relevant safety measures immediately. Alternatively, multiple risk ranges can be preset. Based on the range within which the elderly person's fall risk value falls, the risk level can be determined as high, medium, or low. Subsequently, corresponding physical rehabilitation training programs, such as joint mobilization exercises and flexibility training, can be tailored to the individual's fall risk level. Furthermore, other fall prevention measures, such as environmental modifications and medication management, can be combined to form a comprehensive fall prevention approach. Additionally, this embodiment can continuously update the monitoring time period to achieve continuous real-time monitoring of the elderly person's fall risk.

[0069] As another implementation method, this embodiment can also acquire fall risk values ​​for the elderly over multiple monitoring time periods, as well as fall risk-related feature data for each monitoring time period, such as plantar pressure and walking acceleration data. A training set is constructed based on the feature data and fall risk values ​​for the elderly over multiple monitoring time periods. Then, the neural network is trained using the training set, outputting the fall risk values ​​for the elderly from the trained model. This is compared with the actual fall risk values ​​obtained, and the hyperparameters during model training are adjusted to minimize the difference between the trained output and the actually calculated fall risk values. Cross-validation (such as k-fold cross-validation) is used to ensure the model's stability and generalization ability. If the model's accuracy is insufficient, more features can be added, different models can be selected, or data augmentation can be performed. Finally, a fall prevention model for the elderly is trained. Subsequently, based on the trained fall prevention model, real-time fall risk warnings for the elderly are implemented.

[0070] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0071] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A fall prevention system for the elderly based on gait assessment, characterized in that, Includes a data processor, which is used to execute the following processing strategies: The foot pressure of elderly people during walking is detected, and the degree of gait instability of elderly people in each single-leg support phase is obtained based on the changes in foot pressure; multiple pressure detection points are set at different positions on the elderly people's insoles; Determining the degree of imbalance and loss of control in the elderly includes: integrating the differences in plantar pressure between adjacent single-leg support phases to obtain the overall difference in plantar pressure; based on the overall difference in plantar pressure and the overall situation of gait instability, determining the degree of center of gravity deviation during the elderly's walking process; and based on the difference between every two adjacent time intervals during the walking process, determining the overall difference in time intervals. The determination of the time intervals includes: for any two adjacent single-leg support phases, determining the start time of the preceding single-leg support phase and the start time of the following single-leg support phase, and using the time interval between the start times of the preceding and following single-leg support phases as the time interval between two adjacent single-leg support phases; and based on the degree of center of gravity deviation and the overall difference in time intervals, determining the degree of imbalance and loss of control. The degree of imbalance and loss of control is positively correlated with both the degree of center of gravity deviation and the overall difference in time intervals. Determine the degree of gait instability and changes in walking speed in elderly individuals during walking, and combine this with the trend of gait instability to determine the degree of slowed reaction in the elderly individuals; By combining the degree of imbalance and loss of control with the degree of slow reaction, a fall risk value for the elderly is obtained; The process of obtaining the degree of gait instability includes: Determine the pressure difference between each area of ​​the sole at every two moments during the single-leg support phase to obtain the degree of unevenness in the sole pressure distribution in each area. The degree of gait instability is obtained by considering the difference between the maximum pressure in each pair of regions during the single-leg support phase, combined with the degree of unevenness in the plantar pressure distribution in each region. The degree of gait instability is determined by the difference in maximum pressure between two regions during the single-leg support phase, combined with the degree of unevenness in plantar pressure distribution in each region. This includes: Calculate the average difference between the maximum pressures of all two regions during the single-leg support phase to obtain the overall difference in maximum pressures during the single-leg support phase; Calculate the average value of the unevenness of plantar pressure distribution in each region to obtain the overall unevenness of plantar pressure distribution during the single-foot support phase; The degree of gait instability is obtained by integrating the overall difference in maximum pressure and the overall unevenness of plantar pressure distribution; the degree of gait instability is positively correlated with both the overall difference in maximum pressure and the overall unevenness of plantar pressure distribution.

2. The fall prevention system for the elderly based on gait assessment as described in claim 1, characterized in that, The overall gait instability level is the average value of the gait instability level.

3. The fall prevention system for the elderly based on gait assessment as described in claim 1, characterized in that, The determination of the degree of gait instability and changes in walking speed in the elderly includes: A curve fitting was performed on the gait instability sequence to obtain the first fitting curve; the gait instability sequence was obtained by arranging the various gait instabilities of the elderly in chronological order during the walking process. A second fitted curve is obtained by curve fitting the walking acceleration sequence; the walking acceleration sequence is obtained by arranging the elderly person's walking accelerations in chronological order during the walking process. Obtain the correlation coefficient between the first fitted curve and the second fitted curve, where the correlation is the correlation coefficient.

4. The fall prevention system for the elderly based on gait assessment as described in claim 3, characterized in that, The process of obtaining the trend of gait instability includes: Obtain the first-order difference sequence of the gait instability sequence; The average value of the data in the first-order difference sequence is obtained to obtain the trend of the change in gait instability; the trend of change is positively correlated with the average value of the data.

5. A fall prevention system for the elderly based on gait assessment as described in claim 3, characterized in that, The process of obtaining the degree of sluggishness includes: The degree of sluggishness is obtained based on the correlation coefficient and the trend of change; the degree of sluggishness is positively correlated with both the correlation coefficient and the trend of change.

6. A fall prevention system for the elderly based on gait assessment as described in claim 1, characterized in that, The Euclidean norm of the degree of imbalance and loss of control and the degree of slow reaction is calculated as the fall risk value of the elderly person.

7. A fall prevention system for the elderly based on gait assessment as described in claim 1, characterized in that, The processing strategy also includes: The elderly person's fall risk value is compared with a preset elderly person's fall risk threshold. If the elderly person's fall risk value is greater than or equal to the preset elderly person's fall risk threshold, a fall warning signal is output.

Images