Intelligent training shoes for rehabilitation after lower limb fracture operation and control method thereof

Abstract

The embodiment of the invention relates to rehabilitation auxiliary equipment, in particular to a pair of lower limb fracture postoperative rehabilitation intelligent training shoes and a control method thereof. The intelligent training shoe comprises a shoe body, and the shoe body is provided with an adjustable fixing band and a bearing panel; the pressure sensing modules are arranged in the front area and the rear area of the bearing panel and used for detecting pressure and obtaining pressure data; the data processing module is connected with the pressure sensing module and is used for periodically collecting and storing the pressure data, dynamically tracking a weighing result and generating gait analysis data; and the terminal analysis module is located outside the shoe body, establishes communication connection with the data processing unit, receives the gait analysis data reported by the data processing module, and generates training indication data according to the gait analysis data.

Description

Technical Field

[0001] Multiple embodiments of this specification relate to rehabilitation assistive devices, and specifically to an intelligent training shoe for rehabilitation after lower limb fracture surgery and a control method thereof. Background Art

[0002] Smart training shoes for rehabilitation after lower limb fracture surgery are a type of rehabilitation equipment used to help patients regain their ability to walk. They are mainly used in hospital rehabilitation centers and home rehabilitation environments. In hospital rehabilitation centers, patients can use rehabilitation shoes for training under the professional guidance of doctors or physical therapists, and doctors can adjust treatment plans based on real-time monitoring data. In a home environment, patients can complete training independently through rehabilitation shoes connected to mobile applications, while synchronizing rehabilitation data to the medical team for remote monitoring and guidance. Rehabilitation shoes use a variety of advanced technologies. For example, pressure sensing technology monitors the pressure distribution of various parts of the foot through sensors integrated in the soles to assess whether the patient's gait is normal. The data analysis and feedback system can process the collected data and provide real-time feedback based on the analysis results, such as adjusting the support strength or correcting the gait through vibration prompts. Although the current rehabilitation training shoes have relatively complete data collection capabilities, they cannot restore complex gait data. Gait data includes sagittal plane dorsiflexion / plantar flexion angles, coronal plane inversion / valgus angles, and horizontal plane internal and external rotation angles. Gait cannot be accurately restored only by detecting data from the pressure sensor on the soles. For this reason, further research on rehabilitation training shoes is needed. Summary of the invention

[0003] Multiple embodiments of this specification describe an intelligent training shoe for rehabilitation after lower limb fracture surgery and a control method thereof.

[0004] In a first aspect, the embodiments of this specification provide a smart training shoe for lower limb fracture postoperative rehabilitation, comprising: A shoe body, wherein the shoe body is provided with an adjustable fixing strap and a load-bearing panel; A pressure sensing module is disposed in the front area and the rear area of ​​the load-bearing panel, and is used to detect pressure and obtain pressure data; A data processing module, connected to the pressure sensing module, periodically collects and stores the pressure data, and dynamically tracks the weighing results to generate gait analysis data; The terminal analysis module is located outside the shoe body, establishes a communication connection with the data processing unit and receives the gait analysis data reported by the data processing module, and generates training instruction data according to the gait analysis data.

[0005] In a second aspect, the present specification provides a control method for the aforementioned intelligent training shoe for rehabilitation after lower limb fracture surgery, comprising the steps of: The data processing module periodically reads the detection value of the pressure sensing module; The data processing module periodically collects and stores the pressure data, and dynamically tracks the weighing results to generate gait analysis data; The data processing module obtains the body weight according to the pressure data, and obtains the plantar equivalent body weight distribution sequence data according to the pressure data and the body weight within a preset time period; The data processing module aligns the plantar equivalent weight distribution sequence data according to the time axis as gait analysis data; The terminal analysis module performs three-dimensional gait reconstruction according to the gait analysis data to obtain three-dimensional gait data, and generates a training result report according to the three-dimensional gait data.

[0006] The beneficial effects brought by the technical solutions provided by some embodiments of this specification include at least: In multiple embodiments of the present specification, the provided intelligent training shoes for rehabilitation after lower limb fracture surgery can obtain more accurate weighing results with the help of pressure sensing modules and data processing modules, helping patients and medical staff to more accurately grasp the force of the patient's affected limb, helping to more effectively carry out rehabilitation training and protect the affected limb. With the help of a three-dimensional gait model, gait can be reconstructed according to gait analysis data, and more accurate tracking of gait can be achieved, which helps to more clearly grasp the patient's rehabilitation status. Through further improvements in the restoration of gait by the three-dimensional gait model, the accuracy of gait reconstruction can be more effectively improved by means of the correlation processing of gaits within a preset time. The plantar force is analyzed with the help of a reference point set, and the detection value of the pressure sensor is further processed by filtering, which helps to filter out the interference of plantar force to a certain extent and improve the accuracy of gait recognition.

[0007] Other features and advantages of the various embodiments of the present specification will be further disclosed in the following detailed description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of this specification, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0009] Figure 1 This is a schematic diagram of the rehabilitation smart training shoes provided in the embodiments of this specification.

[0010] Figure 2 This is a schematic diagram of the control architecture of the rehabilitation smart training shoes provided in the embodiments of this specification.

[0011] Figure 3 This is a schematic diagram of a weighing circuit provided in an embodiment of this specification.

[0012] Figure 4 A schematic flow chart of a method for dynamically tracking weighing results provided in an embodiment of this specification.

[0013] Figure 5 A schematic diagram of a method for obtaining plantar equivalent weight distribution sequence data provided in an embodiment of this specification.

[0014] Figure 6 A schematic diagram of the reference points on the sole of the foot provided in the embodiments of this specification.

[0015] Figure 7 A schematic flow chart of a method for obtaining an equivalent pressure of a plantar reference point provided in an embodiment of this specification.

[0016] Figure 8 A schematic flow chart of a method for three-dimensional gait reconstruction provided in an embodiment of this specification. DETAILED DESCRIPTION

[0017] The following is an explanation and description of the technical solutions of the embodiments of this specification in conjunction with the drawings of the embodiments of this specification, but the following embodiments are only preferred embodiments of this specification, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the scope of protection of this specification.

[0018] The terms "first", "second", "third", etc. in the description and claims of this specification and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0019] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation of this specification.

[0020] The data involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection of relevant data complies with the relevant laws, regulations and standards of relevant countries and regions.

[0021] Before describing the technical solution in this specification, an introduction is given to the application scenarios and related technologies of the technical solution.

[0022] The smart training shoes for postoperative rehabilitation of lower limb fractures are a rehabilitation device that combines modern technology to help patients regain their ability to walk. Its main application scenarios include hospital rehabilitation centers and home rehabilitation environments. In hospital rehabilitation centers, this smart training shoe can provide patients with professional rehabilitation guidance and support. Doctors or physical therapists can use the sensor technology in the training shoes to monitor the patient's gait and plantar pressure distribution in real time, and accurately assess the patient's rehabilitation progress. Based on these data, the medical team can tailor a personalized rehabilitation plan and dynamically adjust the training intensity and content according to the patient's recovery status. For example, in the early stage of rehabilitation, more attention may be paid to reducing the burden on the affected limb and preventing the formation of incorrect gait; in the later stage of rehabilitation, it may focus on strengthening muscle strength and improving joint mobility. In addition, by connecting to the hospital's information system, doctors can also remotely track the patient's daily training to ensure the effective implementation of the treatment plan.

[0023] For the home environment, the smart training shoes for postoperative rehabilitation of lower limb fractures have greatly improved the convenience and flexibility of rehabilitation training. Patients do not need to go to the hospital frequently and can conduct effective self-training at home. With the connection to the mobile application, patients can easily record and view their rehabilitation progress, such as daily walking steps, gait analysis results, etc. The application usually also provides a series of rehabilitation training suggestions and goal setting functions to encourage patients to persist in completing training tasks. When encountering problems, patients can also send relevant data to doctors or physical therapists through the application for help, and realize remote consultation and guidance. This method not only saves time and energy in going to and from the hospital, but also allows patients to conduct rehabilitation training more relaxedly in a familiar environment, which helps to improve the overall rehabilitation effect.

[0024] However, current rehabilitation training shoes rely solely on sensors on the soles of the feet and cannot restore the patient's gait very well. To this end, this manual provides a method of using historical relevant data to provide a more accurate three-dimensional gait restoration by building a model, so as to achieve a more accurate grasp of the patient's rehabilitation training status and help improve the recovery training effect. Please refer to the attached Figure 1 The intelligent training shoe for lower limb fracture surgery provided in this specification includes a shoe body, wherein the shoe body is provided with a wrapping lining 10 and an adjustable fixing strap 11, a front baffle 12 is provided at the front end of the shoe body, a training shoe shell 13 is provided at the bottom of the shoe body, and a load-bearing panel 14 is provided in the training shoe shell 13; a pressure sensing module is provided in the front and rear areas of the sole for detecting pressure and obtaining pressure data; A data processing module, connected to the pressure sensing module, periodically collects and stores the pressure data, and dynamically tracks the weighing results to generate gait analysis data; The terminal analysis module is located outside the shoe body, establishes a communication connection with the data processing unit and receives the gait analysis data reported by the data processing module, and generates training instruction data according to the gait analysis data.

[0025] The method provided in this application is applied to Figure 2 The system architecture shown in Figure 2 As shown, the system architecture includes a terminal analysis module and a terminal device 30, and an interactive interface is provided on the terminal device 30, wherein the interactive interface can be run on the terminal device 30 in the form of a browser, or can be run on the terminal device 30 in the form of an independent application (application, APP), etc. The specific display form of the interactive interface is not limited here. The terminal analysis module involved in this application can be an independent physical server 20, or a server 20 cluster or distributed system composed of multiple physical servers 20, or a server 20 that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms, as a terminal analysis module. The terminal device 30 can be a smart phone, a tablet computer, a laptop, a PDA, a personal computer, a smart speaker, a smart TV, a smart watch, a vehicle-mounted device, a wearable device, etc., but is not limited thereto. The terminal device 30 and the terminal analysis module can be directly or indirectly connected by wired or wireless communication, and this application does not limit this. There is no limit to the number of terminal analysis modules and terminal devices 30.

[0026] Please see attached Figure 3The pressure sensor module includes pressure sensors and a weighing circuit respectively arranged in the front area and the rear area of ​​the load-bearing panel, the weighing circuit includes a signal input terminal SIG- and a signal input terminal SIG+, the signal input terminal SIG- and the signal input terminal SIG+ are grounded via capacitors C40 and C47 respectively, a capacitor C44 is connected between the signal input terminal SIG- and the signal input terminal SIG+, anodes of the Zener diodes D24 and D25 are respectively connected to the signal input terminal SIG- and the signal input terminal SIG+, and cathodes of the Zener diodes D24 and D25 are both connected to The signal input terminal SIG- is connected to the first input pin of the weighing chip U17 via a resistor R72, the signal input terminal SIG+ is connected to the second input pin of the weighing chip U17 via a resistor R77, capacitors C45 and C46 are connected between the first input pin and the second input pin of the weighing chip U17, the first input pin and the second input pin of the weighing chip U17 are grounded via capacitors C41 and C48 respectively, and the CS pin, SCLK pin, DIN pin, DOUT / DRDY pin and DRDY pin of the weighing chip U17 are connected to designated signal terminals via resistors respectively.

[0027] Please see attached Figure 4 When the data processing module dynamically tracks the weighing results, the following steps are performed: Step S101) periodically reading the pressure data, determining the effective weighing time according to the pressure data, and discarding the pressure data outside the effective weighing time; Step S102) converting the pressure data into percentage data according to the most recent weight calibration data; Step S103) filtering the percentage data using a filter; Step S104) obtaining a weighing result according to the filtered percentage data and the weight calibration data.

[0028] Wherein, according to the most recent weight calibration data, the method for converting the pressure data into percentage data includes: The percentage value is calculated according to the following formula: Percentage value = (percentage change / pressure detection value change) × pressure detection value + nonlinear compensation value, wherein the percentage change is the percentage value change between loading the preset weight and no load, the pressure detection value change is the pressure detection value change between loading the preset weight and no load, and the pressure detection value is the pressure detection value in the converted pressure data; The nonlinear compensation value is calculated by the following formula: Non-linear compensation value = (no-load percentage value / pressure detection value after loading preset weights - percentage value after loading preset weights / no-load pressure detection value) × change in pressure detection value.

[0029] The pressure sensing module includes a plurality of pressure sensors arranged in the front and rear areas of the sole. The data processing module generates gait analysis data according to the pressure data. The terminal analysis module performs three-dimensional gait reconstruction according to the gait analysis data to obtain three-dimensional gait data, and generates a training result report according to the three-dimensional gait data.

[0030] When the data processing module generates gait analysis data, the following steps are performed: Obtaining body weight according to the pressure data, and obtaining plantar equivalent body weight distribution sequence data according to the pressure data and body weight within a preset time period; The plantar equivalent weight distribution sequence data and the inertia data are aligned along a time axis to serve as gait analysis data.

[0031] Please see attached Figure 5 , according to the pressure data and body weight within a preset time period, a method for obtaining a sequence data of equivalent body weight distribution on the sole of the foot comprises: Step S201) Read the pre-established reference point set, see the attached Figure 6 The reference point set is a set of pre-set plantar reference points 16 .

[0032] Step S202) Obtaining the equivalent pressure of the plantar reference point 16 according to the positional relationship between the pressure sensor 15 and the reference point set, and the pressure data.

[0033] Please see the attached Figure 7 According to the positional relationship between the pressure sensor 15 and the reference point set, and the pressure data, the method for obtaining the equivalent pressure of the plantar reference point 16 includes: Step S301) Read the pressure data within a preset time length. Read the pressure data of a fixed time length (for example, a few seconds) from a data buffer or memory. Efficient data reading can be achieved through a circular queue or other suitable data structure.

[0034] Step S302) extract the amplitude of the frequency components above the preset frequency of the pressure data, and record it as the high-frequency amplitude. Use fast Fourier transform (FFT) or other frequency domain analysis methods to convert the pressure data in the time domain to the frequency domain, and then extract the amplitude of all frequency components above a preset frequency as the "high-frequency amplitude".

[0035] Step S303) Multiply the high-frequency amplitude by a preset factor and superimpose it on the frequency component below the preset frequency to obtain secondary pressure data. Operate in the frequency domain to find all frequency components below the preset frequency, and group the lower frequencies and higher frequencies. The combination pairing method is: obtain multiple frequency multiples of each lower frequency (frequency component below the preset frequency), find the closest frequency multiple for each higher frequency (frequency component above the preset frequency), and include the higher frequency component and the lower frequency component corresponding to the frequency multiple into the same group. Each group has a lower frequency component and zero or several higher frequency components. The amplitude of the higher frequency in each group is reduced by a preset factor and superimposed on the lower frequency component of the same group. After that, perform an inverse Fourier transform to convert the data back to the time domain to obtain the secondary corrected pressure data.

[0036] Step S304) A force relationship model is established according to the positional relationship between the pressure sensor 15 and the reference point set, wherein the input of the force relationship model is the pressure value at the position of the pressure sensor 15, and the output is the pressure value at the sole reference point 16. The force relationship model can predict the pressure distribution at a specific reference point on the sole of the foot according to the pressure value at the position of the pressure sensor 15. Exemplarily, the force relationship model can be established by linear interpolation or a machine learning algorithm (such as a neural network). In this embodiment, the force relationship model is established by means of a neural network model.

[0037] Step S305) After the force relationship model is trained using manually annotated samples, the equivalent pressure of the plantar reference point 16 is obtained according to the response of the force relationship model to the secondary pressure data. First, sufficient manually annotated samples are collected, i.e., known inputs (measured values ​​of each pressure sensor 15) and expected outputs (real pressure of the plantar reference point 16). These data are used to train the force relationship model mentioned above (exemplarily, the neural network is trained by a supervised learning algorithm). After the training is completed, the new secondary pressure data is predicted using the force relationship model to obtain the equivalent pressure at the plantar reference point 16.

[0038] Step S203) converts the equivalent pressure into a proportion of body weight, and obtains the equivalent weight distribution sequence data of the sole according to the proportion. Body weight is actually the weight of the patient and his / her clothes, as well as the weight of the items that the patient may hold in his / her hands. Therefore, the weight of the patient is different each time. Considering that the patient may also hold crutches or use handrails, the support force provided by the crutches and handrails is dynamically changing, so the use of known techniques in the art will not be able to correctly exclude the influence of the crutches and handrails on body weight. In this implementation, it is believed that the support force provided by the crutches and handrails has a certain periodic regularity. By using the previous filtering and frequency grouping and superposition method, the periodic regularity of the support force generated by the crutches and handrails can be taken into account to a certain extent, and the use of crutches and handrails can be uniformly mapped to the force representation of the reference point set by using the equivalent pressure. Then, the equivalent pressure is converted into a proportion of body weight, which basically excludes the absolute value of the influence on body weight when holding crutches or using handrails. As long as holding crutches or using handrails will not cause significant changes in the force proportion distribution law of the sole, it will not have a significant impact on the results. Of course, on the other hand, when the use of crutches and handrails causes a significant change in the force pattern of the sole of the foot during walking, the final effect of this embodiment will be partially affected. However, in subsequent steps, this effect will be compensated by comparing with the gait library and probability judgment.

[0039] Step S103) aligning the plantar equivalent weight distribution sequence data according to the time axis to use them as gait analysis data.

[0040] After the gait analysis data is obtained, the 3D gait reconstruction can be performed. Figure 8 The method for performing three-dimensional gait reconstruction based on the gait analysis data includes: Step S401) Establish a three-dimensional gait model, the input of the three-dimensional gait model is the gait analysis data, and the output is several values ​​of sagittal plane flexion angle, coronal plane roll angle, and horizontal plane rotation angle and their probabilities. The three-dimensional gait model is a neural network model, or other machine learning models. Although the plantar force condition has a strong correlation with the three-dimensional gait, there is a lack of mathematical derivation relationship. This embodiment uses a neural network model to fit the correlation. The three-dimensional gait model includes sagittal plane flexion angle, coronal plane roll angle, and horizontal plane rotation angle. When the sagittal plane flexion angle is positive, it is the sagittal plane dorsiflexion angle; when the sagittal plane flexion angle is negative, it is the sagittal plane plantar flexion angle. The absolute values ​​of the sagittal plane dorsiflexion angle and the sagittal plane plantar flexion angle are both in the range of 0°~45°. When the coronal plane rotation angle is positive, it is the coronal plane inversion angle, and when the coronal plane rotation angle is negative, it is the coronal plane valgus angle. The absolute values ​​of the coronal plane inversion angle and the coronal plane valgus angle are both in the range of 0° to 20°. When the horizontal plane rotation angle is positive, it is the horizontal plane internal rotation angle, and when the horizontal plane rotation angle is negative, it is the horizontal plane external rotation angle. The absolute values ​​of the horizontal plane internal rotation angle and the horizontal plane external rotation angle are both in the range of 0° to 15°.

[0041] Step S402) Using the manually labeled gait analysis data of sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle as sample data, the three-dimensional gait model is trained. The three-dimensional gait model is trained using the labeled sample data, and the training is completed when a preset accuracy rate is achieved.

[0042] Step S403) According to the time axis, a plurality of gait analysis data within a preset time length are read, input into the three-dimensional gait model, and a plurality of sagittal plane flexion angles, coronal plane rotation angles, horizontal plane rotation angles and their probabilities at each time point on the time axis are obtained.

[0043] Relying solely on the three-dimensional gait results obtained by the three-dimensional gait model based on the gait analysis data, the three-dimensional gait results include sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle, and the accuracy can be improved. This embodiment compares multiple three-dimensional gaits within a preset time length, and obtains the three-dimensional gait with the highest credibility as the final gait recognition result, which has higher accuracy. And with the help of the results identified by multiple three-dimensional gait models within the preset time length, mutual verification can smooth out the inaccuracy of the three-dimensional gait model itself.

[0044] Step S404) obtaining multiple groups of combinations of sagittal plane flexion angles, coronal plane rotation angles, and horizontal plane rotation angles along the time axis, and calculating the total probability of each combination.

[0045] Step S405) Compare each combination with a preset gait library to obtain the closest gait and similarity.

[0046] Step S406) According to the maximum weighted sum of the total probability and the similarity, a gait is selected as the identified gait, and the sagittal plane flexion angle, coronal plane flip angle, and horizontal plane rotation angle of the identified gait are used as the three-dimensional gait reconstruction result within the preset time length. The weighted sum of the total probability and the similarity of the closest gait in the gait library is used as the final judgment index, which has higher accuracy. It can effectively ensure the reliability of gait recognition results and provide a basis for the generation of subsequent training result reports.

[0047] Among them, the method of presetting the gait library includes: Obtain the historical combination, manually annotate gaits and their confidences, and obtain a gait library; After obtaining the gait and similarity of the closest combination, the similarity is corrected using the confidence.

[0048] On the other hand, in another embodiment, the method for presetting a gait library includes: Additional sensors for measuring sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle are installed to obtain gait analysis data of the patient walking within a preset time period, as well as the corresponding sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle; The sagittal plane flexion angle, the coronal plane rotation angle, and the horizontal plane rotation angle are associated with corresponding gait analysis data to obtain the gait, and then obtain a gait library.

[0049] On the other hand, in another embodiment, the intelligent training shoe for postoperative rehabilitation of lower limb fractures also includes a feedback module, which is arranged on the side of the shoe body, and the feedback module includes a vibration motor and an LED indicator light. The feedback module is connected to the data processing module. When it is judged that the abnormal plantar pressure distribution continues to exceed the preset time threshold according to the plantar equivalent weight distribution sequence data, the vibration motor and the LED indicator light a vibration and light alarm. The vibration motor is arranged on the side of the shoe body and can generate vibration when receiving an instruction. Vibration, as a tactile feedback method, can immediately attract the user's attention and is suitable for providing instant feedback when the user is focusing on other activities (such as walking or rehabilitation training). When the vibration motor vibrates, it will have a slight effect on the acceleration sensor and the pressure sensor 15. These data can be removed according to the vibration time of the vibration motor, and the frequency of the vibration motor can be filtered out by filtering. The LED indicator light is also located on the side of the shoe body, and can prompt the user by emitting light when an alarm is needed. Lights of different colors can be used to indicate different states or severity, such as red for emergency or high risk, yellow for warning, etc.

[0050] On the other hand, in another embodiment, the data processing module also calculates the difference between the sagittal flexion angle, the coronal plane flip angle, and the horizontal plane rotation angle between two consecutive identified gaits. When any difference of the sagittal flexion angle, the coronal plane flip angle, and the horizontal plane rotation angle is greater than a preset threshold, a vibration and light alarm is issued through the vibration motor and the LED indicator. The data processing module processes the input gait analysis data based on the method described above, and identifies the gait pattern and its sagittal flexion angle, coronal plane flip angle, and horizontal plane rotation angle at the current time point. For each newly identified gait, the data processing module compares its sagittal flexion angle, coronal plane flip angle, and horizontal plane rotation angle with the three angle values ​​corresponding to the previous gait, and calculates the difference. One or more thresholds are preset (corresponding to the sagittal flexion angle, the coronal plane flip angle, and the horizontal plane rotation angle, respectively), and these thresholds represent the limits of the normal gait variation range. Once it is detected that any angle difference exceeds its preset threshold, the vibration motor and the LED indicator will issue a vibration and light alarm.

[0051] On the other hand, in another embodiment, the data processing module also receives a reference gait, and when the difference between the sagittal flexion angle, coronal plane rotation angle, and horizontal plane rotation angle between the identified gait and the reference gait exceeds a preset threshold, a vibration and light alarm is emitted through the vibration motor and the LED indicator. By introducing a reference gait and comparing it with the identified gait in real time, it can not only help patients better master the correct gait, but also effectively prevent potential risks caused by gait abnormalities, thereby accelerating the rehabilitation process.

[0052] On the other hand, this specification provides a control method for a lower limb fracture postoperative rehabilitation smart training shoe, comprising the steps of: The data processing module periodically reads the detection value of the pressure sensing module; The data processing module periodically collects and stores the pressure data, and dynamically tracks the weighing results to generate gait analysis data; The data processing module obtains the body weight according to the pressure data, and obtains the plantar equivalent body weight distribution sequence data according to the pressure data and body weight within a preset time period; The data processing module aligns the plantar equivalent weight distribution sequence data and the inertia data according to the time axis to use them as gait analysis data; The terminal analysis module performs three-dimensional gait reconstruction according to the gait analysis data to obtain three-dimensional gait data, and generates a training result report according to the three-dimensional gait data.

[0053] The method in which the data processing module obtains the plantar equivalent weight distribution sequence data according to the pressure data and weight within a preset time period includes: Reading a pre-established reference point set, wherein the reference point set is a set of pre-set plantar reference points 16; Obtaining the equivalent pressure of the plantar reference point 16 according to the positional relationship between the pressure sensor 15 and the reference point set, and the pressure data; The equivalent pressure is converted into a proportion of body weight, and the plantar equivalent body weight distribution sequence data is obtained according to the proportion.

[0054] The method for the data processing module to obtain the equivalent pressure of the plantar reference point 16 according to the positional relationship between the pressure sensor 15 and the reference point set and the pressure data includes: Extracting the amplitude of the frequency component above the preset frequency of the pressure data, and recording it as the high-frequency amplitude; The high frequency amplitude is multiplied by a preset magnification and then superimposed on a frequency component lower than a preset frequency to obtain secondary pressure data; According to the positional relationship between the pressure sensor 15 and the reference point set, a force relationship model is established, wherein the input of the force relationship model is the pressure value at the position of the pressure sensor 15, and the output is the pressure value of the sole reference point 16; After the force relationship model is trained using manually annotated samples, the equivalent pressure of the plantar reference point 16 is obtained according to the response of the force relationship model to the secondary pressure data.

Claims

1. A smart training shoe for rehabilitation after lower limb fracture surgery, characterized in that: include: A shoe body, wherein the shoe body is provided with an adjustable fixing strap and a load-bearing panel; A pressure sensing module is disposed in the front area and the rear area of ​​the load-bearing panel, and is used to detect pressure and obtain pressure data; A data processing module, connected to the pressure sensing module, periodically collects and stores the pressure data, and dynamically tracks the weighing results to generate gait analysis data; The terminal analysis module is located outside the shoe body, establishes a communication connection with the data processing unit and receives the gait analysis data reported by the data processing module, and generates training instruction data according to the gait analysis data.

2. The intelligent training shoe for rehabilitation after lower limb fracture surgery according to claim 1, characterized in that: The pressure sensor module includes pressure sensors and a weighing circuit respectively arranged in the front area and the rear area of ​​the load-bearing panel, the weighing circuit includes a signal input terminal SIG- and a signal input terminal SIG+, the signal input terminal SIG- and the signal input terminal SIG+ are grounded via capacitors C40 and C47 respectively, a capacitor C44 is connected between the signal input terminal SIG- and the signal input terminal SIG+, the anodes of the Zener diodes D24 and D25 are respectively connected to the signal input terminal SIG- and the signal input terminal SIG+, and the cathodes of the Zener diodes D24 and D25 are both connected to The signal input terminal SIG- is connected to the first input pin of the weighing chip U17 via a resistor R72, the signal input terminal SIG+ is connected to the second input pin of the weighing chip U17 via a resistor R77, capacitors C45 and C46 are connected between the first input pin and the second input pin of the weighing chip U17, the first input pin and the second input pin of the weighing chip U17 are grounded via capacitors C41 and C48 respectively, and the CS pin, SCLK pin, DIN pin, DOUT / DRDY pin and DRDY pin of the weighing chip U17 are connected to designated signal terminals via resistors respectively.

3. The intelligent training shoe for rehabilitation after lower limb fracture surgery according to claim 1 or 2, characterized in that: When the data processing module dynamically tracks the weighing results, the following steps are performed: Periodically reading the pressure data, determining the effective weighing time according to the pressure data, and discarding the pressure data outside the effective weighing time; According to the most recent weight calibration data, the pressure data is converted into percentage data; Filtering the percentage data using a filter; A weighing result is obtained according to the filtered percentage data and the weight calibration data.

4. The intelligent training shoe for rehabilitation after lower limb fracture surgery according to claim 3, characterized in that: According to the most recent weight calibration data, the method of converting the pressure data into percentage data includes: The percentage value is calculated according to the following formula: Percentage value = (percentage change / pressure detection value change) × pressure detection value + nonlinear compensation value, wherein the percentage change is the percentage value change between loading the preset weight and no load, the pressure detection value change is the pressure detection value change between loading the preset weight and no load, and the pressure detection value is the pressure detection value in the converted pressure data; The nonlinear compensation value is calculated by the following formula: Non-linear compensation value = (no-load percentage value / pressure detection value after loading preset weights - percentage value after loading preset weights / no-load pressure detection value) × change in pressure detection value.

5. The intelligent training shoe for rehabilitation after lower limb fracture surgery according to claim 1 or 2, characterized in that: The pressure sensing module includes a plurality of pressure sensors disposed in the front and rear areas of the sole, and the data processing module generates gait analysis data according to the pressure data. The terminal analysis module performs three-dimensional gait reconstruction according to the gait analysis data to obtain three-dimensional gait data, and generates a training result report according to the three-dimensional gait data; When the data processing module generates gait analysis data, the following steps are performed: Obtaining body weight according to the pressure data, and obtaining plantar equivalent body weight distribution sequence data according to the pressure data and body weight within a preset time period; The plantar equivalent weight distribution sequence data are aligned according to the time axis and used as gait analysis data.

6. The intelligent training shoe for rehabilitation after lower limb fracture surgery according to claim 5, characterized in that: The method for obtaining the plantar equivalent weight distribution sequence data according to the pressure data and the weight within a preset time period includes: Reading a pre-established reference point set, wherein the reference point set is a collection of pre-set plantar reference points; Obtaining an equivalent pressure of the plantar reference point according to the positional relationship between the pressure sensor and the reference point set, and the pressure data; The equivalent pressure is converted into a proportion of body weight, and the plantar equivalent body weight distribution sequence data is obtained according to the proportion.

7. The intelligent training shoe for rehabilitation after lower limb fracture surgery according to claim 6, characterized in that: According to the positional relationship between the pressure sensor and the reference point set, and the pressure data, a method for obtaining the equivalent pressure of the plantar reference point comprises: Reading the pressure data within a preset time period; Extracting the amplitude of the frequency component above the preset frequency of the pressure data, and recording it as the high-frequency amplitude; The high frequency amplitude is multiplied by a preset magnification and then superimposed on a frequency component lower than a preset frequency to obtain secondary pressure data; According to the positional relationship between the pressure sensor and the reference point set, a force relationship model is established, wherein the input of the force relationship model is the pressure value at the position of the pressure sensor, and the output is the pressure value of the sole reference point; After the force relationship model is trained using manually annotated samples, the equivalent pressure of the plantar reference point is obtained according to the response of the force relationship model to the secondary pressure data.

8. The intelligent training shoe for rehabilitation after lower limb fracture surgery according to claim 7, characterized in that: The method for performing three-dimensional gait reconstruction according to the gait analysis data comprises: Establishing a three-dimensional gait model, wherein the input of the three-dimensional gait model is the gait analysis data, and the output is a number of values ​​of sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle and their probabilities; Using gait analysis data with manually labeled sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle as sample data to train the three-dimensional gait model; According to the time axis, a plurality of the gait analysis data within a preset time length are read, input into the three-dimensional gait model, and a plurality of sagittal plane flexion angles, coronal plane rotation angles, horizontal plane rotation angles and their probabilities at each time point on the time axis are obtained; Obtain multiple combinations of sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle along the time axis, and calculate the total probability of each combination; Compare each combination with the preset gait library to obtain the closest gait and similarity; According to the maximum weighted sum of the total probability and the similarity, a gait is selected as the identified gait, and the sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle of the identified gait are used as the three-dimensional gait reconstruction result within the preset time length.

9. The intelligent training shoe for rehabilitation after lower limb fracture surgery according to claim 8, characterized in that: The methods for presetting gait libraries include: Obtain the historical combination, manually annotate gaits and their confidences, and obtain a gait library; After obtaining the gait and similarity of the closest combination, modifying the similarity using the confidence; or, The methods for presetting gait libraries include: Additional sensors for measuring sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle are installed to obtain gait analysis data of the patient walking within a preset time period, as well as the corresponding sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle; The sagittal plane flexion angle, the coronal plane rotation angle, and the horizontal plane rotation angle are associated with corresponding gait analysis data to obtain the gait, and then obtain a gait library.

10. The intelligent training shoe for rehabilitation after lower limb fracture surgery according to claim 5, characterized in that: The data processing module also receives a reference gait, and when the difference in sagittal plane flexion angle, coronal plane rotation angle, and horizontal plane rotation angle between the identified gait and the reference gait exceeds a preset threshold, a vibration and light alarm is emitted through the vibration motor and LED indicator light.

11. A control method for the intelligent training shoe for rehabilitation after lower limb fracture surgery according to any one of claims 1 to 10, characterized in that: Includes steps: The data processing module periodically reads the detection value of the pressure sensing module; The data processing module periodically collects and stores the pressure data, and dynamically tracks the weighing results to generate gait analysis data; The data processing module obtains the body weight according to the pressure data, and obtains the plantar equivalent body weight distribution sequence data according to the pressure data and the body weight within a preset time period; The data processing module aligns the plantar equivalent weight distribution sequence data according to the time axis as gait analysis data; The terminal analysis module performs three-dimensional gait reconstruction according to the gait analysis data to obtain three-dimensional gait data, and generates a training result report according to the three-dimensional gait data.

12. A control method according to claim 11, characterized in that: The method for the data processing module to obtain the plantar equivalent weight distribution sequence data according to the pressure data and weight within a preset time period includes: Reading a pre-established reference point set, wherein the reference point set is a collection of pre-set plantar reference points; Obtaining an equivalent pressure of the plantar reference point according to the positional relationship between the pressure sensor and the reference point set, and the pressure data; The equivalent pressure is converted into a proportion of body weight, and the plantar equivalent body weight distribution sequence data is obtained according to the proportion.

13. A control method according to claim 12, characterized in that: The method for the data processing module to obtain the equivalent pressure of the plantar reference point according to the positional relationship between the pressure sensor and the reference point set and the pressure data includes: Extracting the amplitude of the frequency component above the preset frequency of the pressure data, and recording it as the high-frequency amplitude; The high frequency amplitude is multiplied by a preset magnification and then superimposed on a frequency component lower than a preset frequency to obtain secondary pressure data; According to the positional relationship between the pressure sensor and the reference point set, a force relationship model is established, wherein the input of the force relationship model is the pressure value at the position of the pressure sensor, and the output is the pressure value of the sole reference point; After the force relationship model is trained using manually annotated samples, the equivalent pressure of the plantar reference point is obtained according to the response of the force relationship model to the secondary pressure data.

Images