Shape memory spine corrector and spine detection and correction method

Abstract

The invention provides a shape memory spine corrector and a spine detection and correction method. The shape memory spine corrector comprises a correction support, a sensing system, a heating module and a terminal. The correction support is obtained by integrally printing a shape memory polymer and is used for dynamically correcting the spine through thermal response deformation; the sensing system and the heating module are both embedded in the correcting bracket; the sensing system is used for collecting multi-source posture data of a user and sending the received multi-source posture data to the terminal; and the terminal is used for performing spine posture analysis according to the received multi-source posture data to obtain a spine posture and a correction strategy so as to control the heating module to complete correction according to the correction strategy. According to the scheme, the intelligent, personalized and real-time human body posture correcting and monitoring function is achieved, and the defects of a traditional brace in the aspects of comfort, feedback and adaptability are overcome.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, particularly to the field of medical rehabilitation device technology, and especially to a shape memory spinal orthosis and a spinal detection and correction method. Background Technology

[0002] Scoliosis is a common spinal deformity. If left untreated, it can lead to structural spinal deformities, affecting appearance and even causing dysfunction in multiple systems, including the respiratory and circulatory systems. Currently, corrective braces are widely used in conservative clinical treatment, relying primarily on long-term external force to prevent further curvature. However, traditional braces are typically made of rigid thermoplastic materials, molded using molds, resulting in bulky structures, lack of flexibility, difficulty in adapting to the complex body shapes of different individuals, poor wearing experience, and low compliance. Furthermore, these devices generally lack posture sensing capabilities, making it impossible for doctors and users to monitor the corrective effect in real time. Assessment often relies on periodic X-rays, which is not only time-consuming but also carries delays and radiation risks. Therefore, there is an urgent need to develop a shape-memory spinal orthosis and a spinal detection and correction method. Summary of the Invention

[0003] This invention provides a shape memory spinal orthosis and a spinal detection and correction method, which realizes real-time monitoring and intelligent adjustment of the spine, and improves the scientificity and portability of scoliosis correction.

[0004] In a first aspect, the present invention provides a shape memory spinal orthodontic device, comprising: an orthodontic bracket, a sensing system, a heating module, and a terminal;

[0005] The corrective frame is integrally printed from shape memory polymer and is used to dynamically correct the spine through thermally responsive deformation.

[0006] Both the sensing system and the heating module are embedded inside the orthopedic brace; the sensing system is used to collect multi-source body posture data of the user and send the received multi-source body posture data to the terminal;

[0007] The terminal is used to perform spinal posture analysis based on the received multi-source body posture data to obtain spinal posture and correction strategy, so as to control the heating module to complete the correction according to the correction strategy.

[0008] Preferably, the sensing system includes a sensing module, a processing module, and a communication module connected in sequence; the sensing module is used to collect multi-source body posture data of the user, the processing module is used to preprocess the multi-source body posture data and transmit it to the communication module; the communication module is used to send the received body posture data to the terminal.

[0009] Preferably, the corrective frame includes a corrective part and a supporting part; the corrective part is used to dynamically correct the spine through deformation, and the supporting part has a porous structure.

[0010] More preferably, the thermal response temperature of the shape memory polymer is 40–60°C.

[0011] Preferably, the sensing module includes a piezoelectric sensor, a triaxial accelerometer, and a triaxial gyroscope; the piezoelectric sensor is located at key vertebrae and key joints on both sides of the user's spine.

[0012] Preferably, the communication module uses wireless communication.

[0013] Preferably, it further includes an energy storage module and a rectifier circuit; the piezoelectric sensor is connected to the rectifier circuit and the energy storage module to provide power to the shape memory spinal orthodont.

[0014] Preferably, the piezoelectric sensor is used to collect the pressure distribution of the user on the orthodontic bracket; the triaxial accelerometer and the triaxial gyroscope are used to collect the user's dynamic motion behavior data.

[0015] Secondly, the present invention provides a method for detecting and correcting a spine, employing the shape memory spine corrector described in any of the above aspects, comprising:

[0016] Real-time acquisition of multi-source body posture data from users;

[0017] Feature extraction is performed on the multi-source body data to obtain temporal features;

[0018] The temporal features are input into a pre-trained detection model, which outputs the current spinal posture and the current correction parameters.

[0019] Based on the current correction parameters, a correction strategy is determined that includes the deformation amount of the shape memory spinal orthodont, so as to adjust the shape memory spinal orthodont.

[0020] Preferably, the step of extracting features from the multi-source body posture data to obtain time-series features includes:

[0021] The multi-source body data is segmented in chronological order using a sliding time window method to obtain several multi-source time-series data; wherein, there is overlap between adjacent multi-source time-series data.

[0022] Feature extraction is performed on the multi-source time-series data to obtain feature vectors; wherein, the feature vectors include time-domain feature vectors, frequency-domain feature vectors, and attitude feature vectors;

[0023] The time-series features are obtained by standardizing the feature vectors.

[0024] Preferably, the pre-trained detection model is trained using the following method:

[0025] Obtain the user's historical multi-source posture data and the corresponding real spinal posture and real correction parameters, and use the historical multi-source posture data and the corresponding real spinal posture and real correction parameters as a sample set;

[0026] Feature extraction is performed on the historical multi-source body shape data to obtain historical time-series features;

[0027] The historical time-series features are input into the detection model to obtain the predicted probability and predicted correction parameters of each spinal posture, and the predicted spinal posture and predicted correction parameters are output.

[0028] A loss function is constructed based on the predicted probability of each spinal posture, the predicted correction parameters, the actual spinal posture, and the actual correction parameters.

[0029] The detection model is trained using the loss function and the sample set to obtain the pre-trained detection model.

[0030] Preferably, the corrective strategy includes:

[0031] The shape memory spinal orthodont is heated to the thermal response temperature, and the shape memory spinal orthodont is dynamically controlled to achieve the specified deformation amount.

[0032] Thirdly, embodiments of the present invention also provide a computing device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the method described in any of the second aspects of this specification.

[0033] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the methods described in any of the second aspects of this specification.

[0034] Fifthly, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the steps of the method described in any second aspect of this specification.

[0035] This invention provides a shape memory spinal orthosis and a spinal detection and correction method. The shape memory spinal orthosis is made of shape memory polymer and can achieve controllable deformation and recovery under temperature excitation, thereby continuously applying adaptive corrective force to the user's spine. Simultaneously, by integrating a sensing system and a heating module within the orthosis, it can not only receive multi-source postural data from the user but also send this data to a terminal for real-time spinal posture detection and provide corresponding correction strategies. The heating module then dynamically corrects the orthosis based on these strategies. Thus, this invention achieves real-time monitoring and intelligent adjustment of the spine, improving the scientific rigor and portability of scoliosis correction. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a schematic diagram of the structure of a shape memory spinal orthodontic device according to an embodiment of the present invention;

[0038] Figure 2 This is a schematic diagram of the structure of another shape memory spinal orthosis provided in an embodiment of the present invention;

[0039] Figure 3 This is a flowchart of a spinal detection and correction method provided in an embodiment of the present invention;

[0040] Figure 4 This is a hardware architecture diagram of a computing device provided in an embodiment of the present invention;

[0041] Reference numerals: 10-Correction bracket; 20-Sensing system; 30-Heating module; 40-Terminal; 101-Correction part; 102-Support part. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0043] The following describes the specific implementation of the concept in this application.

[0044] Please refer to Figure 1 This invention provides a shape memory spinal orthodontic device, comprising: an orthodontic bracket 10, a sensing system 20, a heating module 30, and a terminal 40;

[0045] The corrective frame 10 is integrally printed from shape memory polymer and is used to dynamically correct the spine through thermally responsive deformation.

[0046] The sensing system 20 and the heating module 30 are both embedded inside the orthopedic bracket 10; the sensing system 20 is used to collect the user's multi-source body posture data and send the received multi-source body posture data to the terminal 40;

[0047] Terminal 40 is used to perform spinal posture analysis based on the received multi-source body posture data to obtain spinal posture and correction strategy, and control heating module 30 to complete correction according to the correction strategy.

[0048] In this embodiment of the invention, the shape memory spinal orthosis is made of shape memory polymer, enabling controllable deformation and recovery under temperature excitation, thereby continuously applying adaptive corrective force to the user's spine. Simultaneously, by integrating a sensing system and a heating module within the orthotic frame, it can not only receive multi-source postural data from the user but also send this data to a terminal for real-time spinal posture detection and provide corresponding correction strategies. The heating module then dynamically corrects the orthotic frame according to these strategies. Thus, this invention achieves real-time monitoring and intelligent adjustment of the spine, improving the scientific rigor and portability of scoliosis correction.

[0049] In this invention, 4D printing technology is used for integrated molding, supporting multi-material printing path planning. The structural integration of the correction bracket, sensor system embedding channel, heating module embedding channel, and signal path is achieved through collaborative printing of multiple materials or the same material, ensuring high integration and manufacturing consistency. It should be noted that the 4D-printed correction bracket consists of an upper cover and a lower shell, which are connected by snaps or screws to form a hollow internal cavity for embedding the sensor system, heating module, and signal path.

[0050] It should be noted that terminals include, but are not limited to, mobile terminals, and mobile terminals may also have a visual interface.

[0051] In a preferred embodiment, the sensing system includes a sensing module, a processing module, and a communication module connected in sequence; the sensing module is used to collect multi-source body posture data of the user, the processing module is used to preprocess the multi-source body posture data and transmit it to the communication module; the communication module is used to send the received body posture data to the terminal.

[0052] Specifically, the processing module is used to receive multi-source body state data sent by the sensing module and transmit it to the communication module after preprocessing; the communication module is used to send the preprocessed multi-source body state data to the terminal.

[0053] In a preferred embodiment, such as Figure 2 As shown, the corrective frame 10 includes a corrective part 101 and a supporting part 102; the corrective part 101 is used to dynamically correct the spine through deformation, and the supporting part 102 is a porous structure.

[0054] In this invention, the support portion has a porous cellular structure to achieve lightweight and breathability, and to form a mechanical gradient structure, improving fit and reducing local pressure. Furthermore, the correction portion and the support portion are preferably made of the same shape memory polymer material and are integrally formed by 4D printing. Structural distinctions are made in different areas according to different functions, with these distinctions completed during digital modeling.

[0055] In a more preferred embodiment, the thermal response temperature of the shape memory polymer is 40–60°C (e.g., 40°C, 45°C, 50°C, 55°C, or 60°C).

[0056] In this invention, the shape memory polymer is a thermoresponsive material, such as shape memory poly-L-lactic acid or shape memory polyurethane, with an activation temperature range of 40°C to 60°C. It possesses reversible shape recovery capabilities, allowing for automatic fit and continuous corrective force output during wear, activated by body temperature or auxiliary heating. Furthermore, the deformation path of the shape memory polymer is pre-set via 4D printing, enabling precise intervention for different scoliosis angles. In addition, the low activation temperature range of this shape memory polymer further reduces the energy consumption required for heating, laying the foundation for self-powered shape memory spinal orthotics.

[0057] In a preferred embodiment, the sensing system includes a sensing module, a processing module, and a communication module connected in sequence; the sensing module is used to collect multi-source body posture data of the user, the processing module is used to preprocess the multi-source body posture data and transmit it to the communication module; the communication module is used to send the received body posture data to the terminal.

[0058] Specifically, the processing module receives multi-source body posture data sent by the sensing module, preprocesses it, and then transmits it to the communication module; the communication module sends the preprocessed multi-source body posture data to the terminal. It should be noted that the sensing module is positioned along the user's spinal force path to obtain rich spinal posture data; the multi-source body posture data includes posture data acquired by various types of sensors. It should also be noted that the processing module has a built-in microcontroller unit with low-power periodic wake-up functionality.

[0059] In a preferred embodiment, the sensing module includes a piezoelectric sensor, a triaxial accelerometer, and a triaxial gyroscope; the piezoelectric sensor is located at key vertebrae and key joints on both sides of the user's spine.

[0060] It should be noted that key joints include the scapula, sacrum, and hip bones.

[0061] In a preferred embodiment, a piezoelectric sensor is used to collect the pressure distribution of the user on the orthopedic brace; a triaxial accelerometer and a triaxial gyroscope are used to collect the user's dynamic motion behavior data.

[0062] It should be noted that the multi-source body posture data includes voltage data collected from multiple points by several piezoelectric sensors, acceleration data collected by a triaxial accelerometer during user activity, and angular acceleration data collected by a triaxial gyroscope during user activity. This allows for the analysis of voltage data to determine the degree of force imbalance on both sides of the user's spine, and the analysis of acceleration and angular acceleration data to determine the dynamic posture changes and three-dimensional posture angles of the user's spine. More specifically, the sensing module may also include deformation sensors for collecting local deformation of the corrective brace; each sensor continuously collects data at a fixed sampling rate to form a multi-channel time series, enabling real-time monitoring and correction of the user's spinal morphology.

[0063] In a preferred embodiment, the communication module employs wireless communication.

[0064] Specifically, the communication module includes, but is not limited to, using the Bluetooth 5.0 low-power communication protocol to connect to the terminal, which can reduce the complexity of wiring.

[0065] In a preferred embodiment, it further includes an energy storage module and a rectifier circuit; the piezoelectric sensor is connected to the rectifier circuit and the energy storage module to provide power to the shape memory spinal orthodont.

[0066] In this invention, the piezoelectric sensor, energy storage module, and rectifier circuit enable self-powered operation of each component in the shape memory spinal orthodon, achieving continuous operation without external power supply and exhibiting high stability and safety. It should be noted that the sensing system and heating module are connected via wires.

[0067] In a more preferred embodiment, an external power supply module is also included, which automatically switches to external power supply mode when the self-powered voltage is lower than a set threshold.

[0068] In this invention, the structural parameters of the shape memory spinal orthosis can be adjusted according to the user's spinal morphology and pathological characteristics, thus making it suitable for different scenarios such as adolescent idiopathic scoliosis, adult postural abnormalities, and postoperative rehabilitation assistance. It can also be expanded to include various sensor types such as temperature, acceleration, and strain to achieve composite monitoring.

[0069] like Figure 3 As shown, the present invention also provides a spinal detection and correction method, applied to a terminal, comprising:

[0070] Step 300: Acquire the user's multi-source body posture data in real time;

[0071] Step 302: Extract features from the multi-source body shape data to obtain temporal features;

[0072] Step 304: Input the temporal features into the pre-trained detection model and output the current spinal posture and current correction parameters;

[0073] Step 306: Based on the current correction parameters, determine a correction strategy that includes the deformation amount of the shape memory spinal orthodont in order to adjust the shape memory spinal orthodont.

[0074] In this embodiment of the invention, feature extraction is performed on multi-source body posture data acquired in real time to obtain temporal features. These temporal features are then input into a pre-trained detection model, which outputs the current spinal posture and current correction parameters. This allows for the determination of the deformation amount of the shape memory spinal orthosis and the corresponding correction strategy, enabling the orthosis to be adjusted according to the corrective strategy. In this way, real-time detection of spinal posture and timely adjustment of the orthosis can be achieved, thereby improving the user experience.

[0075] The following description Figure 3 The execution method of each step is shown.

[0076] First, for step 300, the multi-source body posture data includes voltage data collected at multiple points by several piezoelectric sensors in the shape memory spinal orthodon, acceleration data collected by a triaxial accelerometer during user activity, angular acceleration data collected by a triaxial gyroscope during user activity, etc.

[0077] In step 302, feature extraction is performed on the multi-source body data to obtain temporal features, including:

[0078] The multi-source morphological data was segmented in chronological order using a sliding time window method to obtain several multi-source time series data; among them, there is overlap between adjacent multi-source time series data.

[0079] Feature extraction is performed on multi-source time-series data to obtain feature vectors; among which, feature vectors include time-domain feature vectors, frequency-domain feature vectors, and attitude feature vectors;

[0080] The feature vectors are standardized to obtain the time series features.

[0081] In this invention, to extract effective features, continuous multi-source body posture data is divided into time windows of fixed length. Specifically, a sliding time window method is used to segment the multi-source body posture data into several multi-source time-series data points, thereby ensuring data continuity and temporal characteristics. For example, the window length is T. w seconds, sampling frequency is f s The number of sampling points within the window is N = T w ×f s The step size of the sliding time window is T. s If the time difference is 1 second, then the start times of adjacent multi-source time series data will differ by one step.

[0082] Specifically, in this invention, feature extraction is performed on multi-source time-series data to obtain a feature vector, including:

[0083] Calculate the mean, standard deviation, maximum and minimum values ​​of each type of data in multi-source time series data to obtain several time-domain feature vectors;

[0084] Fast Fourier Transform is used to transform various types of data in multi-source time-series data into frequency domain data. The maximum power frequency and total frequency domain capability of the frequency domain data are calculated to obtain several frequency domain feature vectors.

[0085] Based on acceleration and angular velocity data from multi-source time-series data, complementary filtering is used to calculate pitch, roll, and yaw angles, resulting in several attitude feature vectors.

[0086] Specifically, to eliminate the dimensional differences between different sensors, all extracted feature vectors are standardized to obtain time-series features, including:

[0087] For each feature vector, the following steps are performed: The difference between the feature vector and the mean of the data corresponding to that feature vector is calculated to obtain the difference; the ratio of this difference to the standard deviation of the data corresponding to that feature vector is used as the time-series feature; the time-series feature is determined by the following formula:

[0088]

[0089] Where, x i ′ is used to characterize the i-th temporal feature; x i Used to characterize the i-th eigenvector; μ i The mean value used to characterize the data corresponding to the i-th feature vector (e.g., voltage data); σ iThe standard deviation used to characterize the data corresponding to the i-th eigenvector.

[0090] In this invention, since each sliding time window collects data from each sensing module within that time period, each time period contains multiple features. By extracting multiple temporal features, redundant information can be removed, and the current posture of the spine can be identified more accurately.

[0091] In a preferred embodiment, for step 304, the pre-trained detection model is trained using the following method:

[0092] Obtain the user's historical multi-source posture data and the corresponding real spinal posture and real correction parameters, and use the historical multi-source posture data and the corresponding real spinal posture and real correction parameters as a sample set;

[0093] Feature extraction is performed on historical multi-source body morphology data to obtain historical time-series features;

[0094] Historical time-series features are input into the detection model to obtain the predicted probability and predicted correction parameters of each spinal posture, and the predicted spinal posture and predicted correction parameters are output.

[0095] A loss function is constructed based on the predicted probability, predicted correction parameters, actual spinal posture, and actual correction parameters for each spinal posture.

[0096] The detection model is trained using a loss function and a sample set to obtain a pre-trained detection model.

[0097] It should be noted that the spinal pose corresponding to the maximum predicted probability is defined as the predicted spinal pose. The sample set includes historical temporal features as input and the true spinal pose and true correction parameters corresponding to the historical temporal features as output.

[0098] In a more preferred embodiment, the loss function is determined by the following formula:

[0099]

[0100] Where L is the loss value; K is the total number of spine pose categories; y k p represents the true state of the k-th spinal posture. k S[k][j] represents the predicted probability of the k-th spinal posture; S[k][j] represents the similarity between spinal posture k and spinal posture j; p j Let y be the predicted probability of the j-th spinal posture; N is the number of samples; i These are the actual correction parameters; To predict correction parameters.

[0101] It should be noted that, for example, spinal posture includes five categories: S0: normal; S1: mild scoliosis (Cobb angle < 10°); S2: moderate scoliosis (Cobb angle 10–25°); S3: severe scoliosis (Cobb angle > 25°); and S4: abnormal posture (such as excessive unilateral load, improper bracing, etc.). Therefore, K = 5. Taking k = 1, corresponding to spinal posture S0, as an example; k = 2, corresponding to spinal posture S1, and so on. If the predicted spinal posture is S2, then the true state of that spinal posture is true, and y3 = 1; the true state of other spinal postures is false, and y1 = y2 = y4 = y5 = 0.

[0102] It should be noted that S is a similarity matrix, 0 ≤ S[k][j] ≤ 1, S[k][k] = 1, and the higher the similarity, the larger the value. Continuing from the previous example, S ∈ R 5×5 And S[2][3]>S[2][4]. Since the spinal conditions of S1 and S2 are clinically similar and are mildly abnormal; the spinal conditions of S3 and S4 are clinically similar and are severely abnormal, S[2][3]>S[3][4] can be further limited. Preferably, the detection model uses a lightweight temporal deep learning model BiLSTM combined with a classifier.

[0103] In this invention, since there are clinical similarities among some spinal conditions, the loss is adjusted by introducing a state similarity matrix, so that the predicted spinal condition is more likely to be the true spinal condition, and the model learns a classification boundary that is more in line with clinical logic, thereby achieving accurate prediction.

[0104] In this invention, the method also has online learning capabilities, which can be dynamically optimized through continuously collected user posture data to improve the model's adaptability to individual users and update the correction strategy.

[0105] For step 306, the corrective strategies include:

[0106] The shape memory spinal orthodont is heated to the thermal response temperature, and the shape memory spinal orthodont is dynamically controlled to achieve the required deformation.

[0107] Specifically, in order to precisely control the heating temperature, the heating current of the heating module is determined by the following formula:

[0108]

[0109] Where u(t) is the heating current at time t; K p For proportional gain; K i K is the integral gain; dLet be the differential gain; e(t) is the difference between the deformation amount in the correction strategy and the current deformation amount at time t. It should be noted that the current deformation amount can be obtained in real time using a shape memory spinal orthosis.

[0110] In this invention, the shape memory spinal orthosis continuously collects deformation data from the deformation sensor, updates the difference e(t) in real time, and continuously adjusts the correction strategy to achieve personalized closed-loop correction, ensuring that the wearer's posture is accurately monitored and effectively adjusted in real time.

[0111] In a preferred embodiment, the mobile terminal also supports automatic recording and cloud-synchronized updates of user behavior logs.

[0112] This invention provides a shape memory spinal orthosis with human body detection function. By introducing shape memory polymer material into the orthosis drive structure, it achieves flexible, programmable, and reversible orthotic force output. It can actively adjust the shape of the brace according to body temperature or external heat sources, making the orthosis process more comfortable and compliant, significantly improving user compliance. Simultaneously, by integrating piezoelectric sensors into key structural components, utilizing the characteristic of piezoelectric materials generating electrical signals under stress, it achieves posture detection without external power supply, greatly improving the system's lightweight and stability. Combined with machine learning algorithms, it performs real-time processing and posture recognition of the collected biomechanical signals (i.e., multi-source postural data), enabling intelligent monitoring of user posture changes, early warning of abnormal behavior, and dynamic optimization of the orthotic strategy. In particular, this invention uses 4D printing technology to manufacture the entire orthosis using multiple materials in a single integrated form, effectively improving structural continuity and functional integration, avoiding stress concentration and adaptation difficulties associated with traditional modular structures, and possessing greater customization capabilities and the potential for rapid mass production. In summary, this invention combines the advantages of flexible actuation, self-powered sensing, intelligent feedback, and structural integration. It is significantly innovative in terms of structural design, functional performance, and manufacturing process, and can effectively make up for the shortcomings of existing spinal correction devices. It has good clinical application prospects and industrialization value.

[0113] The spinal detection and correction method provided in this invention can be implemented through software, hardware, or a combination of both. From a hardware perspective, such as… Figure 4 The diagram shown is a hardware architecture diagram of a computing device housing a spinal detection and correction device provided in an embodiment of the present invention. (Except for...) Figure 4 In addition to the processor, memory, network interface, and non-volatile memory shown, the computing device in the embodiment may also include other hardware, such as a forwarding chip responsible for processing messages.

[0114] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on a shape memory spinal orthodontic device. In other embodiments of the present invention, a shape memory spinal orthodontic device may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0115] The information interaction and execution process between the modules in the above-mentioned device are based on the same concept as the method embodiment of the present invention, and the specific details can be found in the description of the method embodiment of the present invention, and will not be repeated here.

[0116] This invention also provides a computing device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements a spinal detection and correction method according to any embodiment of this invention.

[0117] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform a spinal detection and correction method according to any embodiment of this invention.

[0118] Embodiments of this application also provide a computer program product, which includes a computer program. A processor of a computer device reads the computer program from a computer-readable storage medium and executes the computer program, causing the computer device to perform any of the spinal detection and correction methods described in the above embodiments.

[0119] Specifically, a system or apparatus equipped with a storage medium may be provided, on which software program code implementing the functions of any of the embodiments described above is stored, and the computer (or CPU or MPU) of the system or apparatus may read and execute the program code stored in the storage medium.

[0120] In this case, the program code read from the storage medium can itself implement the function of any of the above embodiments, and therefore the program code and the storage medium storing the program code constitute part of the present invention.

[0121] Examples of storage media used to provide program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer via a communication network.

[0122] Furthermore, it should be clear that not only can the program code read by the computer be executed, but also the operating system on the computer can be instructed based on the program code to perform some or all of the actual operations, thereby realizing the function of any of the embodiments described above.

[0123] Furthermore, it is understood that the program code read from the storage medium is written to the memory set in the expansion board inserted into the computer or to the memory set in the expansion module connected to the computer. Then, based on the instructions of the program code, the CPU or other components installed on the expansion board or expansion module execute some and all of the actual operations, thereby realizing the function of any of the above embodiments.

[0124] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0125] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as ROM, RAM, magnetic disk, or optical disk.

[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A shape memory spinal orthosis, characterized in that, include: Correction brace, sensing system, heating module and terminal; The corrective frame is integrally printed from shape memory polymer and is used to dynamically correct the spine through thermally responsive deformation. Both the sensing system and the heating module are embedded inside the orthopedic brace; the sensing system is used to collect multi-source body posture data of the user and send the received multi-source body posture data to the terminal; The terminal is used to perform spinal posture analysis based on the received multi-source body posture data to obtain spinal posture and correction strategy, so as to control the heating module to complete the correction according to the correction strategy.

2. The shape memory spinal orthosis according to claim 1, characterized in that, The sensing system includes a sensing module, a processing module, and a communication module connected in sequence; the sensing module is used to collect multi-source body posture data of the user; the processing module is used to preprocess the multi-source body posture data and transmit it to the communication module; the communication module is used to send the received body posture data to the terminal. And / or, The corrective frame includes a corrective part and a support part; the corrective part is used to dynamically correct the spine through deformation, and the support part has a porous structure; preferably, the thermal response temperature of the shape memory polymer is 40-60℃.

3. The shape memory spinal corrector according to claim 2, characterized in that, The sensing module includes a piezoelectric sensor, a triaxial accelerometer, and a triaxial gyroscope; the piezoelectric sensor is located at key vertebrae and key joints on both sides of the user's spine. And / or, The communication module uses wireless communication.

4. The shape memory spinal corrector according to claim 3, characterized in that, It also includes an energy storage module and a rectifier circuit; the piezoelectric sensor is connected to the rectifier circuit and the energy storage module to provide power to the shape memory spinal orthodon. And / or, The piezoelectric sensor is used to collect the pressure distribution of the user on the orthopedic brace; the triaxial accelerometer and triaxial gyroscope are used to collect the user's dynamic motion behavior data.

5. A method for spinal detection and correction based on a shape memory spinal orthosis according to any one of claims 1 to 4, characterized in that, include: Real-time acquisition of multi-source body posture data from users; Feature extraction is performed on the multi-source body data to obtain temporal features; The temporal features are input into a pre-trained detection model, which outputs the current spinal posture and the current correction parameters. Based on the current correction parameters, a correction strategy is determined that includes the deformation amount of the shape memory spinal orthodont, so as to adjust the shape memory spinal orthodont.

6. The method according to claim 5, characterized in that, The step of extracting features from the multi-source body posture data to obtain temporal features includes: The multi-source body data is segmented in chronological order using a sliding time window method to obtain several multi-source time-series data; wherein, there is overlap between adjacent multi-source time-series data. Feature extraction is performed on the multi-source time-series data to obtain feature vectors; wherein, the feature vectors include time-domain feature vectors, frequency-domain feature vectors, and attitude feature vectors; The time-series features are obtained by standardizing the feature vectors.

7. The method according to claim 5, characterized in that, The pre-trained detection model is trained using the following method: Obtain the user's historical multi-source posture data and the corresponding real spinal posture and real correction parameters, and use the historical multi-source posture data and the corresponding real spinal posture and real correction parameters as a sample set; Feature extraction is performed on the historical multi-source body shape data to obtain historical time-series features; The historical time-series features are input into the detection model to obtain the predicted probability and predicted correction parameters of each spinal posture, and the predicted spinal posture and predicted correction parameters are output. A loss function is constructed based on the predicted probability of each spinal posture, the predicted correction parameters, the actual spinal posture, and the actual correction parameters. The detection model is trained using the loss function and the sample set to obtain the pre-trained detection model.

8. The method according to any one of claims 5 to 7, characterized in that, The corrective strategies include: The shape memory spinal orthodont is heated to the thermal response temperature, and the shape memory spinal orthodont is dynamically controlled to achieve the specified deformation amount.

9. A computing device comprising a memory and a processor, wherein the memory stores a computer program, and the processor, when executing the computer program, implements the method as described in any one of claims 5-8.

10. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the method of any one of claims 5-8.

Images