Body posture monitoring method, device and electronic equipment

Abstract

The present application provides a posture monitoring method, device and electronic device, which relate to the field of data processing. In the method, the bioelectric field data for the human body sent by the electric field sensor array device is obtained; according to the bioelectric field data, the first posture change data of the human body is determined; the magnetic positioning data for the human body sent by the magnetic field sensor array device is obtained; according to the magnetic positioning data, the second posture change data of the human body is determined; the first posture change data and the second posture change data are input into the preset LSTM model to generate a dynamic posture three-dimensional model of the human body, so as to monitor the posture of the human body. The implementation of the technical solution provided by the present application is convenient for real-time monitoring of the patient's human posture.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a body posture monitoring method, device and electronic equipment. Background Art

[0002] In the field of modern medical surgery, especially in minimally invasive interventional treatment, surgical accuracy and safety play a vital role in the treatment effect of patients. Interventional treatment usually requires high-precision equipment to guide surgical instruments to the target area, and slight changes in the patient's body posture, such as displacement caused by breathing, heartbeat, and muscle activity, may cause the target area to shift. This shift may not only reduce the effectiveness of treatment, but also increase surgical risks.

[0003] At present, most traditional posture monitoring methods rely on static or periodic image acquisition and cannot capture the dynamic changes of the patient's posture in real time. However, especially in treatments that require high-precision operations such as radiofrequency ablation, the inability to obtain timely dynamic changes in posture will have a significant impact on target positioning, thereby reducing surgical accuracy and safety. Therefore, the above monitoring methods are not conducive to real-time monitoring of the patient's human posture.

[0004] Therefore, there is an urgent need for a body posture monitoring method, device and electronic equipment. Summary of the invention

[0005] The present application provides a posture monitoring method, device and electronic equipment, which are convenient for real-time monitoring of a patient's human posture.

[0006] In a first aspect of the present application, a posture monitoring method is provided, the method comprising: obtaining bioelectric field data for a human body sent by an electric field sensing array device; determining first posture change data of the human body based on the bioelectric field data; obtaining magnetic positioning data for the human body sent by a magnetic field sensing array device; determining second posture change data of the human body based on the magnetic positioning data; inputting the first posture change data and the second posture change data into a preset LSTM model to generate a dynamic posture three-dimensional model of the human body to perform posture monitoring on the human body.

[0007] By adopting the above technical solutions, by combining the advantages of bioelectric field and magnetic positioning data, the patient's body changes can be fully captured. Electric field data is highly sensitive to subtle deformations on the human body surface and can accurately reflect local dynamic changes caused by breathing, muscle activity, etc. Magnetic field data provides three-dimensional positioning support for the overall position of the human body and the displacement of key points, enhancing the spatial consistency of the overall modeling. Through data fusion, it can effectively make up for the monitoring blind spots or errors that may be caused by a single technology. Through high-frequency acquisition and LSTM model reasoning, it can quickly respond to changes in the patient's body shape, realize real-time updates of dynamic three-dimensional models, and ensure reliable decision support during surgery. The long and short-term memory capabilities of the LSTM model are used to fully capture the time series characteristics of the patient's body shape changes. Compared with traditional static monitoring methods, dynamic three-dimensional body models can more accurately reflect the patient's real-time body shape. Directly generating a dynamic three-dimensional model helps doctors intuitively understand the process of changes in the patient's body shape and facilitates the formulation of accurate surgical plans. Especially in complex interventional surgeries, the dynamic update of the model can prompt the target point offset in real time. Therefore, it is convenient to monitor the patient's body shape in real time.

[0008] Optionally, the obtaining of bio-electric field data for a human body sent by an electric field sensing array device specifically includes: receiving a first electric field signal for the human body sent by the electric field sensing array device at a first timestamp, the electric field sensing array device comprising a plurality of electric field sensors circumferentially distributed on the human body in a rectangular grid; amplifying the first electric field signal according to the bio-signal characteristics of the human body using a high-gain operational amplifier to obtain a first analog electric field signal; filtering the first analog electric field signal using a filter to obtain a second analog electric field signal, the filter comprising a low-pass filter and a band-pass filter; converting the second analog electric field signal into a first digital signal; obtaining a second digital signal corresponding to the second electric field signal for the human body sent by the electric field sensing array device at a second timestamp; and determining the bio-electric field data according to the first digital signal and the second digital signal.

[0009] By adopting the above technical solution, the electric field sensing array device adopts a rectangular grid-shaped multi-sensor distribution, which can fully cover the human body and capture the electric field changes in different parts of the body. This dense layout effectively improves the spatial resolution of signal acquisition and can accurately detect small electric field changes caused by posture changes, muscle activity or physiological fluctuations. By receiving electric field signals at different timestamps, comparative analysis can be achieved to help identify and quantify the dynamic process of posture changes over time. This method ensures that signal acquisition is not limited to a certain moment, but can provide a full picture of time series changes, so as to better capture the changes in the patient's posture. Using a high-gain operational amplifier to amplify the first electric field signal can effectively increase the intensity of the electric field signal, making the weak bioelectric field changes more obvious and enhancing the ability to respond to subtle changes. This is particularly important for capturing posture fine-tuning caused by physiological activities such as breathing and heartbeat, which helps to obtain accurate dynamic posture data. Using a low-pass filter and a band-pass filter to filter the analog signal can effectively remove noise and irrelevant frequency components. This not only improves the purity of the signal, but also prevents errors caused by signal interference, thereby ensuring the reliability of the sensing data. By converting the filtered analog electric field signal into a digital signal, the processing accuracy and compatibility of the data can be improved. Digital signals are easy to process later, avoiding the distortion or noise reduction problems that may be caused by analog signals during transmission. The second electric field signal at the second timestamp can provide a comparison with the first timestamp data, which not only ensures the continuity of the data, but also reflects the trend of body posture changes over time. This processing method supports more detailed timing analysis and helps to build a more accurate dynamic body posture model.

[0010] Optionally, determining the first body shape change data of the human body according to the bioelectric field data specifically includes: determining a first signal strength corresponding to the human body at the first timestamp according to the first digital signal; determining a second signal strength corresponding to the human body at the second timestamp according to the second digital signal; and obtaining the first body shape change data corresponding to the human body by differential calculation based on the first signal strength and the second signal strength.

[0011] By adopting the above technical solution, by comparing the digital signal strength at the first timestamp and the second timestamp, the electric field changes between the two moments can be accurately captured. This method has high sensitivity and can identify small changes in posture caused by breathing, heartbeat, posture changes, etc. Compared with collecting data separately, this differential calculation method can effectively amplify the signal of posture change, reduce the influence of background noise, and provide more accurate monitoring results. The posture changes of the human body are very small, especially in complex surgery or interventional treatment, where slight movements may cause the treatment target area to shift. Through differential calculation, the recognizability of these slight changes can be enhanced, thereby improving the real-time positioning accuracy during surgery. In dynamic posture monitoring, there may be factors such as sensor noise, environmental interference or individual differences. Differential calculation provides a continuous data stream that can continuously track the posture changes of the human body. This is crucial for surgical navigation and real-time positioning, because the slight changes in the patient's position can be continuously monitored during treatment to avoid treatment deviations caused by position shifts. The posture change data obtained by differential calculation not only enhances the clarity and accuracy of the signal, but also makes the generated 3D posture model more timely and dynamic. The real-time updated 3D posture model can help doctors accurately identify any deviation of the patient's posture and adjust the treatment strategy in time.

[0012] Optionally, the acquiring of magnetic positioning data for the human body sent by the magnetic field sensing array device specifically includes: sending a magnetic field acquisition signal to a passive magnetic tag through the magnetic field sensing array device, the installation position of the passive magnetic tag on the human body is determined according to the type of surgery corresponding to the human body; acquiring a magnetic field array signal sent by the magnetic field sensing array device, the magnetic field array signal is determined by the magnetic field sensing array device receiving a magnetic field return signal sent by the passive magnetic tag; and generating the magnetic positioning data according to the magnetic field array signal.

[0013] By adopting the above technical solution, the magnetic field sensor array device sends a collection signal to the passive magnetic tag, and receives the magnetic field return signal sent back by the tag, so that the position change of the tag can be accurately captured. This signal reflection method has high accuracy in the positioning system, especially in complex surgical environments, and can provide doctors with real-time and accurate position data to ensure the positioning accuracy of the treatment target area. By receiving magnetic field signals by multiple magnetic field sensors and performing data fusion, higher-precision positioning can be achieved. The magnetic field array signal provides multi-dimensional information in space, so that positioning is not limited to changes in a single direction, but can fully capture the position information of the tag, improving the positioning accuracy and stability. Passive magnetic tags do not require batteries or external power supplies, and they only perform position calibration through the reflection of magnetic field signals. This makes it unnecessary to have any physical contact with the human body, will not cause discomfort to the patient or interfere with the surgical operation, and ensure the comfort and safety of the patient during the operation. Compared with other devices that require conduction current or contact sensing, passive tags provide a gentler monitoring method. Due to the use of passive magnetic tags, this positioning method does not rely on batteries or current, and is positioned through the response of magnetic field signals, so the potential risks caused by electromagnetic interference can be avoided. Interference to medical equipment and precision instruments is effectively suppressed, increasing safety during surgery. Through the feedback of the magnetic field sensor array, the patient's position changes can be tracked in real time to ensure the accuracy of the surgical operation.

[0014] Optionally, determining the second body posture change data of the human body according to the magnetic positioning data specifically includes: determining the magnetic field strength signal corresponding to each of the passive magnetic tags according to the magnetic positioning data; inverting a plurality of three-dimensional coordinates marked by the passive magnetic tag on the human body through a magnetic dipole model according to the magnetic field strength signal corresponding to each of the passive magnetic tags; and determining the second body posture change data corresponding to the human body according to the plurality of three-dimensional coordinates.

[0015] By adopting the above technical solution, the magnetic field strength signal of each passive magnetic tag can be inverted through the magnetic dipole model, and the three-dimensional coordinates of the tag on the human body can be accurately calculated. This method can upgrade the traditional two-dimensional positioning to three-dimensional space, making the posture change data more comprehensive and accurate. In surgery and interventional treatment, three-dimensional coordinates can more accurately reflect the patient's posture changes and avoid positioning deviations caused by relying only on plane data. Through this three-dimensional inversion method, the slight posture changes of the human body in space can be accurately captured, providing higher accuracy than traditional posture monitoring technology. This is particularly important for complex surgeries or medical treatments that require precise operations, because slight changes in body position may have a significant impact on the accuracy of the treatment target. The multi-dimensional coordinates obtained through magnetic positioning data can continuously track changes in the patient's body position and update the body change data in real time. For interventional treatment and refined surgery, the patient's slight displacement or posture changes can be monitored in time. This enables doctors to adjust the operation strategy in real time during the treatment process to ensure the accuracy of the treatment. During complex surgeries, physiological changes such as breathing and heartbeat of the patient may cause slight deviations in body position. Precise three-dimensional coordinate inversion can help doctors monitor and correct these changes to avoid affecting surgical accuracy, thereby ensuring the success rate and safety of the surgery. Magnetic positioning data can not only provide information on a single dimension of body posture changes, but also combine other sensor information such as bioelectric field data and image data to generate comprehensive three-dimensional body posture data. Through this multi-source data fusion, subtle changes in the patient's body position can be captured more comprehensively, improving the precision and accuracy of overall monitoring.

[0016] Optionally, the step of inputting the first posture change data and the second posture change data into a preset LSTM model to generate a dynamic three-dimensional body model of the human body specifically includes: determining deformation distribution data of the human body according to the first posture change data; determining three-dimensional displacement data of the human body according to the second posture change data, wherein the first posture change data and the second posture change data have the same corresponding timestamp; performing feature extraction on the deformation distribution data and the three-dimensional displacement data respectively using a feedforward neural network to obtain a first feature vector and a second feature vector; concatenating the first feature vector and the second feature vector to generate a high-dimensional vector; inputting the high-dimensional vector into the preset LSTM model to obtain three-dimensional body change parameters corresponding to the timestamp, wherein the three-dimensional body change parameters include key point skeleton parameters and body surface deformation parameters; and constructing the dynamic three-dimensional body model according to the key point skeleton parameters and the body surface deformation parameters.

[0017] By adopting the above technical solution, the posture change data corresponding to the bioelectric field data and the magnetic positioning data are fused, which can provide a more comprehensive and accurate input to describe the dynamic posture changes of the human body. This enables the model to identify small changes in the human body at different time points, especially in complex surgery or treatment processes, and can effectively avoid the deviation of the treatment target area caused by small changes in posture. By extracting deformation distribution data and three-dimensional displacement data, the structure and displacement of the human body can be described in detail, and the posture changes of the human body can be further accurately predicted in combination with the deep learning model. These data provide higher accuracy for building a dynamic posture three-dimensional model, ensuring that the model can reflect the actual dynamic changes of the human body. Using a feedforward neural network to extract features from deformation distribution data and three-dimensional displacement data respectively, key features can be extracted from complex input data. In this way, the original data can be converted into a more useful representation, so that the subsequent LSTM model can learn and predict more efficiently and accurately. The two feature vectors are spliced ​​into a high-dimensional vector, which comprehensively considers the deformation and displacement information of the posture and improves the multidimensionality of the model input. The construction of high-dimensional vectors enables the model to capture multiple aspects of human posture changes during the learning process, thereby enhancing the expression ability of the model. The key point skeleton parameters generated by the LSTM model can accurately describe the changes in the human body's skeletal structure. These parameters provide strong support for surgical navigation and real-time adjustment of patient posture. Doctors can use these parameters to obtain the patient's skeletal position in three-dimensional space, so as to adjust the operation strategy in real time and avoid errors caused by posture changes during surgery. In addition to bone position, surface deformation parameters can also reflect subtle changes in the external surface of the human body during surgery. A dynamic body posture three-dimensional model can be constructed through key point skeleton parameters and surface deformation parameters. This model is not only static, but can reflect changes in human body posture in real time. The dynamic three-dimensional model provides strong support for surgical navigation and real-time adjustment, allowing doctors to accurately grasp the patient's posture during surgery and make corresponding operations based on the changes.

[0018] Optionally, the method also includes: using a SLAM algorithm to correct the actual position of the human body and the position of the dynamic body three-dimensional model to obtain a corrected dynamic body model; marking the position of the surgical target in the corrected dynamic body model to obtain a target visualization model; and displaying the target visualization model to a surgeon corresponding to the human body through AR glasses, wherein the surgeon wears the AR glasses.

[0019] By adopting the above technical solution, the SLAM algorithm is used to correct the deviation between the actual position of the human body and the position in the dynamic body three-dimensional model. Due to the slight changes in the patient's body shape during the operation, the traditional static three-dimensional model may deviate from the actual situation. SLAM can track the patient's position in real time and adjust the dynamic body shape model through sensor data to make the model more in line with the actual situation, thereby ensuring the accuracy of surgical navigation. During the operation, the patient's body position may change slightly due to factors such as breathing and heartbeat. The SLAM algorithm can effectively track these changes and adjust the model in time to avoid the dislocation of the surgical target area due to position deviation. This is crucial for high-precision interventional treatment, ensuring that the treatment target is always consistent with the actual body position, and improving the success rate and safety of the operation. Applying the SLAM algorithm to the correction of the dynamic body shape three-dimensional model can not only ensure the accuracy of the model, but also reflect the changes in the patient's body position in real time. Any slight changes in body shape during the operation will be captured by the SLAM algorithm in time and fed back to the dynamic body shape model, thereby ensuring that the body shape information seen by the doctor is always synchronized with the actual patient's body shape, avoiding surgical operation errors caused by static data. The corrected dynamic body model can be continuously updated, and doctors can operate according to the real-time adjusted model to avoid relying on outdated or error-prone data. This dynamic correction capability allows every moment of the operation to be consistent with the most accurate body model, improving the controllability and accuracy of the operation. In the corrected dynamic body model, marking the location of the surgical target can provide doctors with intuitive and accurate target locations. Especially in complex interventional treatments, the accuracy of the target location is directly related to the treatment effect. The accurate marking of the target in the dynamic body model allows doctors to clearly understand the location and morphology of the target area during the operation, further improving the accuracy of the operation. Through precise marking and combined with AR technology, the surgical target can be presented in a three-dimensional and realistic way in the doctor's field of view, enhancing the doctor's spatial perception of the target. This allows doctors to locate the target area more clearly during the operation and avoid misoperation caused by changes in body position.

[0020] In a second aspect of the present application, a posture monitoring device is provided, which includes an acquisition module and a processing module, wherein the acquisition module is used to acquire bioelectric field data for a human body sent by an electric field sensing array device; the processing module is used to determine first posture change data of the human body based on the bioelectric field data; the acquisition module is also used to acquire magnetic positioning data for the human body sent by a magnetic field sensing array device; the processing module is also used to determine second posture change data of the human body based on the magnetic positioning data; the processing module is also used to input the first posture change data and the second posture change data into a preset LSTM model to generate a dynamic posture three-dimensional model of the human body to perform posture monitoring on the human body.

[0021] In the third aspect of the present application, an electronic device is provided, which includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device performs the method described above.

[0022] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, the method described above is executed.

[0023] In summary, one or more technical solutions provided in this application have at least the following technical effects or advantages:

[0024] By combining the advantages of bioelectric field and magnetic positioning data, the patient's posture changes can be fully captured. Electric field data is highly sensitive to subtle deformations on the human body surface and can accurately reflect local dynamic changes caused by breathing, muscle activity, etc. Magnetic field data provides three-dimensional positioning support for the overall position of the human body and the displacement of key points, enhancing the spatial consistency of the overall modeling. Through data fusion, it can effectively make up for the monitoring blind spots or errors that may be caused by a single technology. Through high-frequency acquisition and LSTM model reasoning, it can quickly respond to changes in the patient's posture, realize real-time updates of dynamic three-dimensional models, and ensure reliable decision support during surgery. The long-term and short-term memory capabilities of the LSTM model are used to fully capture the time series characteristics of the patient's posture changes. Compared with traditional static monitoring methods, dynamic three-dimensional posture models can more accurately reflect the patient's real-time posture state. Directly generating a dynamic three-dimensional model helps doctors intuitively understand the process of changes in the patient's posture and facilitates the formulation of accurate surgical plans. Especially in complex interventional surgeries, the dynamic update of the model can prompt the target point offset in real time. Therefore, it is convenient to monitor the patient's human posture in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A flow chart of a body posture monitoring method provided in an embodiment of the present application.

[0026] Figure 2 Another schematic diagram of a flow chart of a body posture monitoring method provided in an embodiment of the present application.

[0027] Figure 3 A schematic diagram of a module of a body posture monitoring device provided in an embodiment of the present application.

[0028] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0029] Explanation of the reference numerals: 31, acquisition module; 32, processing module; 41, processor; 42, communication bus; 43, user interface; 44, network interface; 45, memory. DETAILED DESCRIPTION

[0030] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0031] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0032] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0033] In the field of modern medical surgery, especially in minimally invasive interventional treatment, the accuracy and safety of the surgery are crucial to the treatment effect of the patient. Interventional treatment usually relies on high-precision equipment to guide surgical instruments to accurately reach the target area. However, slight changes in the patient's posture, such as displacement caused by breathing, heartbeat or muscle activity, may cause the target area to shift. This shift not only affects the treatment effect, but also may increase the risk of surgery.

[0034] At present, most traditional posture monitoring methods rely on static or periodic image acquisition and cannot capture the dynamic changes of the patient's posture in real time. In interventional treatments that require high-precision operations, especially surgeries such as radiofrequency ablation, the inability to obtain the dynamic changes of the patient's posture in real time will lead to inaccurate target positioning, which in turn affects the accuracy and safety of the surgery. Therefore, existing monitoring methods have obvious deficiencies in real-time tracking of patient posture changes and cannot effectively support precise operations during complex surgeries.

[0035] In order to solve the above technical problems, the present application provides a body posture monitoring method, referring to Figure 1 , Figure 1 The following is a flow chart of a posture monitoring method provided in an embodiment of the present application. The posture monitoring method is applied to a server and includes steps S110 to S150, which are as follows:

[0036] S110: Acquire bioelectric field data of a human body sent by an electric field sensor array device.

[0037] Specifically, the server refers to a computer system or server cluster used to process various data in the operating room. The electric field sensing array device is a device composed of multiple electric field sensors, which is used to capture the bioelectric field signals generated by the human body or other objects in real time. Bioelectric field data refers to weak electric field signals emitted by the human body. These signals originate from physiological activities in the human body, such as nerve signals, muscle activities, etc. These data can be used to monitor body changes, posture adjustments, etc. The electric field sensing array device detects the bioelectric field generated by the human body through a group of arranged electric field sensors. These sensors collect bioelectric field signals at different time points and transmit these signals to the server for processing. The electric field signal captured by each sensor corresponds to the electric field strength of the human body at that location.

[0038] In a possible implementation, obtaining bioelectric field data for a human body sent by an electric field sensing array device specifically includes: receiving a first electric field signal for a human body sent by the electric field sensing array device at a first timestamp, wherein a plurality of electric field sensors included in the electric field sensing array device are circumferentially distributed on the human body in a rectangular grid; amplifying the first electric field signal according to the biological signal characteristics of the human body using a high-gain operational amplifier to obtain a first analog electric field signal; filtering the first analog electric field signal using a filter to obtain a second analog electric field signal, wherein the filter includes a low-pass filter and a band-pass filter; converting the second analog electric field signal into a first digital signal; obtaining a second digital signal corresponding to the second electric field signal for a human body sent by the electric field sensing array device at a second timestamp; and determining the bioelectric field data according to the first digital signal and the second digital signal.

[0039] Specifically, first, the server receives the first electric field signal from the electric field sensing array device. The electric field sensing array device collects the electric field generated by the bioelectric activity of the human body through a group of sensors distributed on the surface of the human body. Assuming that the electric field sensing array device is placed on the patient's chest, it can sense the electric field changes caused by the patient's breathing or heartbeat. Among them, multiple electric field sensors are arranged in a grid matrix, surrounding a certain area of ​​the human body, such as the chest, abdomen, etc. Each sensor is responsible for capturing the electric field signal of a specific area. Assuming that some sensors are placed on the patient's chest, these sensors are arranged in a matrix, and the captured electric field signals can reflect the muscle activity of the chest or the electrical activity of the heart.

[0040] Secondly, the electric field signal is usually very weak, so it needs to be amplified by a high-gain operational amplifier. This helps to enhance the signal strength and make subsequent signal processing clearer. For analog signals, filters are needed to remove noise or irrelevant high-frequency components. Low-pass filters remove high-frequency noise, while band-pass filters only retain signals within a specific frequency band to ensure that the filtered signal is more consistent with the characteristics of the bioelectric field signal. For example, if there is high-frequency noise caused by muscle activity in the electrocardiogram signal, the low-pass filter will remove these irrelevant frequencies and only retain the effective signal of the electrocardiogram. After filtering, the analog signal is still a continuous analog signal. For further processing, it needs to be converted into a digital signal through an analog-to-digital converter. Digital signals are easier for the server to process and analyze. The server will also receive the electric field signal at a second time point. In the same process, the sensor obtains the second bioelectric field signal and then converts it into a digital signal. At this time, the capture of the signal is time-series, and the dynamic changes of the signal can be analyzed. For example, during the patient's breathing, the electric field signal will change over time, and the signal captured by the first timestamp and the signal of the second timestamp can reflect the changes in the patient's body posture during breathing.

[0041] Finally, the digital signals of the first and second timestamps are combined and analyzed to extract the actual bioelectric field data. This data can be used to analyze the dynamic physiological changes of the human body or evaluate the slight changes in the patient's posture during surgery. For example, by comparing the electric field signals of the two timestamps, the chest movement of the patient during inhalation and exhalation can be analyzed. This data helps doctors understand the patient's posture in real time and avoid the displacement of the treatment target due to posture changes during treatment.

[0042] S120: Determine first body posture change data of the human body according to the bioelectric field data.

[0043] Specifically, the first posture change data refers to certain physiological or physical changes of the human body within a specific time period, such as posture adjustment, muscle activity, breathing depth changes, etc. These data are helpful for real-time monitoring and analysis of the patient's posture, especially during surgery, to help doctors adjust the position of instruments. The server uses the previously collected electric field signal data to analyze and infer the first posture change of the human body through certain algorithms or computational models. These algorithms can include signal processing, pattern recognition, time series analysis and other technologies.

[0044] In a possible implementation, first body posture change data of a human body is determined based on bioelectric field data, specifically including: determining a first signal strength corresponding to a first timestamp of the human body based on a first digital signal; determining a second signal strength corresponding to a second timestamp of the human body based on a second digital signal; and obtaining the first body posture change data corresponding to the human body by differential calculation based on the first signal strength and the second signal strength.

[0045] Specifically, the first digital signal and the second digital signal represent the electric field signal data acquired at two different time stamps. The first time stamp and the second time stamp refer to the specific time points of signal acquisition. For example, the first time stamp may be the start time of the operation, and the second time stamp is a certain time after 1 second. Signal strength refers to the amplitude of the received electric field signal. In the embodiment of the present application, the first signal strength and the second signal strength respectively represent the strength of the electric field signal received at two different time stamps. For example, if the patient's breathing causes the electric field of the chest to change, the signal strength will fluctuate with the change of breathing. Differential calculation is used to compare the change of two data points. The embodiment of the present application is used to compare the signal strength difference of two time stamps. By calculating the difference between the first signal strength and the second signal strength, the change of signal strength can be determined. The signal strength change obtained by the server through differential calculation is the first posture change data, which reflects the change of the posture of the human body during this period. For example, if the patient is taking a deep breath, the strength of the electric field signal will change due to the ups and downs of breathing, and the patient's posture changes during this period, such as the depth or frequency of breathing, can be determined by differential calculation.

[0046] For example, suppose a scenario is that during radiofrequency ablation treatment, the doctor needs to accurately know the patient's posture changes, such as slight displacement of the chest, to ensure the stability of the treatment target position. Assume that at the beginning of the operation, that is, the first timestamp, the signal strength received by the electric field sensing array device is S1: 0.75, indicating the electric field strength in the patient's chest area, such as changes caused by breathing or muscle movement. One second later, that is, the second timestamp, the device receives a signal strength of S2: 0.78 again. At this time, the patient's breathing or other factors may cause the strength of the chest electric field to change, that is, the first posture change data 0.03. This change value indicates that the patient's posture has changed slightly during this period, for example, it may be due to chest electric field fluctuations caused by breathing. This data will help doctors determine whether the patient's posture is stable, so as to make adjustments to ensure the accurate positioning of the treatment target.

[0047] S130: Acquire magnetic positioning data for a human body sent by a magnetic field sensor array device.

[0048] Specifically, a magnetic field sensing array device is a device for detecting surrounding magnetic fields. It is composed of a plurality of magnetic field sensors that can detect and measure changes in magnetic fields in the environment. In an embodiment of the present application, a magnetic field sensing array device is specifically used to track and locate the human body. Magnetic positioning data refers to data related to the position and posture of the human body collected by a magnetic field sensing array device. The magnetic field sensor determines the precise position of the passive magnetic tag by detecting changes in the magnetic field. These data are used to locate or track the position and movement of an object in space, and are particularly suitable for scenarios such as dynamic posture monitoring and surgical navigation. Among them, the passive magnetic tags are attached to the human body. These tags are small magnetic elements that do not require batteries or external power supplies. Their function is to help the sensor determine its position by reflecting or sending magnetic field signals. The magnetic field sensor array sends magnetic field acquisition signals to the area where the passive magnetic tags are installed. When the tag receives these signals, it reflects or generates certain magnetic field changes. The magnetic field sensing array device calculates the precise position of the tag based on the received reflected magnetic field signal.

[0049] In a possible implementation, obtaining magnetic positioning data for a human body sent by a magnetic field sensor array device specifically includes: sending a magnetic field acquisition signal to a passive magnetic tag through the magnetic field sensor array device, and the installation position of the passive magnetic tag on the human body is determined according to the type of surgery corresponding to the human body; obtaining a magnetic field array signal sent by the magnetic field sensor array device, and the magnetic field array signal is determined by the magnetic field sensor array device receiving a magnetic field return signal sent by the passive magnetic tag; and generating magnetic positioning data according to the magnetic field array signal.

[0050] Specifically, the magnetic field sensing array device actively generates a magnetic field signal and sends it to a specific area to stimulate the passive magnetic tags located on the human body. The location design of the passive magnetic tags is determined according to the type of surgery. For example, thoracic surgery: the tags may be placed around the chest. Joint replacement surgery: the tags may be placed near the surgical site. Laparoscopic surgery: the tags may be attached to the abdomen or torso area. This design ensures that the tag location is associated with the surgical site, thereby more accurately reflecting the body posture changes in the surgical area. After receiving the magnetic field acquisition signal sent by the device, the passive magnetic tag will reflect or resend a magnetic field signal. The intensity, direction and phase of this return signal will vary depending on the specific location and direction of the tag. The magnetic field sensing array device receives the return signal through multiple sensors, so that the spatial position of the passive magnetic tag can be determined. This process is equivalent to reverse triangulation, but is based on magnetic field signals. The server preprocesses the received magnetic field array signal, such as denoising and filtering, to extract useful information. Then the magnetic dipole model or other magnetic field inversion algorithm is applied to calculate the three-dimensional coordinates of the passive magnetic tag. Among them, the magnetic positioning data includes the three-dimensional spatial position of the tag, the spatial orientation information of the tag, such as the tilt angle, and signal strength-related data, which are used to evaluate the stability and reliability of the tag.

[0051] S140: Determine second body posture change data of the human body according to the magnetic positioning data.

[0052] Specifically, each passive magnetic tag corresponds to a three-dimensional coordinate at each timestamp. At timestamp t1, the position of tag A is (10.2, 5.8, 3.4). At timestamp t2, the position of tag A is (10.5, 6.0, 3.6). The displacement of tag A between timestamps t1 and t2 is calculated to be 0.374 units. If there are multiple tags, such as tags A, B, and C, the server will integrate their displacement information to form the overall change data of different parts of the human body. For example: tag A is located at the shoulder; tag B is located at the waist; tag C is located at the abdomen. Through the change data of multiple tags, the overall posture adjustment of the patient can be inferred, such as the tilt, rotation or position movement of the torso. Based on the displacement data of all tags, the server calculates the dynamic changes of the human body as a whole, that is, the second posture change data. The data can be specifically described as: displacement direction, such as movement from left to right or front and back; displacement amplitude, for example, the change of tag A at the shoulder is 3° rotation and 5mm upward movement; relative changes between tags, such as the torsion angle of a part of the body.

[0053] For example, when the second posture change data monitored is shoulder elevation: +3mm; chest expansion angle: 5°; abdominal movement: +3mm upward, it indicates that the patient's shoulders, chest and abdomen are slightly raised and expanded outward, indicating that the patient is inhaling.

[0054] In a possible implementation, second body posture change data of the human body is determined based on the magnetic positioning data, specifically including: determining the magnetic field strength signal corresponding to each passive magnetic tag based on the magnetic positioning data; inverting multiple three-dimensional coordinates marked by the passive magnetic tags on the human body through a magnetic dipole model based on the magnetic field strength signal corresponding to each passive magnetic tag; and determining the second body posture change data corresponding to the human body based on the multiple three-dimensional coordinates.

[0055] Specifically, the magnetic field strength signal refers to the strength of the return signal generated by the passive magnetic tag by receiving the magnetic field acquisition signal. The magnetic field sensor array device detects these signals and records the strength corresponding to each passive tag. The purpose is to construct an estimate of the position of each passive magnetic tag in three-dimensional space. Passive magnetic tags will form a specific magnetic field distribution in space, and the strength and direction of the magnetic field are related to the position of the tag, the magnetic strength and the position of the sensor. The strength of the magnetic field is stronger in the area closer to the tag and gradually weakens with the increase of distance. The distribution characteristics of this magnetic field can be modeled by describing the behavior of the tag as a magnetic dipole, that is, similar to a pair of north and south micro magnets. Furthermore, the magnitude and direction of the magnetic field have a certain dependence on the distance from the tag to the measurement point. The closer the distance, the stronger the magnetic field; the farther the distance, the faster the magnetic field strength decays. The direction of the magnetic field is affected by the direction of the magnetic moment inside the tag, and the spatial position of the measurement point relative to the tag will also affect the direction distribution of the magnetic field. The magnetic field distribution of the tag is symmetrical, and its strength and direction in space can be obtained by multi-point sampling of the detector. The sensor array collects magnetic field strength and direction data from multiple locations, and combines the distribution law of magnetic dipoles to infer the collected data into the specific three-dimensional coordinate position of the passive tag. This process is the inversion calculation of determining the spatial position of the tag through magnetic field distribution information. Finally, the server combines the three-dimensional coordinates of multiple passive tags to understand the dynamic position of different parts of the human body. By comparing the coordinate changes at consecutive time points, body posture changes such as rotation, tilt, and stretching are calculated. For example, displacement is calculated through differential methods, and rotation and deformation are inferred through matrix transformation.

[0056] S150, inputting the first posture change data and the second posture change data into a preset LSTM model to generate a dynamic posture three-dimensional model of the human body to perform posture monitoring on the human body.

[0057] Specifically, LSTM is a deep learning model that can process time series data and is good at extracting dynamic change rules from continuous time data. In an embodiment of the present application, LSTM predicts the dynamic changes of human body posture on the time axis by learning the time series characteristics of the input data, and constructs a dynamic three-dimensional body posture model. The LSTM model generates a dynamic body posture three-dimensional model based on the two input body posture change data, covering the key point parameters and surface deformation parameters of the human skeleton, reflecting the patient's dynamic posture. Real-time monitoring of human body posture is achieved, especially the accurate capture of subtle posture changes during surgery, providing doctors with instant visual reference.

[0058] In a possible implementation, the first posture change data and the second posture change data are input into a preset LSTM model to generate a dynamic three-dimensional model of a human body, specifically including: determining deformation distribution data of the human body according to the first posture change data; determining three-dimensional displacement data of the human body according to the second posture change data, wherein the first posture change data and the second posture change data have the same corresponding timestamp; using a feedforward neural network to perform feature extraction on the deformation distribution data and the three-dimensional displacement data respectively to obtain a first feature vector and a second feature vector; concatenating the first feature vector and the second feature vector to generate a high-dimensional vector; inputting the high-dimensional vector into a preset LSTM model to obtain three-dimensional posture change parameters corresponding to the timestamp, wherein the three-dimensional posture change parameters include key point skeleton parameters and body surface deformation parameters; and constructing a dynamic three-dimensional model according to the key point skeleton parameters and the body surface deformation parameters.

[0059] Specifically, the deformation distribution data is based on the first posture change data to extract the deformation characteristics of the human body surface in space, such as local displacement or expansion of muscles or skin. The three-dimensional displacement data is based on the second posture change data to determine the three-dimensional motion trajectory of the human body as a whole, such as the tilt of the trunk, the movement of the arms, etc. These data are all timestamped to ensure that they can be analyzed synchronously on the time axis. The server uses a feedforward neural network, such as a multi-layer perceptron, MLP to extract features from the deformation distribution data and the three-dimensional displacement data respectively. The feature extraction of the deformation distribution data is to extract the key features describing the surface deformation, such as the distribution pattern of the deformation amplitude, and the feature extraction of the three-dimensional displacement data is to extract the trend of spatial displacement, such as the direction of change of the position coordinates. The result of the feature extraction is the first eigenvector and the second eigenvector. Next, the server splices the first eigenvector and the second eigenvector to form a high-dimensional vector that comprehensively expresses the overall posture change information of the human body at the timestamp.

[0060] Among them, LSTM analyzes continuous input data through time series modeling to generate three-dimensional posture change parameters corresponding to the current timestamp. Key point skeleton parameters reflect the spatial distribution of the human skeletal structure, such as the positions of key points such as shoulders, knees, and elbows. Body surface deformation parameters reflect subtle deformations of the human surface, such as chest expansion or contraction caused by breathing. Based on the key point skeleton parameters and body surface deformation parameters, a dynamic body posture three-dimensional model is generated using three-dimensional modeling technology to provide intuitive visualization of the patient's posture.

[0061] Furthermore, the construction and training process of the preset LSTM model is as follows: collect time series data of human body posture changes, including deformation distribution data and three-dimensional displacement data, and normalize and timestamp-align the data. Use a feedforward neural network to extract the features of deformation distribution data and three-dimensional displacement data respectively, obtain two sets of feature vectors and splice them into a high-dimensional feature. Input the time series high-dimensional features to the model, and capture the dynamic characteristics of posture changes through the LSTM layer. The model outputs predicted three-dimensional posture change parameters, including key point skeletons and surface deformation. Finally, use the labeled data for model training, and deploy it on the server after verification to receive sensor data in real time and generate a three-dimensional model of human dynamic posture.

[0062] In one possible implementation, refer to Figure 2 , Figure 2 Another flow chart of a posture monitoring method provided in an embodiment of the present application. Specifically comprising steps S210 to S230, the steps are as follows: S210, using the SLAM algorithm to correct the actual position of the human body and the position of the dynamic posture three-dimensional model, to obtain a corrected dynamic posture model; S220, marking the position of the surgical target in the corrected dynamic posture model, to obtain a target visualization model; S230, using AR glasses, showing the target visualization model to the surgeon corresponding to the human body, the surgeon wearing AR glasses.

[0063] Specifically, the server uses the SLAM algorithm to calibrate the position of the actual human body with the position of the three-dimensional model. For example, the predicted position of the dynamic body three-dimensional model is input, and the algorithm combines the external environment, such as the markers on the operating table or the reference position provided by the sensor, to obtain a more accurate corrected dynamic body three-dimensional model. In the corrected model, the doctor selects and marks the position of the surgical target through interactive devices such as touch screens or voice commands. The target information includes its spatial coordinates in the three-dimensional model to ensure complete alignment with the real human body position. The server combines the corrected dynamic body model with the marked surgical targets to generate a three-dimensional visualization model containing the surgical target area. The model dynamically displays the patient's body changes and target positions, allowing doctors to intuitively understand the surgical area. Finally, the target visualization model is displayed in the doctor's field of view in an overlaid form, aligned with the patient's actual position in real time, providing real-time body updates, surgical navigation information and target guidance, so that doctors can directly observe the dynamic three-dimensional model and target through AR glasses without relying on external screens, improving surgical efficiency and accuracy.

[0064] The present application also provides a posture monitoring device, referring to Figure 3 , Figure 3 A module schematic diagram of a posture monitoring device provided in an embodiment of the present application. The posture monitoring device is a server, and the server includes an acquisition module 31 and a processing module 32, wherein the acquisition module 31 acquires bioelectric field data for a human body sent by an electric field sensing array device; the processing module 32 determines first posture change data of a human body according to the bioelectric field data; the acquisition module 31 acquires magnetic positioning data for a human body sent by a magnetic field sensing array device; the processing module 32 determines second posture change data of a human body according to the magnetic positioning data; the processing module 32 inputs the first posture change data and the second posture change data into a preset LSTM model to generate a dynamic posture three-dimensional model of a human body to perform posture monitoring on the human body.

[0065] In a possible implementation, the acquisition module 31 acquires the bioelectric field data for the human body sent by the electric field sensing array device, specifically including: the acquisition module 31 receives the first electric field signal for the human body sent by the electric field sensing array device at the first timestamp, and the electric field sensing array device includes a plurality of electric field sensors distributed circumferentially on the human body in a rectangular grid; the processing module 32 amplifies the first electric field signal according to the biological signal characteristics of the human body using a high-gain operational amplifier to obtain a first analog electric field signal; the processing module 32 filters the first analog electric field signal using a filter to obtain a second analog electric field signal, and the filter includes a low-pass filter and a band-pass filter; the processing module 32 converts the second analog electric field signal into a first digital signal; the acquisition module 31 acquires the second digital signal corresponding to the second electric field signal for the human body sent by the electric field sensing array device at the second timestamp; the processing module 32 determines the bioelectric field data according to the first digital signal and the second digital signal.

[0066] In a possible implementation, the processing module 32 determines the first body posture change data of the human body based on the bioelectric field data, specifically including: the processing module 32 determines the first signal strength corresponding to the human body at the first timestamp based on the first digital signal; the processing module 32 determines the second signal strength corresponding to the human body at the second timestamp based on the second digital signal; the processing module 32 obtains the first body posture change data corresponding to the human body through differential calculation based on the first signal strength and the second signal strength.

[0067] In a possible implementation, the acquisition module 31 acquires magnetic positioning data for a human body sent by a magnetic field sensing array device, specifically including: the processing module 32 sends a magnetic field acquisition signal to a passive magnetic tag through the magnetic field sensing array device, and the installation position of the passive magnetic tag on the human body is determined according to the type of surgery corresponding to the human body; the acquisition module 31 acquires the magnetic field array signal sent by the magnetic field sensing array device, and the magnetic field array signal is determined by the magnetic field sensing array device receiving a magnetic field return signal sent by the passive magnetic tag; the processing module 32 generates magnetic positioning data according to the magnetic field array signal.

[0068] In a possible implementation, the processing module 32 determines the second body posture change data of the human body based on the magnetic positioning data, specifically including: the processing module 32 determines the magnetic field strength signal corresponding to each passive magnetic tag based on the magnetic positioning data; the processing module 32 inverts the multiple three-dimensional coordinates marked by the passive magnetic tags on the human body through the magnetic dipole model based on the magnetic field strength signal corresponding to each passive magnetic tag; the processing module 32 determines the second body posture change data corresponding to the human body based on the multiple three-dimensional coordinates.

[0069] In a possible implementation, the processing module 32 inputs the first posture change data and the second posture change data into a preset LSTM model to generate a dynamic three-dimensional model of the human body, specifically including: the processing module 32 determines the deformation distribution data of the human body according to the first posture change data; the processing module 32 determines the three-dimensional displacement data of the human body according to the second posture change data, and the first posture change data and the second posture change data have the same corresponding timestamp; the processing module 32 uses a feedforward neural network to perform feature extraction on the deformation distribution data and the three-dimensional displacement data respectively to obtain a first feature vector and a second feature vector; the processing module 32 concatenates the first feature vector and the second feature vector to generate a high-dimensional vector; the processing module 32 inputs the high-dimensional vector into the preset LSTM model to obtain three-dimensional posture change parameters corresponding to the timestamp, and the three-dimensional posture change parameters include key point skeleton parameters and body surface deformation parameters; the processing module 32 constructs a dynamic three-dimensional model according to the key point skeleton parameters and the body surface deformation parameters.

[0070] In a possible implementation, the processing module 32 uses a SLAM algorithm to correct the actual position of the human body and the position of the dynamic body three-dimensional model to obtain a corrected dynamic body model; the processing module 32 marks the position of the surgical target in the corrected dynamic body model to obtain a target visualization model; the processing module 32 displays the target visualization model to the surgeon corresponding to the human body through AR glasses, and the surgeon wears AR glasses.

[0071] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0072] The present application also provides an electronic device, referring to Figure 4 , Figure 4 The electronic device may include: at least one processor 41 , at least one network interface 44 , a user interface 43 , a memory 45 , and at least one communication bus 42 .

[0073] The communication bus 42 is used to realize the connection and communication between these components.

[0074] The user interface 43 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 43 may also include a standard wired interface and a wireless interface.

[0075] The network interface 44 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0076] Among them, the processor 41 may include one or more processing cores. The processor 41 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 45, and calling data stored in the memory 45. Optionally, the processor 41 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 41 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 41, and it can be implemented separately through a chip.

[0077] Among them, the memory 45 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 45 includes a non-transitory computer-readable storage medium. The memory 45 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 45 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 45 may also be optionally at least one storage device located away from the aforementioned processor 41. As Figure 4 As shown, the memory 45 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of a body posture monitoring method.

[0078] exist Figure 4 In the electronic device shown, the user interface 43 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 41 can be used to call an application program storing a body posture monitoring method in the memory 45. When executed by one or more processors, the electronic device executes one or more methods in the above-mentioned embodiments.

[0079] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0080] The present application also provides a computer-readable storage medium, which stores instructions. When executed by one or more processors, the electronic device executes one or more of the methods described in the above embodiments.

[0081] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0082] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0083] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0085] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0086] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the embodiments, those skilled in the art will easily think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The specification and the embodiments are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A body posture monitoring method, characterized in that: The method comprises: The electric field sensing array device includes a plurality of electric field sensors distributed circumferentially on the human body in a rectangular grid; receiving a first digital signal corresponding to a first electric field signal for the human body sent by the electric field sensing array device at a first timestamp, Acquire a second digital signal corresponding to a second electric field signal for the human body sent by the electric field sensing array device at a second timestamp; Determine, according to the first digital signal, a first signal strength corresponding to the human body at the first timestamp; Determine, according to the second digital signal, a second signal strength corresponding to the human body at the second timestamp; Obtaining first body posture change data corresponding to the human body through differential calculation according to the first signal strength and the second signal strength; Acquiring magnetic positioning data for the human body sent by a magnetic field sensor array device; Determining second body posture change data of the human body according to the magnetic positioning data; Determining deformation distribution data of the human body according to the first body posture change data; determining the three-dimensional displacement data of the human body according to the second body posture change data, wherein the timestamps corresponding to the first body posture change data and the second body posture change data are the same; Using a feedforward neural network to extract features from the deformation distribution data and the three-dimensional displacement data, respectively, to obtain a first feature vector and a second feature vector; Concatenate the first eigenvector and the second eigenvector to generate a high-dimensional vector; Input the high-dimensional vector into a preset LSTM model to obtain three-dimensional body posture change parameters corresponding to the timestamp, wherein the three-dimensional body posture change parameters include key point skeleton parameters and body surface deformation parameters; A dynamic body three-dimensional model is constructed based on the key point skeleton parameters and the body surface deformation parameters to perform body posture monitoring.

2. The body posture monitoring method according to claim 1, characterized in that: The step of acquiring magnetic positioning data for the human body sent by the magnetic field sensor array device specifically includes: Sending a magnetic field acquisition signal to a passive magnetic tag through the magnetic field sensor array device, wherein the installation position of the passive magnetic tag on the human body is determined according to the type of surgery corresponding to the human body; Acquiring a magnetic field array signal sent by the magnetic field sensing array device, wherein the magnetic field array signal is determined by receiving a magnetic field return signal sent by the passive magnetic tag from the magnetic field sensing array device; The magnetic positioning data is generated according to the magnetic field array signal.

3. The body posture monitoring method according to claim 2, characterized in that: Determining the second body posture change data of the human body according to the magnetic positioning data specifically includes: Determine the magnetic field strength signal corresponding to each of the passive magnetic tags according to the magnetic positioning data; According to the magnetic field strength signal corresponding to each of the passive magnetic tags, inverting multiple three-dimensional coordinates marked by the passive magnetic tag on the human body through a magnetic dipole model; According to the multiple three-dimensional coordinates, second body posture change data corresponding to the human body is determined.

4. The body posture monitoring method according to claim 1, characterized in that: The method further comprises: Using a SLAM algorithm to correct the actual position of the human body and the position of the dynamic body three-dimensional model to obtain a corrected dynamic body model; In the corrected dynamic body model, the surgical target position is marked to obtain a target visualization model; The target visualization model is displayed to a surgeon corresponding to the human body through the AR glasses, and the surgeon wears the AR glasses.

5. A body posture monitoring device, characterized in that: The body posture monitoring device comprises an acquisition module (31) and a processing module (32), wherein the electric field sensor array device comprises a plurality of electric field sensors which are distributed circumferentially on the human body in a rectangular grid shape; The acquisition module (31) is used to receive a first digital signal corresponding to a first electric field signal for the human body sent by the electric field sensing array device at a first timestamp, The acquisition module (31) is further used to acquire a second digital signal corresponding to a second electric field signal for the human body sent by the electric field sensing array device at a second timestamp; The processing module (32) is used to determine a first signal strength of the human body corresponding to the first timestamp according to the first digital signal; The processing module (32) is further used to determine, based on the second digital signal, a second signal strength of the human body corresponding to the second timestamp; The processing module (32) is further used to obtain first posture change data corresponding to the human body through differential calculation according to the first signal strength and the second signal strength; The acquisition module (31) is also used to acquire magnetic positioning data for the human body sent by the magnetic field sensor array device; The processing module (32) is further used to determine second body posture change data of the human body according to the magnetic positioning data; The processing module (32) is further used to determine the deformation distribution data of the human body according to the first body shape change data; The processing module (32) is further used to determine the three-dimensional displacement data of the human body according to the second body posture change data, and the timestamps corresponding to the first body posture change data and the second body posture change data are the same; The processing module (32) is further used to use a feedforward neural network to perform feature extraction on the deformation distribution data and the three-dimensional displacement data respectively to obtain a first feature vector and a second feature vector; The processing module (32) is further used to concatenate the first feature vector and the second feature vector to generate a high-dimensional vector; The processing module (32) is further used to input the high-dimensional vector into a preset LSTM model to obtain three-dimensional body posture change parameters corresponding to the timestamp, wherein the three-dimensional body posture change parameters include key point skeleton parameters and body surface deformation parameters; The processing module (32) is also used to construct a dynamic body posture three-dimensional model based on the key point skeleton parameters and the body surface deformation parameters to perform body posture monitoring.

6. An electronic device, characterized in that: The electronic device comprises a processor (41), a memory (45), a user interface (43) and a network interface (44), wherein the memory (45) is used to store instructions, the user interface (43) and the network interface (44) are both used to communicate with other devices, and the processor (41) is used to execute the instructions stored in the memory (45) so that the electronic device executes the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 4 is performed.

Images