Motion artifact monitoring method and system for emergency equipment

Abstract

The invention discloses a motion artifact monitoring method and system for first-aid equipment, and belongs to the technical field of first-aid monitoring, and the method comprises the steps: introducing a space-time code, carrying out the collection of a motion signal and a physiological signal for a target user, and determining the state data of the user, the space-time code comprising a synchronization timestamp and a space phase; an artifact processing module is introduced and deployed in a data center of the first-aid equipment in an embedded mode, user state data are received, effective state data are determined by triggering the artifact processing module and executing data directional recognition and artifact monitoring processing, and the first-aid equipment executes response based on the effective state data. Wherein the artifact processing module is of a double-buffer-area structure, a front buffer area executes filter initialization under artifact identification, and a rear buffer area executes artifact processing. The technical problems that in the prior art, motion and physiological signal collection in an emergency scene is prone to being interfered by artifacts, data are not synchronous, and consequently vital sign recognition is inaccurate, and intervention is delayed are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency monitoring, and in particular to a motion artifact monitoring method and system for emergency equipment. Background Art

[0002] With the widespread application of wearable physiological monitoring devices and intelligent first aid systems, the real-time collection and analysis of vital signs such as heart rate, blood oxygen, and body temperature has become one of the key links in the rescue process.

[0003] Existing motion artifact suppression methods often use simple filtering algorithms or fixed threshold recognition mechanisms. However, in applications with frequent dynamic posture changes or heterogeneous multi-source devices, these methods suffer from the following shortcomings: They cannot synchronize the acquisition and control of the user's current posture and motion state in both time and space, resulting in inherent synchronization deviations in the source data; they lack dynamic threshold recognition and feedback optimization mechanisms, which can easily lead to misjudgments or missed detections; and the processing process generally uses single-buffer serial processing, resulting in high response latency and difficulty meeting the requirements of real-time artifact removal. Therefore, a new method with high-precision artifact recognition and adaptive processing capabilities for high-interference dynamic environments is urgently needed to ensure the stability and reliability of emergency equipment monitoring under complex motion conditions. Summary of the Invention

[0004] This application provides a motion artifact monitoring method and system for first aid equipment, aiming to solve the technical problems in the prior art that motion and physiological signal acquisition in first aid scenarios are easily interfered by artifacts and data is out of sync, resulting in inaccurate vital sign recognition and delayed intervention.

[0005] In view of the above problems, the present application provides a motion artifact monitoring method and system for emergency equipment.

[0006] The first aspect disclosed in the present application provides a motion artifact monitoring method for emergency equipment, which includes introducing a space-time code, collecting motion signals and physiological signals of a target user, and determining user status data, wherein the space-time code includes a synchronization timestamp and a spatial phase; introducing an artifact processing module, which is embedded and deployed in a data center of the emergency equipment, receives the user status data, triggers the artifact processing module, performs data-oriented identification and artifact monitoring processing, determines valid status data, and the emergency equipment executes a response based on the valid status data; wherein the artifact processing module is a dual-buffer architecture, the front buffer performs filter initialization under artifact identification, and the back buffer performs artifact processing.

[0007] Another aspect disclosed in the present application provides a motion artifact monitoring system for emergency equipment, the system including a user status data determination module for introducing a space-time code, collecting motion signals and physiological signals of a target user, and determining user status data, wherein the space-time code includes a synchronization timestamp and a spatial phase; a valid status data determination module for introducing an artifact processing module, embeddedly deployed in a data center of the emergency equipment, receiving the user status data, performing data-oriented identification and artifact monitoring processing by triggering the artifact processing module, determining valid status data, and the emergency equipment executing a response based on the valid status data; wherein the artifact processing module has a dual-buffer architecture, the front buffer performs filter initialization under artifact identification, and the back buffer performs artifact processing.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] By introducing a space-time code containing a synchronized timestamp and spatial phase to synchronously acquire motion signals and physiological signals, and combining an embedded artifact processing module with a dual-buffer architecture, the front buffer performs artifact identification and filter initialization, and the back buffer performs artifact processing. This technical solution solves the technical problems of asynchronous multi-source signal acquisition and severe artifact interference in emergency scenarios in the existing technology, which lead to distortion of vital sign recognition and delayed response, and achieves the technical effect of improving data synchronization, enhancing signal purification accuracy and ensuring the real-time response capability of equipment.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic flow chart of a method for monitoring motion artifacts of emergency equipment is provided for an embodiment of the present application;

[0012] Figure 2 A schematic structural diagram of a motion artifact monitoring system for emergency equipment is provided for an embodiment of the present application.

[0013] Description of reference numerals: user status data determination module 11 , valid status data determination module 12 . DETAILED DESCRIPTION

[0014] The overall idea of the technical solution provided by this application is as follows:

[0015] The present invention provides a motion artifact monitoring method and system for emergency medical equipment. By introducing space-time codes to achieve temporal and spatial synchronization of motion and physiological signals, an artifact processing module is embedded in the emergency medical equipment. A dual-buffer structure is employed: a front buffer for artifact identification and filter initialization, and a back buffer for specific artifact processing. This outputs artifact-free valid status data, enabling stable and accurate vital sign monitoring in high-interference environments.

[0016] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced in conjunction with the drawings in the specification.

[0017] Example 1

[0018] like Figure 1 As shown, an embodiment of the present application provides a method for monitoring motion artifacts of emergency equipment, the method comprising:

[0019] Step S100 , introducing a space-time code, collecting motion signals and physiological signals of a target user, and determining user status data, wherein the space-time code includes a synchronous timestamp and a spatial phase.

[0020] Specifically, a space-time code refers to a coding structure used to identify the "time and space information" during data collection. In this application, the space-time code consists of two elements: a synchronized timestamp: which records the exact time when the data is collected to ensure that the data between different devices are synchronized under the same time reference; and a spatial phase: which is based on the user's real-time posture, orientation, or displacement state, and is used to describe the spatial position parameters of the device or user at the time of collection. A motion signal refers to data reflecting the user's motion, posture, or body movement state collected by sensors such as an inertial measurement unit (IMU), accelerometer, and gyroscope. A physiological signal refers to the user's vital sign parameters collected by a physiological monitoring device, such as heart rate (ECG), blood oxygen (SpO2), body temperature, respiratory rate, etc. User status data refers to a multimodal joint data set that combines "motion signal + physiological signal" and is synchronized based on the space-time code, and is used to reflect the user's current physiological and behavioral state.

[0021] During implementation, emergency equipment integrates multi-source sensor modules (such as IMUs, ECG sensors, and body temperature sensors) and introduces a higher-level controller (such as an edge computing processor) to deploy a "space-time code" mechanism. This mechanism implements the following steps: First, a synchronized timestamp is generated for each data collection task. This timestamp can be provided by a GPS module, a master clock source, or a Network Time Protocol (NTP) module to ensure that all devices use the same sampling reference. With synchronized timestamps, motion monitoring devices and physiological monitoring devices collect data streams in parallel. Simultaneously, the user's posture or spatial position information (such as prone, sitting up, or side-lying) is acquired as spatial phase information, typically using an IMU, gyroscope, camera visual tracking, or UWB (ultra-wideband) positioning system. Finally, the "acquisition timestamp + spatial phase information" is combined to generate a space-time code, which is then bound to the collected motion and physiological signal data to form a unified, tagged user status data packet.

[0022] By introducing a space-time code containing "synchronous timestamp + spatial phase", time-synchronous acquisition control of heterogeneous multiple devices is achieved to ensure data alignment; dynamic spatial state label assignment improves the signal context understanding ability; provides a "data occurrence background" basis for subsequent artifact identification, significantly improving the accuracy of interference identification and the matching degree of processing strategies; and ultimately outputs higher-precision and more robust user status data to support the precise response of the emergency system.

[0023] Step S200, introduce an artifact processing module, which is embedded and deployed in the data center of the emergency equipment, receives the user status data, triggers the artifact processing module, performs data-oriented identification and artifact monitoring processing, determines valid status data, and the emergency equipment executes a response based on the valid status data.

[0024] Specifically, the artifact processing module is an intelligent processing unit deployed within emergency medical equipment, specifically designed to identify and eliminate motion artifacts caused by user movement, device jitter, or sensor noise, thereby improving the authenticity and accuracy of physiological data. Data-directed recognition involves the targeted identification of locations or time periods where artifacts may exist based on spatial information, motion patterns, or abnormal characteristics in user status data. Valid status data refers to high-quality signal data after artifact interference has been eliminated, serving as the basis for subsequent device decisions and responses.

[0025] Deploy edge computing units (such as embedded ARM processors, FPGAs, DSP chips, etc.) in emergency equipment and install an artifact processing module as an independent data processing core. Receive user status data with time and space codes in real time from the host device or sensor network, including acceleration signals, electrocardiogram signals, blood oxygen curves, etc. Upon receiving a new status data packet or detecting an event such as a sudden change in posture or abnormal signal fluctuation, the artifact processing module is automatically triggered. Potential artifacts related to changes in user posture, time period, and frequency are identified; adaptive filters, wavelet denoising, pattern matching, or machine learning models are applied to identify and remove artifacts; after removal, portions with high signal-to-noise ratios are retained and valid status data is output. Emergency equipment (such as portable defibrillators and electrocardiogram monitors) performs alarms, recording, intervention, and other operations based on this valid status data.

[0026] This step achieves real-time identification and elimination of motion artifacts, improving data reliability. Based on the directional recognition mechanism, it can optimize processing strategies based on user behavior status. It avoids signal misunderstandings caused by artifacts and improves the accuracy of vital sign judgment. It ensures that emergency equipment can make accurate and rapid response decisions in high-dynamic scenarios on site, such as timely defibrillation and early warning, significantly improving emergency efficiency and safety.

[0027] The artifact processing module is a double buffer architecture, wherein the front buffer performs filter initialization under artifact identification, and the back buffer performs artifact processing.

[0028] Specifically, the dual-buffer architecture divides the processing process into two independent but collaborative buffers (i.e., the front buffer and the back buffer), each of which performs different functions to parallelize and asynchronously process artifact identification and removal tasks: the front buffer is used for preprocessing, preliminary identification, threshold judgment, and filter parameter preparation; the back buffer is used for actual filtering processing, artifact removal, and signal reconstruction.

[0029] The front buffer receives a user status data stream marked with time and space codes, including acceleration, electrocardiogram, and blood oxygen levels. Artifact identification and filter initialization are performed in the front buffer. Artifact analysis is performed using the front buffer: posture variables are read; time-frequency characteristics are compared; abnormal changes (such as sudden amplitude increases or frequency drift) are determined; artifact probability instructions are generated based on the artifact identification results; and based on these instructions, the adaptive filter is initialized. For example, if the artifact frequency is concentrated in the 1–3 Hz range, a notch filter is initialized; if the posture changes dramatically, a bandpass filter + Kalman filter combination is initialized.

[0030] The back buffer performs artifact processing, using the filter configuration initialized by the front buffer. It filters, removes noise, and smoothly reconstructs the complete data segment, outputting a "motion-clean signal" as valid status data for the system response module. The dual buffer can operate using a ring cache structure or a time-sharing scheduling strategy, improving overall system response efficiency and processing throughput.

[0031] It achieves decoupling of front-end and back-end processing, supports more complex artifact recognition and dynamic adjustment of processing strategies; improves the system's real-time processing capability and response speed, and avoids signal blocking or delay caused by serial processing; the filter can adaptively adjust parameters according to artifact characteristics to improve signal purification effect; significantly improves the flexibility, robustness and processing accuracy of the artifact processing system, and ensures that subsequent emergency intervention is performed based on trusted data.

[0032] Furthermore, the determination of user status data includes: performing upper-level association between the motion monitoring device and the physiological monitoring device to construct an upper device; introducing space-time codes and deploying the upper device; generating a synchronous acquisition timestamp according to the upper device, driving the motion monitoring device and the physiological monitoring device, determining the spatial phase based on the target user posture, performing directional monitoring, transmitting and integrating, and determining the user status data.

[0033] Specifically, upper-level association refers to the integration and unified management of multiple underlying monitoring devices in the control logic, and the coordination and scheduling through the upper-level control module to achieve information fusion and collection. The upper-level device refers to the central software system responsible for unified drive, time synchronization, and data aggregation. The target user posture refers to the user's body posture, orientation or spatial movement status at a specific point in time, which is usually obtained through IMU (inertial measurement unit), posture estimation algorithm or visual positioning technology. Directional monitoring refers to the dynamic adjustment of data acquisition parameters or sensor sampling direction based on the target user's posture status to achieve more targeted signal extraction.

[0034] First, the motion monitoring module (such as a triaxial accelerometer and gyroscope) and the physiological monitoring module (such as an ECG sensor, a PPG blood oxygen sensor, and a body temperature sensor) in the emergency equipment are integrated into a host device via bus protocols (such as I2C / SPI / UART) or wireless protocols (such as BLE / WiFi). Specifically, the motion monitoring device and the physiological monitoring device are hardware-integrated via bus or wireless communication protocols. Second, a unified control processing unit (such as an embedded MCU or edge computing module) is deployed, loading the acquisition driver and space-time code management program. A synchronous acquisition trigger mechanism is configured to ensure time consistency between the different devices. Finally, a data fusion interface is implemented to integrate and output the collected data according to time and space tags, forming a unified user status data stream. The host device has an embedded space-time code management module, which uses the system clock or Network Time Protocol (NTP) as the time source and combines it with the IMU attitude fusion algorithm to calculate the spatial phase in real time. The host device issues a unified start command, triggering the motion and physiological sensors to synchronize sampling using the current timestamp and attitude state as tags. The sampling process is dynamically optimized based on the target user's current posture, such as prioritizing chest ECG signals when the user is supine or adjusting the posture perception threshold when the user is rolling over. All sampled data is time-space coded and returned to the host computer, which aligns and fuses the multi-source information to form a structured user status data packet for subsequent artifact processing modules.

[0035] It achieves unified temporal and spatial control of multi-source monitoring equipment; it can sense changes in user posture and implement adaptive adjustment of sampling methods; data is uniformly marked as "user status data" to provide accurate and structured input for subsequent artifact processing and anomaly detection; it can significantly improve data quality and response timeliness at the emergency scene, and enhance monitoring and intervention efficiency.

[0036] Furthermore, the front buffer performs filter initialization under artifact recognition, and the back buffer performs artifact processing, including: identifying the user status data, and the artifact processing module extracts the motion signal through matching; for the motion signal, performing artifact recognition and filter initialization based on the front buffer, triggering the back buffer to perform artifact processing, and outputting a motion clean signal.

[0037] Specifically, motion clean signal refers to high-quality signal data that has been output after artifact processing, with interference components removed and the characteristics of the true motion state retained.

[0038] First, the artifact processing module performs signal matching and pattern recognition on the received user status data to extract the motion signal portion representing the user's body movements. For example, it accurately separates the three-axis acceleration sequence from a mixed data stream containing ECG, blood oxygen, and acceleration. After extraction, the motion signal is fed into the front buffer for artifact identification. The system uses joint time-frequency analysis and comparison with posture variables to determine the presence of artifacts. For example, artifacts are considered when periodic high-frequency perturbations or sudden posture changes are detected in the signal. Based on the judgment result, the system uses the calculated artifact probability command as input to select and initialize an appropriate filter strategy, such as using a bandpass filter to filter artifacts within a specified frequency band or using a wavelet denoising function to locally remove time-varying perturbations. After filter initialization, a control signal triggers the back buffer to initiate operation, invoking the configured filter parameters and algorithm modules to perform substantial artifact removal on the motion signal. This process may include weighted filtering, adaptive filter application, baseline correction, and threshold suppression. Once processed, the system outputs a clean motion signal—high-quality, artifact-free body motion data—for subsequent modules to perform critical operations such as vital state assessment and intervention strategy selection. The entire process is built on an embedded data processing platform, utilizing edge processors (such as ARM Cortex-M series or DSP chips) to perform signal separation and processing in real time.

[0039] For example, during first aid on the scene, a patient is placed on a stretcher, causing an abnormal mutation in the three-axis acceleration signal (such as the X-axis amplitude increasing from 0.2g to 2.5g). The front buffer recognizes this mutation and determines it as a strong artifact area, initializing the notch filter and sliding mean filter. The back buffer then takes over, clearing the abnormal fluctuations and outputting a smooth, recognizable posture trajectory.

[0040] This step enables the rapid and accurate extraction of motion components from complex multimodal data; supports dynamic adjustment of filter configuration based on artifact characteristics to enhance the adaptability of artifact processing; ensures processing efficiency and real-time output through the collaboration of front and back buffers; and ultimately improves the system's ability to restore real-world user motion characteristics, providing highly reliable input for downstream decision-making modules and significantly improving the stability and accuracy of emergency response.

[0041] Furthermore, artifact identification based on the front buffer is performed, including: introducing an interference threshold, wherein the interference threshold is set as a critical value under the joint constraint of posture variable threshold-time-frequency, and the posture variable includes at least one posture feature; using the time-frequency as the first independent variable, the posture variable threshold as the second independent variable, and the interference threshold as the third dependent variable to construct an identification linear relationship; based on the identification linear relationship, performing artifact identification on the motion signal to generate an artifact probability instruction.

[0042] Specifically, the interference threshold refers to the numerical limit used to determine whether a signal is interfered with. This application is dynamically adaptable by setting a joint constraint of posture variables and time-frequency features. Posture variables refer to variables that reflect the user's body posture or body movement state, such as roll angle, pitch angle, acceleration modulus, angular velocity, etc. Time-frequency features include the instantaneous amplitude, frequency distribution, energy density, frequency domain power, etc. of the signal, which are usually extracted through methods such as Fourier transform and wavelet analysis. Identifying linear relationships means taking "time-frequency features" and "posture variable thresholds" as two independent variables to construct a functional relationship model with the "interference threshold" for quantitatively determining whether there are artifacts. Artifact probability instructions refer to numerical outputs that reflect the possibility of artifacts in a certain signal segment, which can be used to guide whether the filter is enabled and its intensity setting.

[0043] First, the input signal's attitude variables, such as the pitch angle change rate or acceleration modulus change amplitude obtained from the IMU, are extracted. The signal's time-frequency features, such as the fluctuation frequency range, energy change rate, and peak density, are also extracted simultaneously. This extraction method can be based on the Fast Fourier Transform (FFT) or Continuous Wavelet Transform (CWT). A discriminative linear model is then constructed, using the time-frequency features as the first independent variable, the attitude variable threshold as the second independent variable, and the target interference determination threshold as the dependent variable. Specifically, a multivariate linear regression or support vector regression (SVR) model trained on historically labeled data is used to collect a large number of annotated historical motion signal samples and extract their corresponding time-frequency features and attitude variable values. A ternary variable dataset is then constructed based on these samples, with the time-frequency features as the first independent variable, the attitude variable threshold as the second independent variable, and the actual interference level as the dependent variable. A multivariate linear regression or support vector regression algorithm is then used for fitting and training to generate a mapping function. This function is then embedded in the front buffer for real-time calculation of the interference threshold, completing artifact identification and judgment.

[0044] During real-time operation, if the joint features of the current motion signal exceed the interference threshold given by the model, it is identified as an artifact segment. Based on this, the system generates an "artifact probability instruction," which represents the degree of signal contamination in probabilistic form (e.g., an artifact probability of 0.87 indicates severe interference). This instruction allows the filtering module to select whether to enable filtering and set the filter strength. The entire process is executed in real time on the embedded edge computing platform, and the identification model can be periodically updated to support adaptive interference adjustment in different scenarios.

[0045] This step implements a data-driven dynamic interference threshold setting mechanism to replace the static threshold method; combines posture and time-frequency information to improve the contextual association ability and accuracy of artifact judgment; and realizes flexible control and adaptive response of the artifact processing module through artifact probability instruction output. It significantly reduces misidentification and missed identification, improves the reliability of motion artifact processing, ensures signal purification accuracy, and provides stable support for the response of emergency equipment.

[0046] Furthermore, filter initialization is performed, including: if the artifact probability instruction is empty, the motion signal is used as the motion clean signal; if the artifact probability instruction is not empty, the filtering standard is determined according to the probability value, wherein the probability value is determined by the amplitude of the attitude variable threshold exceeding the limit; according to the filtering standard, the filter is initialized.

[0047] Specifically, the posture variable threshold excursion amplitude refers to the degree to which a user's posture variables (such as acceleration, angular velocity, and posture angle) exceed the normal physiological range. This is often used to quantitatively assess interference intensity and is a key factor in calculating artifact probability. The filtering standard refers to the filter parameter configuration rules, such as filter type (bandpass, notch, etc.), cutoff frequency, and order, that are set based on the artifact probability instruction and the excursion amplitude.

[0048] First, the artifact probability command output by the artifact identification module determines whether the current signal requires filtering. If this command is empty, indicating no significant artifact interference is detected in the current signal segment, the raw motion signal is directly output as a "motion-clean signal" without filtering. If the artifact probability command is valid, this probability value is used as a basis for evaluating the severity of the signal interference, combined with the magnitude of the attitude variable's overshoot (for example, the degree to which acceleration exceeds the normal range of ±1.5g). This value is then used to determine the severity of the signal interference and to determine the filtering criteria. These criteria may include the selected filter type (e.g., notch filter to remove interference in a specific frequency band, bandpass filter to retain the primary frequency band), filter order (affecting response speed and accuracy), and cutoff frequency range. Based on these parameters, the system then initializes the filter configuration module, setting the specific algorithm flow and parameter values. Common filter implementation tools include digital IIR filters, FIR filters, wavelet denoising modules, or Kalman filter combinations. After initialization, the processing chain automatically applies the configured filter parameters for subsequent artifact removal, ensuring the authenticity and stability of the output signal.

[0049] This step implements an on-demand filtering mechanism to avoid over-processing of interference-free signals; by combining probability values and out-of-limit amplitudes, a dynamic and adaptive filtering standard system is constructed; the accuracy and adaptability of artifact processing are improved, reducing the false filtering rate and signal distortion risk; and the system's ability to respond to scenarios with different interference intensities is enhanced, ensuring the continuity and credibility of signal output, thereby improving the accuracy and reliability of emergency response.

[0050] Furthermore, before executing the artifact recognition based on the front buffer, it includes: locating the interference period based on whether there is a posture variable for the motion signal; intercepting the motion signal based on the interference period to determine the recognition signal segment; and triggering the artifact recognition based on the front buffer based on the recognition signal segment.

[0051] Specifically, posture variables represent the physical quantities of the user's body posture or movement state, commonly including three-axis acceleration, angular velocity, roll angle, pitch angle, etc., usually collected by IMU (inertial measurement unit). Interference period refers to the abnormal data segment in the signal caused by drastic changes in posture or sudden movements. The signal during this period has a high possibility of artifacts. The identification signal segment refers to the local signal segment intercepted from the original motion signal according to the positioning result of the interference period, which is used for subsequent artifact identification processing.

[0052] Before performing the artifact identification operation, the received complete motion signal is first pre-processed and analyzed. Specifically, the changes in the posture variables in the signal are monitored in real time, such as sudden changes in the pitch angle, abnormal increases in the acceleration modulus, etc., to determine whether there are preliminary signs of artifact interference. When a significant change in the posture variables within a certain time period is detected (such as exceeding the set threshold of ±15° or the acceleration is greater than 1.8g), the system marks the time period as an interference period. Subsequently, the system intercepts the signal during the interference period and generates a shorter local analysis window, called the "identification signal segment", the time length of which can be set to 1 to 3 seconds, which is used to concentrate on analyzing whether there is real artifact interference. This signal segment is then input into the front buffer, triggering the artifact identification process. The front buffer performs in-depth analysis based on the posture variables, frequency changes, etc. of the segment and generates an artifact probability instruction.

[0053] This step effectively avoids the redundant computational burden of performing artifact recognition on the complete data stream; quickly determines whether there is artifact interference through posture priors, thereby improving the system response speed and processing efficiency; focuses on possible interference areas to improve the pertinence and accuracy of artifact recognition; retains the original structure of interference-free signal segments to reduce unnecessary signal loss or excessive filtering; and overall improves the real-time performance, accuracy, and resource utilization of the artifact monitoring module in emergency equipment.

[0054] Furthermore, the method further includes: obtaining the artifact operation chain and storing it in a temporary database,

[0055] The temporary database is updated based on a preset period; according to the temporary database, artifact identification misjudgment statistics and artifact processing deviation statistics under the preset period are performed to determine the abnormal operation data; according to the abnormal operation data, the artifact processing module is updated.

[0056] Specifically, the artifact operation chain refers to the chain of operation process records of the system during each artifact identification and processing process, including the calculated value of the artifact probability, the type of filter used, parameter settings, processing output results, etc. The temporary database is a module used to cache the artifact processing history records. It has a short-cycle automatic refresh feature to facilitate statistical analysis without occupying long-term storage resources. The preset period refers to the time interval set by the system (such as every 30 seconds, every 5 minutes), which automatically triggers an update and statistical process after the end of this time period. False positive statistics are used to identify situations where the artifact identification results are inconsistent with the true label (or subsequent verification data), such as when a valid signal was originally identified as an artifact. Processing deviation statistics refer to the degree of deviation between the statistical processing results and the expected signal characteristics, such as when the signal after filtering is over-compressed or the effective component is lost. Deviating operation data refers to operation records that are marked as having differences, misjudgments or processing deviations in statistics, which are used for subsequent parameter correction and model optimization of the artifact processing module.

[0057] By periodically executing a feedback mechanism to monitor the artifact processing effect, the artifact processing module is dynamically optimized and adaptively updated. Specifically, after each artifact processing operation, the current artifact operation chain is recorded, including the input signal characteristics, the generated artifact probability instructions, the filter selection and its parameters, the output signal quality, and other contents. These records are then stored in a temporary database that has the ability to automatically refresh and review based on a set time period. After each preset period (such as every 30 seconds or every 1 minute), the historical records in the database are called up to perform two key statistical analyses: one is to count whether there are any misjudgments in artifact identification, that is, to compare subsequent device feedback or manually annotated data to determine whether the recognition system mistakenly identifies valid signals as artifacts or vice versa; the other is to evaluate whether there are any deviations in the artifact processing results, such as distortion of the signal morphology after filtering or weakening of important physiological characteristic bands. By analyzing these records, the system generates a set of "existing deviation operation data", which represents sample data and corresponding operation parameters that have identification or processing deviations. Then, the system uses these data as samples to trigger parameter updates or model adjustment mechanisms within the artifact processing module, such as retraining the artifact probability regression model, adjusting the filter selection strategy, and optimizing the judgment threshold.

[0058] This step introduces a self-feedback closed-loop mechanism to achieve continuous self-optimization of the artifact processing module; improves the robustness and accuracy of artifact recognition and processing through misjudgment statistics and processing deviation evaluation; supports enhanced scene adaptation capabilities to cope with changes in artifact characteristics under different users, body postures, and environmental changes; significantly reduces the misjudgment rate and signal loss rate, improves the credibility of the final signal and data utilization efficiency, and provides more reliable vital signs data support for emergency equipment.

[0059] In summary, the motion artifact monitoring method for emergency equipment provided by the embodiments of the present application has the following technical effects:

[0060] 1. By introducing space-time codes and a double-buffered artifact processing architecture, emergency medical equipment can achieve synchronous acquisition of motion and physiological signals and real-time artifact monitoring, improving signal spatiotemporal consistency and processing accuracy, significantly enhancing the ability to extract valid vital sign data in high-dynamic environments, and providing reliable data support for emergency response.

[0061] 2. A collaborative processing mechanism between the front and rear buffers is adopted. The front buffer performs artifact identification and filter initialization, while the rear buffer performs signal purification. This achieves the separation and asynchronization of identification and processing, greatly improving processing efficiency and real-time performance, effectively reducing signal response delay and improving artifact removal accuracy.

[0062] Example 2

[0063] Based on the same inventive concept as the motion artifact monitoring method of emergency equipment in the aforementioned embodiment, Figure 2 As shown, an embodiment of the present application provides a motion artifact monitoring system for emergency equipment, the system comprising:

[0064] The user status data determination module 11 is used to introduce a space-time code, collect motion signals and physiological signals of the target user, and determine the user status data, wherein the space-time code includes a synchronization timestamp and a spatial phase; the valid status data determination module 12 is used to introduce an artifact processing module, which is embedded and deployed in the data center of the emergency equipment, receives the user status data, triggers the artifact processing module, performs data-oriented identification and artifact monitoring processing, determines the valid status data, and the emergency equipment executes a response based on the valid status data; wherein the artifact processing module has a dual-buffer architecture, the front buffer performs filter initialization under artifact identification, and the back buffer performs artifact processing.

[0065] Furthermore, the user status data determination module 11 is also used to perform the following steps: perform upper association of the motion monitoring device and the physiological monitoring device to construct an upper device; introduce space-time code and deploy the upper device; generate a synchronous acquisition timestamp according to the upper device, drive the motion monitoring device and the physiological monitoring device, determine the spatial phase based on the target user posture, perform directional monitoring, transmit and integrate, and determine the user status data.

[0066] Furthermore, the valid state data determination module 12 is also used to perform the following steps: identifying the user state data, and the artifact processing module extracts the motion signal through matching; for the motion signal, performing artifact identification and filter initialization based on the front buffer, triggering the back buffer to perform artifact processing, and outputting a motion clean signal.

[0067] Furthermore, the valid state data determination module 12 is also used to perform the following steps: introducing an interference threshold, wherein the interference threshold is set as a critical value under the posture variable threshold-time-frequency joint constraint, and the posture variable includes at least one posture feature; using the time-frequency as the first independent variable, the posture variable threshold as the second independent variable, and the interference threshold as the third dependent variable to construct an identification linear relationship; based on the identification linear relationship, performing artifact identification on the motion signal to generate an artifact probability instruction.

[0068] Furthermore, the valid state data determination module 12 is also used to perform the following steps: if the artifact probability instruction is empty, the motion signal is used as the motion clean signal; if the artifact probability instruction is not empty, the filtering standard is determined according to the probability value, wherein the probability value is determined by the amplitude of the attitude variable threshold exceeding the limit; according to the filtering standard, the filter is initialized.

[0069] Furthermore, the valid state data determination module 12 is also used to perform the following steps: for the motion signal, locate the interference period based on whether there is a posture variable; intercept the motion signal based on the interference period to determine the identification signal segment; for the identification signal segment, trigger artifact identification based on the front buffer.

[0070] Furthermore, the system is also used to perform the following steps: obtaining an artifact operation chain and storing it in a temporary database, wherein the temporary database is updated based on a preset period; according to the temporary database, performing artifact identification misjudgment statistics and artifact processing deviation statistics under a preset period to determine abnormal operation data; and updating the artifact processing module according to the abnormal operation data.

[0071] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0072] Furthermore, the terms "first" or "second" as described above may not only represent an order relationship but may also represent a specific concept and / or refer to the selection of multiple elements individually or collectively. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, if such modifications and variations fall within the scope of this application and its equivalents, this application is intended to include such modifications and variations.

Claims

1. A method for monitoring motion artifacts of emergency equipment, characterized in that: The method comprises: Introducing a space-time code to collect motion and physiological signals of a target user and determine user status data, wherein the space-time code includes a synchronized timestamp and spatial phase; An artifact processing module is introduced and embedded in the data center of the emergency equipment. The module receives the user status data, triggers the artifact processing module, performs data-oriented identification and artifact monitoring processing, determines valid status data, and the emergency equipment executes a response based on the valid status data. The artifact processing module is a double buffer architecture, wherein the front buffer performs filter initialization under artifact identification, and the back buffer performs artifact processing.

2. The motion artifact monitoring method for emergency equipment according to claim 1, characterized in that: The determining of user status data includes: Conduct a superior association between motion monitoring equipment and physiological monitoring equipment to build a superior device; Introducing space-time codes and deploying the host device; According to the host device, a synchronous acquisition timestamp is generated, the motion monitoring device and the physiological monitoring device are driven, the spatial phase is determined based on the target user posture, directional monitoring is performed, and the data is transmitted back and integrated to determine the user status data.

3. The motion artifact monitoring method for emergency equipment according to claim 1, characterized in that: The front buffer performs filter initialization under artifact identification, and the back buffer performs artifact processing, including: Identifying the user status data, the artifact processing module extracts the motion signal through matching; For the motion signal, artifact identification and filter initialization are performed based on the front buffer, and the back buffer is triggered to perform artifact processing and output a motion clean signal.

4. The motion artifact monitoring method for emergency equipment according to claim 3, characterized in that: Performing artifact identification based on the front buffer, comprising: An interference threshold is introduced, wherein the interference threshold is set by a critical value under a posture variable threshold-time-frequency joint constraint, and the posture variable includes at least one posture feature; Using the time frequency as a first independent variable, the posture variable threshold as a second independent variable, and the interference threshold as a third dependent variable, an identification linear relationship is constructed; According to the identified linear relationship, artifact identification is performed on the motion signal to generate an artifact probability instruction.

5. The motion artifact monitoring method for emergency equipment according to claim 4, characterized in that: Perform filter initialization, including: If the artifact probability instruction is empty, the motion signal is used as the motion clean signal; If the artifact probability instruction is not empty, determining a filtering standard according to a probability value, wherein the probability value is determined by an amplitude exceeding a threshold of a posture variable; An initialization process is performed on the filter according to the filtering standard.

6. The motion artifact monitoring method for emergency equipment according to claim 3, characterized in that: Before performing artifact recognition based on the front buffer, the method includes: For the motion signal, locating the interference period based on whether there is a posture variable; intercepting the motion signal during the interference period to determine an identification signal segment; For the identified signal segment, triggering artifact identification based on the front buffer.

7. The motion artifact monitoring method for emergency equipment according to claim 1, characterized in that: The method further comprises: Acquire an artifact operation chain and store it in a temporary database, wherein the temporary database is updated based on a preset period; According to the temporary database, performing artifact identification misjudgment statistics and artifact processing deviation statistics in a preset period to determine abnormal operation data; The artifact processing module is updated according to the difference operation data.

8. A motion artifact monitoring system for emergency equipment, characterized in that: A system for performing the motion artifact monitoring method for emergency equipment according to any one of claims 1 to 7, comprising: A user status data determination module is configured to introduce a space-time code, collect motion signals and physiological signals of a target user, and determine user status data, wherein the space-time code includes a synchronized timestamp and a spatial phase; The valid status data determination module is used to introduce the artifact processing module, is embedded in the data center of the emergency equipment, receives the user status data, triggers the artifact processing module, performs data-oriented identification and artifact monitoring processing, determines the valid status data, and the emergency equipment executes a response based on the valid status data; wherein, the artifact processing module has a dual-buffer architecture, the front buffer performs filter initialization under artifact identification, and the back buffer performs artifact processing.

Images