Automatic recognition system, method and device for supine and prone position change and storage medium

Abstract

This application provides an automatic system, method, device, and storage medium for recognizing changes in body position (lying or standing). The system includes: a pressure acquisition module for acquiring a first pressure distribution matrix of the head region using a first pressure pad, and a second pressure distribution matrix of the foot region using a second pressure pad; a data transmission module for acquiring the first and second pressure distribution matrices and transmitting them to the body position recognition module; a body position recognition module for calculating the pressure distribution characteristics within each distribution matrix, determining the corresponding time of the body position change based on the pressure distribution characteristics and using a body position threshold rule, predicting the risk of fall using a risk threshold rule, and obtaining the recognition result; a result output module for displaying the recognition result on a display device, marking the corresponding time of the body position change, and indicating the risk of fall; and a data storage module for storing the distribution matrices and the recognition result. This application can accurately mark the time corresponding to the body position change.

Description

Technical Field

[0001] This invention relates to the field of automatic body position recognition technology, and in particular to an automatic recognition system, method, device and storage medium for changes in body position in supine and standing positions. Background Technology

[0002] The Head-Up Tilt Test (HUTT) ​​is an important clinical examination method for assessing autonomic function, orthostatic hypotension, syncope, and cerebral hemodynamic regulation. By having the subject gradually transition from a supine to a standing position, changes in physiological parameters such as blood pressure, heart rate, and cerebral blood flow velocity can be observed, providing important diagnostic information for doctors.

[0003] In related technologies, existing prostituent position change tests often rely on manual observation and recording, which is easily affected by the observer's subjective factors or other influences, leading to deviations in recording timing and thus affecting the synchronization with physiological monitoring curves and the accuracy of results. Moreover, in actual tests, the standing process for some patients with difficulty standing often takes 5-10 seconds, and recording only one standing time point does not meet the actual clinical needs. Clinicians prefer to record the entire standing process, such as "head off the bed" representing the moment of starting to stand and "fully standing" representing the moment of completing standing. With the development of medical monitoring and sensor technologies, position recognition and human condition monitoring based on pressure sensors are gradually gaining attention. By placing pressure sensors or pressure pads on surfaces such as beds, chairs, or the ground, changes in pressure distribution between the human body and the contact surface can be collected, thereby inferring the body's position and its changes. Existing position recognition technologies based on pressure sensors are mostly used to roughly distinguish between states such as "in bed," "out of bed," "standing," and "lying down," focusing on long-term status monitoring or safety alarms. However, they are still significantly insufficient for clinical supine-standing tests, which have clear procedures and strict time requirements. They cannot accurately capture the time points of key positional changes such as "head off the bed," "fully standing," and "returning to a supine position." For example, the patent with publication number CN119296282A only identifies the approximate state of supine and standing positions using posture sensors.

[0004] In addition, existing intelligent technologies also have some shortcomings. For example, the patent with publication number CN120496833A uses skeleton reconstruction or image contour extraction to assist in body position judgment, and the patent with publication number CN119302650A uses a training model as a body position recognition model. The above solutions rely on visual recognition, which is affected by environmental factors and differences in subjects, and the accuracy and reliability of recognition are affected. Moreover, the model training process is relatively complicated. Furthermore, most of them focus on the post-judgment of body position status and lack real-time monitoring and risk warning mechanisms for the stability of body position changes, making it difficult to detect potential dangers in time and posing certain safety hazards.

[0005] Based on the above analysis of the development status of this technology field, the existing technologies lack a solution that directly senses pressure pads, uses precise threshold rules to customize the identification of key movements in lying and standing positions, automatically marks the accurate times corresponding to "head off the bed," "fully standing," and "returning to a supine position," identifies potential fall risks in advance, and forms a data linkage with display devices for intuitive display. Summary of the Invention

[0006] The purpose of this invention is to provide an automatic recognition system and method for changes in body position (lying or standing), aiming to solve the above-mentioned problems in the prior art.

[0007] According to a first aspect of the present invention, an automatic body position change recognition system is provided, comprising: The pressure acquisition module is used to acquire a first pressure distribution matrix of the head region through a first pressure pad and a second pressure distribution matrix of the foot region through a second pressure pad. The data transmission module is used to acquire the first pressure distribution matrix and the second pressure distribution matrix, and transmit them to the body position recognition module; The body position recognition module is used to calculate the pressure distribution characteristics within each distribution matrix, determine the corresponding time of body position change based on the pressure distribution characteristics and body position threshold rules, predict the risk of fall using risk threshold rules, and obtain the recognition result. The results output module is used to display the recognition results through a display device, mark the time corresponding to the body position change, and indicate the risk of falling. The data storage module is used to store the distribution matrix and recognition results.

[0008] According to a second aspect of the present invention, an automatic method for recognizing changes in body position in supine or upright positions is provided, comprising: The pressure acquisition module acquires the first pressure distribution matrix of the head area through the first pressure pad and the second pressure distribution matrix of the foot area through the second pressure pad. The first pressure distribution matrix and the second pressure distribution matrix are obtained through the data transmission module and transmitted to the body position recognition module; The pressure distribution characteristics within each distribution matrix are calculated by the body position recognition module. Based on the pressure distribution characteristics, the corresponding time of body position change is determined using the body position threshold rule. The risk of fall is predicted using the risk threshold rule to obtain the recognition result. The results output module displays the recognition results on a display device, marking the time corresponding to the body position change and indicating the risk of falling. The distribution matrix and recognition results are stored through the data storage module.

[0009] According to a third aspect of the present invention, an electronic device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the automatic identification method for changes in supine and upright positions as provided in the second aspect of the present disclosure.

[0010] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, on which an information transmission implementation program is stored, which, when executed by a processor, implements the steps of the automatic identification method for changes in supine and upright body positions provided in the second aspect of the present disclosure.

[0011] The technical solution provided by this invention has the following beneficial effects: Based on direct sensing of the pressure pad, it avoids the inaccuracy problems caused by factors such as lighting conditions, shooting angle, occlusion, or individual differences in subjects in video recognition schemes, thus improving the reliability of the system in complex clinical environments; through a unified pressure signal acquisition method and clear rules for judging body position changes, it automatically labels the accurate times corresponding to "head off the bed," "fully standing," and "returning to a supine position" by precisely identifying key movements in the supine and standing positions, reducing subjective errors caused by manual observation and improving the objectivity and consistency of body position change time point recognition; it also identifies potential fall risks in advance, thereby improving the safety level of elderly subjects or high-risk groups during the supine and standing position test; and it forms data linkage with display devices for intuitive display, improving the automation level and operational efficiency of the overall examination process.

[0012] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in one or more embodiments of this specification or in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic diagram of the automatic recognition system for changes in body position in lying and standing positions according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the pressure pad according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the baseline pressure distribution of the first pressure pad in an embodiment of the present invention; Figure 4 This is a schematic diagram of the baseline pressure distribution of the second pressure pad according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the identification system architecture according to an embodiment of the present invention; Figure 6 This is a flowchart of the automatic identification method for changes in body position in supine and standing positions according to an embodiment of the present invention; Figure 7 This is a schematic diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0015] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0016] System Implementation Examples According to an embodiment of the present invention, an automatic body position change recognition system is provided. Figure 1 This is a schematic diagram of the automatic recognition system for changes in body position (lying or standing) according to an embodiment of the present invention, as shown below. Figure 1 As shown, the automatic body position change recognition system according to an embodiment of the present invention specifically includes: The pressure acquisition module 10 is used to acquire a first pressure distribution matrix of the head region through the first pressure pad and a second pressure distribution matrix of the foot region through the second pressure pad. Figure 2 This is a schematic diagram of the pressure pad according to an embodiment of the present invention, as shown below. Figure 2 The image shows the experimental scenario.

[0017] The pressure acquisition module 10 is specifically used for: By using pressure sensors as the first and second pressure pads, the pressure acquisition module 10 can be further divided into a first pressure acquisition module and a second pressure acquisition module. Preferably, the pressure sensor can be a flexible pressure sensor array, which includes multiple pressure sampling units for acquiring the pressure distribution matrix. The supine-standing test involves lying flat on the test bed for a period of time, then quickly standing up, stabilizing for another period of time, and then returning to the supine position. Based on this test procedure, a first pressure pad is placed on the test bed below the subject's head to collect pressure data at the head position when the subject is lying flat. A second pressure pad is placed on the ground next to the test bed at the foot position when the subject is standing to collect pressure data at the soles of the feet. In this embodiment of the invention, pressure data refers to a pressure distribution matrix. Before the experiment, the pressure pads were baseline calibrated to establish an individualized pressure distribution reference model. Subjects were guided to stand on the second pressure pad, and the foot pressure distribution matrix data when standing normally was collected as the baseline data of the second pressure pad. Subjects were guided to lie flat on the test bed, and the head pressure distribution matrix data of the first pressure pad when lying flat was collected as the baseline data of the first pressure pad. The first pressure distribution matrix when lying flat and the second pressure distribution matrix when standing normally are collected as the baseline distribution matrix for the judgment of whether the threshold is exceeded in the rules. That is, the subsequent modules are called based on the baseline distribution matrix to calculate the baseline distribution characteristics, which serve as the judgment criteria for some rules. Figure 3 This is a schematic diagram of the baseline pressure distribution of the first pressure pad according to an embodiment of the present invention, as shown below. Figure 3 The diagram shows the pressure distribution formed in the head; Figure 4 This is a schematic diagram of the baseline pressure distribution of the second pressure pad according to an embodiment of the present invention, as shown below. Figure 4 As shown, this illustrates the pressure distribution created by the two feet.

[0018] The baseline calibration process described above can effectively eliminate the influence of individual differences such as body shape, weight, and stress habits of different subjects on stress data, thereby providing a stable and reliable reference for subsequent postural change recognition and improving the accuracy and stability of postural change recognition.

[0019] During the experiment, the actual collected data were used as the distribution matrices, i.e., the data to be judged.

[0020] The data transmission module 12 is connected to the pressure acquisition module 10, that is, it is connected to the first pressure acquisition module and the second pressure acquisition module respectively, and is used to acquire the first pressure distribution matrix and the second pressure distribution matrix and transmit them to the body position recognition module. The data transmission module 12 is specifically used for: The collected pressure distribution matrix data is transmitted to the target device to achieve real-time feedback and processing of pressure data. The target device can be a transcranial Doppler (TCD) device, a computer, a tablet terminal, or other terminal devices with data processing and display capabilities. The first and second pressure distribution matrices are encapsulated using a transmission protocol and marked with timestamps to obtain data information, ensuring the consistency and synchronization of data from different pressure acquisition modules and physiological data collected by other devices in time. Data information can be transmitted using wired or wireless communication, where wired communication includes, but is not limited to, USB interface communication; and wireless communication includes, but is not limited to, Bluetooth communication, Wi-Fi communication, etc.

[0021] The body position recognition module 14 is used to calculate the pressure distribution characteristics within each distribution matrix, determine the corresponding time of body position change based on the pressure distribution characteristics and body position threshold rules, predict the risk of fall using risk threshold rules, and obtain the recognition result. The body position recognition module 14 receives the transmitted pressure data, calculates the pressure distribution characteristics, automatically determines the key moments of body position changes during the supine-standing test based on the pressure change characteristics, and provides early warning of the risk of falling.

[0022] The body position recognition module 14 specifically includes: The pressure distribution feature calculation unit is used to extract features from the first pressure distribution matrix and the second pressure distribution matrix respectively, and to extract spatial distribution features and statistical features from the pressure matrix data p(i,j) collected from the two pressure pads to obtain the pressure distribution features corresponding to each distribution matrix. The pressure distribution characteristic calculation unit is specifically used for: Formula 1 is used to perform binary processing on the distribution matrix based on a preset noise threshold. The matrix cells with a value of 1 after binary processing are taken as effective pressure regions, and the first pressure pad effective pressure region image B1 and the second pressure pad effective pressure region image B2 are generated respectively. There may be multiple effective pressure regions in each image. Formula 1; in, This represents the result of binary processing. Indicates the pressure value of the matrix unit. Represents the x-axis, Represents the ordinate, Indicates the preset noise threshold; The effective pressure area reflects the change in the contact area between the body and the pressure pad. Formula 2 is used to calculate the effective pressure area. : Formula 2; The center of gravity reflects the overall positional shift of pressure distribution and is used to determine postural changes such as standing up and landing. The center of gravity is calculated within each effective pressure region; it can only be calculated if an effective pressure region exists. If multiple effective pressure regions exist, the center of gravity can be calculated within any of them. For example, in a standing position, two centers of gravity are typically calculated. and Use Formula 3 to calculate the x-coordinate of the pressure center of gravity. and ordinate : Formula 3; The pressure distribution entropy quantifies the uniformity and dispersion of the pressure distribution. In a steady state, the pressure distribution is relatively uniform, and the pressure distribution entropy is low. Formula 4 is used to normalize the distribution matrix into a probability distribution, and Formula 5 is used to calculate the pressure distribution entropy based on this probability distribution. : Formula 4; Formula 5; in, This represents the result of probability distribution processing; Calculate the average pressure value within the effective pressure region to obtain the average pressure value; The effective pressure region, effective pressure area, pressure centroid, pressure distribution entropy, and average pressure value are used as pressure distribution characteristics.

[0023] The body position event determination unit, based on multiple pressure distribution features output by the pressure distribution feature calculation unit, comprehensively reflects the changes in the contact range, spatial position, force concentration and distribution complexity between the human body and the two pressure pads by analyzing the changes in features within a certain time window, thereby achieving automatic identification of different body position states and body position change processes. The body position event determination unit is used to obtain the body position threshold rule composed of the first rule, the second rule and the third rule; determine the time of head leaving the bed according to the pressure distribution characteristics of the first pressure pad and the first rule; after the head leaves the bed, determine the time of complete standing body position change according to the pressure distribution characteristics of the second pressure pad and the second rule; after standing, determine the time of returning to supine body position according to the pressure distribution characteristics of the first pressure pad and the third rule. The specific rules for body position thresholds include: (1) The first rule for determining head off the bed When the subject got up from a supine position, the contact between the head and the first pressure pad gradually changed from full contact to partial contact within a short period of time, and finally reached complete separation; during this process, the pressure distribution characteristics of the first pressure pad showed a phased and continuous change in the time dimension.

[0024] During the body position recognition process, the body position recognition module first confirms that the subject has not yet entered a standing state based on the pressure distribution of the second pressure pad, that is, the effective pressure area of ​​the second pressure pad is zero or below the noise threshold, thus limiting the current body position change to the stage of getting up from the bed, as described below according to the first rule, combined with the process characteristic changes during this stage: The effective pressure area corresponding to the first pressure pad decreases at a rate exceeding the decrease threshold compared to the baseline area of ​​the first pressure pad within a preset time window; that is, the number of effective pressure areas of the first pressure pad continues to decrease, and the pressure distribution area decreases rapidly within the preset time window. The decrease threshold is similar to decreasing to below 50% of the baseline pressure area within 1 second, which represents a rapid reduction in the contact area between the head and the bed surface. The center of gravity of the first pressure pad gradually deviates from the center of gravity of the supine baseline. That is, the center of gravity of the pressure pad deviates from the allowable fluctuation area set by the center of gravity of the first pressure pad baseline. The movement rate exceeds or reaches the movement threshold. This is used to distinguish between the overall lifting of the head off the bed surface and the slight change in the center of gravity caused by local posture adjustment. During the transition from full contact to partial contact at the head, the pressure distribution of the first pressure pad changes from a relatively uniform state to a discrete state, and the pressure distribution entropy value increases to the discrete state threshold. The pressure distribution entropy value is significantly higher than the baseline state, which is used to characterize the change of the contact pattern from stable fit to non-uniform contact. When the average pressure value decreases from near the baseline pressure level to the noise threshold within a preset time window, and eventually falls below the noise threshold, it indicates that the head has basically detached from the bed support.

[0025] When at least three characteristic changes are met and the duration reaches the interval threshold, each pressure distribution characteristic is considered as a feature. That is, if the rule is met for a certain period of time, the time corresponding to the head leaving the bed is determined to be the time before the interval threshold. For example, the position recognition module is based on the consistent changes of the above multiple pressure distribution characteristics in the time dimension. When the relevant features jointly indicate that the head and bed surface contact state has transitioned from partial contact to complete separation, and the duration of this state is not less than a preset time threshold, such as 0.5 seconds, the corresponding time point is determined to be the time when the head leaves the bed, that is, the time before 0.5 seconds.

[0026] (2) The second rule for determining a fully standing position The "fully standing" moment corresponds to the process by which the subject gradually transitions from a state of getting up to a stable upright state after completing the "head off the bed" transition. During this process, the subject's body support method changes from bed surface support to bipedal support, and the pressure distribution characteristics of the second pressure pad show an evolution from no support, single-leg or partial support to stable bipedal support over time.

[0027] During the body position recognition process, after recognizing the moment when the head leaves the bed, the body position recognition module first performs a constraint judgment on the pressure distribution of the first pressure pad. When the effective pressure area of ​​the first pressure pad is zero or lower than the preset noise threshold, it confirms that the subject's head has completely left the bed support, thus entering the standing state determination stage, which corresponds to the second rule as follows, combined with the description of the changes in process characteristics during this stage: Two effective pressure areas appear in the second pressure pad. The effective pressure area reaches the preset range threshold within the preset percentage range of the baseline area of ​​the second pressure pad. The effective pressure area of ​​the second pressure pad gradually appears from no effective area and evolves from a local concentrated area into two relatively stable pressure concentrated areas. The overall pressure distribution area reaches and is maintained within the expected range set based on the baseline standing state, which is used to characterize that the subject's feet have formed effective support with the ground. In the initial stage of standing up, the center of gravity of the second pressure pad moves rapidly or shifts significantly. As the subject gradually stands up, the magnitude of the shift in the center of gravity gradually decreases. The center of gravity stabilizes from the shifted state to the allowable fluctuation area set by the baseline center of gravity of the second pressure pad. The movement rate is lower than the movement threshold, which is used to characterize the body's overall tendency to balance.

[0028] A symmetrical analysis of the pressure distribution in the left and right areas of the second pressure pad was performed. The difference between the effective pressure area and the average pressure value between the two effective pressure areas was lower than the difference threshold, indicating that the bipedal support state had been basically established, thus distinguishing the transitional states such as single-foot treading and posture adjustment. During the standing transition phase, the pressure distribution of the second pressure pad exhibits a change from discrete and locally concentrated to a relatively stable distribution. The pressure distribution entropy value gradually decreases from the dry level to the stable state threshold, which is used to characterize the transformation of plantar pressure from unstable support to stable support.

[0029] When at least three characteristic changes are met and the duration reaches the interval threshold, the time corresponding to the fully standing position is determined to be the time before the interval threshold. The position recognition module determines the time point as the fully standing moment based on the consistent changes of the above multiple pressure distribution characteristics in the time dimension. When the pressure distribution of the second pressure pad simultaneously meets the requirements of establishing bipedal support, stabilizing the pressure center of gravity, and achieving the symmetry of pressure distribution, and the duration of the above states is not less than the preset time threshold, such as 0.5 seconds.

[0030] (3) The third rule for determining a fully standing position The "return to supine position" moment corresponds to the transition from standing to lying down again after the subject has completed the "fully standing" state. During this process, the subject's body support gradually shifts from foot support to bed surface support. The foot pressure distribution of the second pressure pad gradually disappears, while the head pressure distribution of the first pressure pad gradually recovers and eventually approaches the pressure distribution characteristics of the initial supine state.

[0031] During the body position recognition process, after the body position recognition module has identified the moment of "fully standing", it performs joint analysis on the pressure distribution characteristics of the two pressure pads to determine whether the subject has completed the return to a supine position.

[0032] The body position recognition module constrains and judges the pressure distribution of the second pressure pad. When the effective pressure area of ​​the second pressure pad gradually decreases and eventually disappears, or when its overall pressure value is lower than the preset noise threshold and continues for a preset time, it is determined that the subject's feet have left the second pressure pad, thus entering the bed surface contact state determination stage. Subsequently, the body position recognition module comprehensively analyzes the head pressure distribution characteristics based on the array pressure data collected by the first pressure pad, which corresponds to the third rule as follows, combined with the description of the changes in process characteristics during this stage: The effective pressure area of ​​the first pressure pad gradually appears from the absence of an effective area and expands rapidly. When an effective pressure area appears in the first pressure pad, the effective pressure area rises to the baseline area of ​​the first pressure pad within a preset time window, which is used to characterize the head re-establishing stable contact with the bed surface. During the transition to lying flat, the center of gravity of the first pressure pad showed obvious movement or instability. As the subject completed the lying position, the center of gravity returned to the allowable fluctuation area set by the baseline center of gravity of the first pressure pad, and the movement rate was lower than the movement threshold, which is used to characterize that the head position tends to be fixed. The pressure distribution of the first pressure pad gradually evolves from a discrete distribution in the transition phase to a stable and continuous distribution pattern. The pressure distribution entropy value decreases to the stable state threshold, which is used to characterize the recovery of the head's fit with the bed surface. The average pressure value of the first pressure pad gradually increases from close to the noise level. When the average pressure value rises to the rising noise threshold within a preset time window, the noise threshold can be a pressure range to eliminate misjudgments caused by instantaneous contact with the bed or brief contact.

[0033] When at least three characteristic changes are met and the duration reaches the interval threshold, the time corresponding to the return to the supine position is determined to be the time before the interval threshold. The position recognition module determines the time point as the time of return to the supine position based on the consistent changes of the above multiple pressure distribution characteristics in the time dimension. When the pressure distribution of the second pressure pad disappears, and the pressure distribution of the first pressure pad simultaneously meets the constraints of the baseline supine state in multiple characteristics such as area, center of gravity, entropy value and average pressure, and the duration is not less than a preset time threshold, such as 0.5 seconds, the corresponding time point is determined to be the time of return to the supine position.

[0034] The fall risk warning unit is used to obtain the risk threshold rule composed of the fourth rule and the fifth rule; the fourth rule is used to predict fall risk during the transition from head-off-bed position to fully standing position, and the fifth rule is used to predict fall risk during the transition from fully standing position to supine position.

[0035] During the supine-standing test, especially during the transition from a supine to a standing position and back to a supine position, elderly subjects or those with autonomic nervous system dysfunction are prone to postural instability and abnormal center of gravity shift, posing a risk of falls. To improve the safety of the supine-standing test, this embodiment of the invention further incorporates a fall risk warning unit in the position recognition module. This unit is used to assess the stability of the subject's positional changes in real time based on pressure distribution characteristics and to output warning information when an abnormal state is detected.

[0036] The fall risk warning unit analyzes the dynamic characteristics of pressure distribution during the subjects’ standing, standing and returning to a supine position by combining the foot array pressure data collected by the second pressure pad with the head pressure data of the first pressure pad. It is important to note that the judgment process does not aim at whether a fall has already occurred, but rather assesses the stability of pressure distribution, the pattern of center of gravity changes, and the continuity of pressure space shape to provide early warning of potential fall risks.

[0037] The risk threshold rules specifically include: (1) The fourth rule for risk prediction in the first transition phase Specifically, during the transition from "head off the bed" to "fully standing," the fall risk warning unit focuses on monitoring changes in the pressure distribution characteristics of the second pressure pad, as described below in accordance with the fourth rule, combined with the changes in process characteristics during this stage: The position of the center of gravity of the foot pressure is detected within a certain time window. When the displacement of the center of gravity of the pressure detected by the second pressure pad exceeds the large displacement threshold, the movement rate is higher than the movement threshold, or the smoothness of the movement point trajectory is higher than the jump threshold, i.e., it shows discontinuous jump characteristics, it is determined that the subject's center of gravity control ability is insufficient during the standing process and there is a risk of postural instability. In a normal standing state, the pressure distribution of the two feet usually presents a relatively symmetrical spatial structure. If, during the standing process, the difference between the effective pressure area and the average pressure value between the two effective pressure areas is higher than the difference threshold, that is, the subject is in a state of obvious asymmetry for a long time or the symmetry index fluctuates drastically in a short period of time, it is determined that the subject has a risk of unbalanced loading or postural imbalance. When the two parameters are judged by the difference threshold, they are compared independently and by their respective thresholds. If there is a difference in both, it is determined to be asymmetric. The effective area, average pressure, and pressure distribution entropy of the plantar pressure distribution are continuously monitored. When the time series fluctuations of the effective pressure area, pressure distribution entropy, and average pressure are higher than the fluctuation threshold, that is, when the above features show high frequency and large amplitude fluctuations within a preset time window and cannot converge to a stable range within a reasonable time, a fall risk signal is output, specifically indicating that the subject's standing stability is insufficient.

[0038] (2) The Fifth Rule for Risk Forecasting in the Second Transition Phase Rule 5 was applied during the transition of subjects from a standing position to a "return to supine" position; The fall risk warning unit further combines the pressure distribution characteristics of the first pressure pad for joint judgment. When the pressure value in the second pressure pad has not completely disappeared, the first pressure pad shows non-connected pressure distribution characteristics, that is, scattered, brief or abnormally positioned pressure distribution characteristics have appeared. The duration of the appearance is less than the brief threshold duration, and after the scattered characteristics reach the stable threshold duration, they do not return to the state corresponding to the supine position of the first pressure pad. That is, the characteristics have failed to quickly evolve into a stable supine pressure distribution pattern, and a fall risk signal is output. Specifically, it is determined that the subject has a risk of loss of balance or insufficient control of movement during the lying process.

[0039] When the fall risk warning unit detects any of the above abnormal features that meet the preset risk judgment conditions and the duration of the abnormal state exceeds the preset threshold, the system outputs fall risk warning information. The warning information can be displayed on the result output module 16 through the display device, or reminded medical staff or subjects to pay attention to postural stability through voice, sound and light, etc., so as to intervene before the actual fall occurs and improve the safety of the supine-standing test process.

[0040] It should be noted that the expression of globally instantiable parameters has been standardized across the five rules. For example, the preset time threshold is set to 0.5 when judging different rules. For parameters with the same meaning but instantiated differently at different stages, the expression "first", "second", etc. are used to distinguish them.

[0041] The result output module 16 is used to display the recognition results through a display device, mark the time corresponding to the change in body position, and indicate the risk of fall. Specifically, it is used for: Connected to body position recognition module 14; The identified time points of body position changes are displayed in real time through a display device, or the identification results are sent to a transcranial Doppler (TCD) device for data linkage display. Once the time points of body position changes are identified and determined, the system can display the real-time trend of the subject's body position changes through a display device, including the time points of key moments such as "head off the bed", "fully standing" and "returning to supine", and mark them on the trend chart. The results output module 16 can not only output the results after the test, but also continuously output the pressure changes of the two pressure acquisition modules in the form of pressure heatmaps throughout the entire supine-standing test. When the body position recognition module 14 detects that the subject's movements are unstable or the pressure fluctuates abnormally, it can remind the doctor or subject to pay attention to maintaining stability through voice or graphics, and mark it on the trend graph, thereby improving the accuracy of the test data.

[0042] Through real-time data interaction with the TCD device, the result output module 16 achieves precise synchronization of body position change time points with physiological signals, improving the intelligence and operational efficiency of the examination process.

[0043] Data storage module 18 is used to store the distribution matrix and recognition results, specifically for: Connected to data transmission module 12 and body position recognition module 14; The system stores the subject's pressure spatial distribution data, identified time points of body position changes, and corresponding physiological data trend graphs. Specifically, the data storage module stores the real-time pressure spatial distribution data collected by the first pressure acquisition module and the second pressure acquisition module, saves the key time points identified by the body position recognition module 14, and saves the physiological data trend graphs and their annotation information collected by the TCD device. The stored data can be viewed and analyzed in subsequent clinical settings, enabling doctors to make comprehensive diagnoses based on complete records of body position changes and physiological signal trends. The data storage module 18 can use local storage or cloud storage, supporting long-term data preservation and convenient retrieval, ensuring that data is not lost during the trial, and meeting data access requirements under different clinical needs, ensuring the integrity and traceability of the trial data.

[0044] The above technical solutions of the embodiments of the present invention will be illustrated with reference to the following accompanying drawings.

[0045] Figure 5 This is a schematic diagram of the identification system architecture according to an embodiment of the present invention, as shown below. Figure 5 As shown, the detailed connection architecture of the automatic recognition system for changes in supine and upright body position based on pressure pads is illustrated. The pressure acquisition module 10 is further divided into a first pressure acquisition module and a second pressure acquisition module. In addition, it also includes a data transmission module 12, a body position recognition module 14, a result output module 16, and a data storage module 18.

[0046] The pressure acquisition module collects pressure distribution matrix data for the subject's head position when lying flat and the soles of their feet when standing, and transmits this data to the position recognition module via the data transmission module. The position recognition module first calculates pressure characteristics using the pressure distribution feature calculation unit, and the position event determination unit identifies key moments in the supine-standing test based on pressure change characteristics, including "head off the bed," "fully standing," and "returning to a supine position." Simultaneously, during the test, the fall risk warning unit provides fall warnings based on the pressure characteristics of the second pressure pad. The results output module can display the identified key time points via a display device or send the recognition results to a TCD device for real-time display and annotation in the physiological data trend graph. The data storage module saves the subject's pressure data, the identified position change time points, and the physiological data trend graph for subsequent viewing and analysis.

[0047] In summary, to address the existing problems, this invention presents an automatic postural change recognition system. Based on a pressure pad, it directly senses the pressure distribution matrix, avoiding the inaccuracies caused by factors such as lighting conditions, shooting angle, occlusion, or individual differences in subjects, which are common in video recognition solutions. This improves the system's reliability in complex clinical environments. Through a unified pressure signal acquisition method and clearly defined postural change judgment rules, it precisely identifies key postural movements, automatically labeling the accurate times of "head off the bed," "fully standing," and "returning to a supine position," reducing subjective errors from manual observation and improving the accuracy of postural change recognition. The system ensures the objectivity and consistency of intermittent point identification; analyzes the pressure distribution and its changing characteristics; monitors postural stability in real time during the subjects' rising, standing, and returning to a supine position; identifies potential fall risks in advance, thereby improving the safety of elderly subjects or high-risk groups during the supine-standing test; and establishes data linkage with the display device for intuitive display. Through the communication interface, the data collected by the pressure pad or the key moments identified are transmitted to the TCD device in real time, and the corresponding time points are automatically marked on the TCD trend chart. It also supports the synchronous control of the start and end commands of the supine-standing test, improving the automation level and operational efficiency of the overall examination process.

[0048] Method Implementation Examples According to an embodiment of the present invention, an automatic method for recognizing changes in body position from supine to upright is provided. Figure 6 This is a flowchart of the automatic identification method for changes in body position in supine and upright positions according to an embodiment of the present invention, as follows: Figure 6 As shown, the automatic identification method for changes in body position in supine or standing positions according to an embodiment of the present invention specifically includes: In step S610, the pressure acquisition module acquires the first pressure distribution matrix of the head region through the first pressure pad and the second pressure distribution matrix of the foot region through the second pressure pad. In step S620, the first pressure distribution matrix and the second pressure distribution matrix are obtained through the data transmission module and transmitted to the body position recognition module; In step S630, the pressure distribution characteristics within each distribution matrix are calculated by the body position recognition module. Based on the pressure distribution characteristics, the corresponding time of body position change is determined using the body position threshold rule. The risk of fall is predicted using the risk threshold rule to obtain the recognition result. In step S640, the recognition results are displayed on the display device through the result output module, the corresponding time of the body position change is marked, and the risk of falling is indicated; In step S650, the distribution matrix and recognition results are stored through the data storage module.

[0049] In summary, to address the existing problems, this invention presents an automatic method for recognizing changes in body position (lying or standing). Based on a pressure pad, it directly senses the pressure distribution matrix, avoiding inaccuracies caused by factors such as lighting conditions, shooting angle, occlusion, or individual differences in subjects, which are common in video recognition solutions. This improves the system's reliability in complex clinical environments. Through a unified pressure signal acquisition method and clearly defined rules for recognizing changes in body position, and by precisely identifying key movements in lying or standing positions, the method automatically labels the accurate times for "head off the bed," "fully standing," and "returning to a supine position," reducing subjective errors from manual observation and improving the accuracy of recognizing changes in body position. The system ensures the objectivity and consistency of intermittent point identification; analyzes the pressure distribution and its changing characteristics; monitors postural stability in real time during the subjects' rising, standing, and returning to a supine position; identifies potential fall risks in advance, thereby improving the safety of elderly subjects or high-risk groups during the supine-standing test; and establishes data linkage with the display device for intuitive display. Through the communication interface, the data collected by the pressure pad or the key moments identified are transmitted to the TCD device in real time, and the corresponding time points are automatically marked on the TCD trend chart. It also supports the synchronous control of the start and end commands of the supine-standing test, improving the automation level and operational efficiency of the overall examination process.

[0050] Electronic device examples Figure 7 This is a schematic diagram of an electronic device according to an embodiment of the present invention. The electronic device 700 may include at least one processor 710 and a memory 720. The processor 710 can execute instructions stored in the memory 720. The processor 710 is communicatively connected to the memory 720 via a data bus. In addition to the memory 720, the processor 710 can also be communicatively connected to an input device 730, an output device 740, and a communication device 750 via the data bus.

[0051] The processor 710 can be any conventional processor, such as a commercially available CPU. The processor may also include graphics processing units (GPUs), field-programmable gate arrays (FPGAs), systems on chips (SoCs), application-specific integrated circuits (ASICs), or combinations thereof.

[0052] The memory 720 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0053] In this embodiment of the present disclosure, the memory 720 stores executable instructions, and the processor 710 can read the executable instructions from the memory 720 and execute the instructions to implement all or part of the steps of the automatic identification method for changes in supine and standing positions in any of the exemplary embodiments described above.

[0054] Computer-readable storage medium embodiments In addition to the methods and systems described above, exemplary embodiments of this disclosure may also be a computer program product or a computer-readable storage medium storing the computer program product, the computer product including computer program instructions that can be executed by a processor to implement all or part of the steps described in any of the above exemplary embodiments of the automatic identification method for changes in supine or upright body position.

[0055] Computer program products can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. Programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages, and scripting languages ​​(e.g., Python). The program code can be executed entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0056] Computer-readable storage media may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example,, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media include: static random access memory (SRAM) having one or more electrically connected wires; electrically erasable programmable read-only memory (EEPROM); erasable programmable read-only memory (EPROM); programmable read-only memory (PROM); read-only memory (ROM); magnetic storage; flash memory; magnetic disk or optical disk; or any suitable combination thereof.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An automatic body position change recognition system, characterized in that, include: The pressure acquisition module is used to acquire a first pressure distribution matrix of the head region through a first pressure pad and a second pressure distribution matrix of the foot region through a second pressure pad. The data transmission module is used to acquire the first pressure distribution matrix and the second pressure distribution matrix, and transmit them to the body position recognition module; The body position recognition module is used to calculate the pressure distribution characteristics within each distribution matrix, determine the corresponding time of body position change based on the pressure distribution characteristics and body position threshold rules, predict the risk of fall using risk threshold rules, and obtain the recognition result. The result output module is used to display the recognition results through a display device, mark the time corresponding to the body position change, and indicate the risk of fall. The data storage module is used to store the distribution matrix and the identification results.

2. The system according to claim 1, characterized in that, The pressure acquisition module is specifically used for: Pressure sensors are used as the first pressure pad and the second pressure pad. The first pressure pad is placed on the test bed below the subject's head, and the second pressure pad is placed on the ground next to the test bed at the subject's feet when standing. Before the experiment, the first pressure distribution matrix when lying flat and the second pressure distribution matrix when standing normally were collected as baseline distribution matrices for judging whether the threshold was exceeded in the rules. During the experiment, the data actually collected were used as each distribution matrix.

3. The system according to claim 1, characterized in that, The data transmission module is specifically used for: The first pressure distribution matrix and the second pressure distribution matrix are encapsulated using a transmission protocol and marked with timestamps to obtain data information; The data information is transmitted using wired or wireless communication.

4. The system according to claim 1, characterized in that, The body position recognition module specifically includes: The pressure distribution feature calculation unit is used to extract features from the first pressure distribution matrix and the second pressure distribution matrix respectively to obtain the pressure distribution features corresponding to each distribution matrix. The body position event determination unit is used to obtain the body position threshold rule composed of the first rule, the second rule and the third rule; determine the time of head leaving the bed according to the pressure distribution characteristics corresponding to the first pressure pad and the first rule; after the head leaves the bed, determine the time of complete standing body position change according to the pressure distribution characteristics corresponding to the second pressure pad and the second rule; after standing completely, determine the time of returning to supine body position according to the pressure distribution characteristics corresponding to the first pressure pad and the third rule. The fall risk warning unit is used to obtain a risk threshold rule consisting of a fourth rule and a fifth rule; the fourth rule is used to predict fall risk during the transition from a head-off-bed position to a fully standing position, and the fifth rule is used to predict fall risk during the transition from a fully standing position to a supine position.

5. The system according to claim 4, characterized in that, The pressure distribution characteristic calculation unit is specifically used for: Formula 1 is used to perform binary processing on the distribution matrix based on a preset noise threshold, and the matrix cells with a value of 1 after binary processing are taken as the effective pressure area. Official 1; in, This represents the result of binary processing. Indicates the pressure value of the matrix unit. Represents the x-axis, Represents the ordinate, Indicates the preset noise threshold; Calculate the effective pressure area using Formula 2. : Official 2; Calculate the pressure centroid within each effective pressure region, and use Formula 3 to calculate the abscissa of the pressure centroid. and ordinate : Official 3; The distribution matrix is ​​normalized to a probability distribution using Equation 4, and the pressure distribution entropy is calculated based on the probability distribution using Equation 5. : Official 4; Official 5; in, This represents the result of probability distribution processing; Calculate the average pressure value within the effective pressure region to obtain the average pressure value; The effective pressure region, the effective pressure area, the pressure centroid, the pressure distribution entropy value, and the average pressure value are used as the pressure distribution characteristics.

6. The system according to claim 5, characterized in that, The body position threshold rules specifically include: The first rule is that when the effective pressure area corresponding to the first pressure pad decreases at a rate exceeding a threshold relative to the baseline area of ​​the first pressure pad within a preset time window, the pressure center of gravity deviates from the allowable fluctuation area set by the baseline center of gravity of the first pressure pad, the movement rate exceeds the movement threshold, the pressure distribution entropy value increases to the discrete state threshold, and the average pressure value decreases to the decrease noise threshold within a preset time window, at least three characteristic changes are satisfied and the duration reaches the interval threshold, and the time corresponding to the head-off-bed position is determined to be the time before the interval threshold. The second rule is as follows: when two effective pressure areas appear in the second pressure pad, the effective pressure area reaches a preset range threshold within a preset percentage range of the baseline area of ​​the second pressure pad, the pressure center of gravity stabilizes from the offset state to the allowable fluctuation area set by the baseline center of gravity of the second pressure pad, the movement rate is lower than the movement threshold, the difference between the effective pressure area and the average pressure value between the two effective pressure areas is lower than the difference threshold, and the pressure distribution entropy value decreases to the stable state threshold, satisfying at least three characteristic changes and the duration reaching the interval threshold, the time corresponding to the fully standing position is determined to be the time before the interval threshold. The third rule is that when an effective pressure area appears in the first pressure pad, the effective pressure area rises to the baseline area of ​​the first pressure pad within a preset time window, the pressure center of gravity returns to the allowable fluctuation area set by the baseline center of gravity of the first pressure pad, the movement rate is lower than the movement threshold, the pressure distribution entropy value decreases to the stable state threshold, and the average pressure value rises to the rising noise threshold within a preset time window, at least three characteristic changes are satisfied and the duration reaches the interval threshold, and the time corresponding to the restoration of the supine position is determined to be the time before the interval threshold.

7. The system according to claim 5, characterized in that, The risk threshold rules specifically include: The fourth rule is that when the second pressure pad detects that the displacement of the pressure center of gravity exceeds the large displacement threshold, the movement rate is higher than the movement threshold, or the smoothness of the movement point trajectory is higher than the jump threshold, the difference between the effective pressure area and the average pressure value between the two effective pressure areas is higher than the difference threshold, and the time series fluctuation formed by the effective pressure area, the pressure distribution entropy value and the average pressure value is higher than the fluctuation threshold, a fall risk signal is output. The fifth rule states that when the pressure value in the second pressure pad has not completely disappeared, a non-connected pressure distribution characteristic appears in the first pressure pad. The duration of this non-connected characteristic is less than the short-term threshold duration, and after reaching the stable threshold duration, the body does not return to the state corresponding to the supine position of the first pressure pad. In this case, a fall risk signal is output.

8. An automatic method for recognizing changes in supine and upright body positions, used in the automatic recognition system for changes in supine and upright body positions as described in any one of claims 1 to 7, characterized in that, include: The pressure acquisition module acquires the first pressure distribution matrix of the head area through the first pressure pad and the second pressure distribution matrix of the foot area through the second pressure pad. The first pressure distribution matrix and the second pressure distribution matrix are obtained through the data transmission module and transmitted to the body position recognition module; The pressure distribution characteristics within each distribution matrix are calculated by the body position recognition module. Based on the pressure distribution characteristics, the corresponding time of body position change is determined using the body position threshold rule. The risk of fall is predicted using the risk threshold rule to obtain the recognition result. The recognition results are displayed on a display device through the result output module, marking the time corresponding to the body position change and indicating the risk of fall. The distribution matrix and the identification results are stored through the data storage module.

9. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the automatic identification method for changes in supine and upright body positions as described in claim 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an information transmission implementation program, which, when executed by a processor, implements the steps of the automatic identification method for changes in supine and standing positions as described in claim 8.

Images